https://kallimachos.de/index.php?action=history&feed=atom&title=Tutorial_anonymus_translators_%28en%29Tutorial anonymus translators (en) - Versionsgeschichte2026-04-07T10:54:29ZVersionsgeschichte dieser Seite in KallimachosMediaWiki 1.43.6https://kallimachos.de/index.php?title=Tutorial_anonymus_translators_(en)&diff=3913&oldid=prevDeletedUser: /* Composition of a text corpus */2017-09-19T09:51:14Z<p><span class="autocomment">Composition of a text corpus</span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 19. September 2017, 09:51 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l5">Zeile 5:</td>
<td colspan="2" class="diff-lineno">Zeile 5:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>===Composition of a text corpus===</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>===Composition of a text corpus===</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>The aim of research for '''[[Identifikation_von_Übersetzern:Main |the project <del style="font-weight: bold; text-decoration: none;">in</del>]]''' is the identification of anonymous arabic-latin translations in medieval times by means of philological and computer-aided methods of style analysis.</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>The aim of research for '''[[Identifikation_von_Übersetzern:Main |the project]]''' is the identification of anonymous arabic-latin translations in medieval times by means of philological and computer-aided methods of style analysis.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>For this purpose, a corpus of electronic latin texts must be constructed. It's advisable to restrict the corpus to a certain arabic author, e.g. Averroes, or to a technical discipline, e.g. philosophy, astronomy/astrology, medicine, mathematics, alchemy/macig/prophecy or religion. </div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>For this purpose, a corpus of electronic latin texts must be constructed. It's advisable to restrict the corpus to a certain arabic author, e.g. Averroes, or to a technical discipline, e.g. philosophy, astronomy/astrology, medicine, mathematics, alchemy/macig/prophecy or religion. </div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>However, this is only possible if the corpus is large enough. At Wuerzburg University an Averroes-based corpus (Hasse 2010) and two corpora with philosophical and astronomical/astrological translations of 12 century were formed and employed (Hasse 2016 and Hasse-Büttner in print). Herein, we were able to benefit from a list of philosophical arabic-latin translations already provided by Burnett in 2005, as well as Carmody in 1956 with a list of astronomic-astrologic translations (which are imprecisely and obsolete, though). In other branches of science, such lists have yet to be created.</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>However, this is only possible if the corpus is large enough. At Wuerzburg University an Averroes-based corpus (Hasse 2010) and two corpora with philosophical and astronomical/astrological translations of 12 century were formed and employed (Hasse 2016 and Hasse-Büttner in print). Herein, we were able to benefit from a list of philosophical arabic-latin translations already provided by Burnett in 2005, as well as Carmody in 1956 with a list of astronomic-astrologic translations (which are imprecisely and obsolete, though). In other branches of science, such lists have yet to be created.</div></td></tr>
</table>DeletedUserhttps://kallimachos.de/index.php?title=Tutorial_anonymus_translators_(en)&diff=3912&oldid=prevDeletedUser: /* (II) Computergestützte Stilometrie mit Burrows Delta */2017-09-06T10:33:28Z<p><span class="autocomment">(II) Computergestützte Stilometrie mit Burrows Delta</span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 6. September 2017, 10:33 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l52">Zeile 52:</td>
<td colspan="2" class="diff-lineno">Zeile 52:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>For very short texts, it my be advisibale to expand the analysis to less frequent words. However, in this case, the less frequent words of other translators have to be compared as well. Experience shows, that only a huge amassment of these less typical words and phrases in an anonymously translated text allow for a credible attribution.</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>For very short texts, it my be advisibale to expand the analysis to less frequent words. However, in this case, the less frequent words of other translators have to be compared as well. Experience shows, that only a huge amassment of these less typical words and phrases in an anonymously translated text allow for a credible attribution.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>===(II) <del style="font-weight: bold; text-decoration: none;">Computergestützte Stilometrie mit </del>''Burrows Delta''===</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>===(II) <ins style="font-weight: bold; text-decoration: none;">Computerized Stylometry using </ins>''Burrows Delta''===</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div><del style="font-weight: bold; text-decoration: none;">Die zweite Methode basiert auf der Idee von </del>John <del style="font-weight: bold; text-decoration: none;">Burrows</del>, <del style="font-weight: bold; text-decoration: none;">dass Autorschaft computergestützt durch den Vergleich der standardisierten relativen Häufigkeiten der </del>most frequent words (MFW) <del style="font-weight: bold; text-decoration: none;">einzelner Texte ermittelt werden kann – ein Verfahren, das sich bei der computergestützten Autorschaftsattribuierung als ausgesprochen erfolgreich herausgestellt hat</del>. <del style="font-weight: bold; text-decoration: none;">Es gibt verschiedene frei im Web zugängliche Implementierungen dieses Verfahrens</del>. <del style="font-weight: bold; text-decoration: none;">Ein nutzerfreundliches Interface wird innerhalb des </del>Stylo-R<del style="font-weight: bold; text-decoration: none;">-Pakets von </del>Maciej Eder und Jan Rybicki <del style="font-weight: bold; text-decoration: none;">angeboten</del>. <del style="font-weight: bold; text-decoration: none;">Wir haben eine eigene Implementierung </del>in Python <del style="font-weight: bold; text-decoration: none;">verwendet</del>, <del style="font-weight: bold; text-decoration: none;">die auf </del>Fotis Jannidis‘ [https://github.com/fotis007/pydelta pydelta] <del style="font-weight: bold; text-decoration: none;">aufbaut</del>. <del style="font-weight: bold; text-decoration: none;">In der Regel kann man bei solchen Implementierungen zwischen verschiedenen Abstandsmaßen („Deltas“) wählen</del>, <del style="font-weight: bold; text-decoration: none;">also zwischen verschiedenen Methoden</del>, <del style="font-weight: bold; text-decoration: none;">in denen der Computer den Abstand zwischen den Texten berechnet (bzw</del>. <del style="font-weight: bold; text-decoration: none;">genauer: den Abstand zwischen den Listen der Worthäufigkeiten der häufigsten Wörter berechnet)</del>. <del style="font-weight: bold; text-decoration: none;">Vergleichsstudien der jüngsten Vergangenheit haben gezeigt</del>, <del style="font-weight: bold; text-decoration: none;">dass ein sehr performantes stilometrische Abstandsmaß das sogenannte „Cosine Delta“ ist</del>. <del style="font-weight: bold; text-decoration: none;">Auch wir haben die besten Ergebnisse mit </del>Cosine Delta <del style="font-weight: bold; text-decoration: none;">erzielt</del>. </div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">The second method is based on the ideas of </ins>John <ins style="font-weight: bold; text-decoration: none;">Burrow´s</ins>, <ins style="font-weight: bold; text-decoration: none;">which assume that authorship can be identified by comparing the standarized relative frequencies of the </ins>most frequent words (MFW) <ins style="font-weight: bold; text-decoration: none;">in texts. The method has proven itself to be highly successfull for computerized authorship attribution</ins>. <ins style="font-weight: bold; text-decoration: none;">Many different open-source implementation of this method can be found in the web</ins>. <ins style="font-weight: bold; text-decoration: none;">A user-friendly interface is employed as part of the </ins>Stylo-<ins style="font-weight: bold; text-decoration: none;">package for </ins>R <ins style="font-weight: bold; text-decoration: none;">by </ins>Maciej Eder und Jan Rybicki. <ins style="font-weight: bold; text-decoration: none;">We used an own implementation </ins>in Python, <ins style="font-weight: bold; text-decoration: none;">based on </ins>Fotis Jannidis‘ [https://github.com/fotis007/pydelta pydelta]. <ins style="font-weight: bold; text-decoration: none;">Usually</ins>, <ins style="font-weight: bold; text-decoration: none;">these implementation offer the choice over different distance mesaurements or "Deltas"</ins>, <ins style="font-weight: bold; text-decoration: none;">i.e</ins>. <ins style="font-weight: bold; text-decoration: none;">different methods for the computerized calculation of the stylistic "distance" of two texts</ins>. <ins style="font-weight: bold; text-decoration: none;">Recent studies have shown</ins>, <ins style="font-weight: bold; text-decoration: none;">that the so-called "Cosine Delta" is an especially high-performant stylometric distance measurement</ins>. <ins style="font-weight: bold; text-decoration: none;">We got our best results with </ins>Cosine Delta <ins style="font-weight: bold; text-decoration: none;">as well</ins>. <ins style="font-weight: bold; text-decoration: none;"> </ins></div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>In <del style="font-weight: bold; text-decoration: none;">einem ersten Schritt werden nur diejenigen Texte des Korpus analysiert</del>, <del style="font-weight: bold; text-decoration: none;">deren Übersetzer bekannt sind</del>. <del style="font-weight: bold; text-decoration: none;">Die Zahl der häufigsten Wörter, also </del>100, 200 <del style="font-weight: bold; text-decoration: none;">oder mehr, lässt sich </del>in <del style="font-weight: bold; text-decoration: none;">den meisten Implementierungen einstellen</del>. <del style="font-weight: bold; text-decoration: none;">Wir haben sehr gute Ergebnisse mit den häufigsten </del>150 <del style="font-weight: bold; text-decoration: none;">Wörtern der Texte erzielt</del>. <del style="font-weight: bold; text-decoration: none;">Jeder Text des Korpus wird intern durch einen Vektor dargestellt</del>, <del style="font-weight: bold; text-decoration: none;">der die standardisierten relativen Häufigkeiten dieser Wörter enthält</del>. <del style="font-weight: bold; text-decoration: none;">Der Abstand zwischen diesen Vektoren wird dann mit </del>Cosine Delta <del style="font-weight: bold; text-decoration: none;">berechnet</del>. <del style="font-weight: bold; text-decoration: none;">Der Computer formt dann Gruppen oder Cluster auf Basis dieser Abstände</del>, <del style="font-weight: bold; text-decoration: none;">die </del>in <del style="font-weight: bold; text-decoration: none;">einem Dendrogramm</del>, <del style="font-weight: bold; text-decoration: none;">einem Baum-Diagramm</del>, <del style="font-weight: bold; text-decoration: none;">visualisiert werden</del>. <del style="font-weight: bold; text-decoration: none;">Mithilfe dieses Verfahrens konnte der Computer im Korpus philosophischer Übersetzungen des 12</del>. <del style="font-weight: bold; text-decoration: none;">Jahrhunderts tatsächlich die Übersetzungen bekannter Übersetzer jeweils in eine Gruppe sortieren: die Gruppe der Übersetzungen des </del>Dominicus Gundisalvi, <del style="font-weight: bold; text-decoration: none;">des </del>Gerhard <del style="font-weight: bold; text-decoration: none;">von </del>Cremona etc. <del style="font-weight: bold; text-decoration: none;">Wenn das gelungen ist</del>, <del style="font-weight: bold; text-decoration: none;">ist die Methode sozusagen kalibriert</del>. </div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>In <ins style="font-weight: bold; text-decoration: none;">the first step</ins>, <ins style="font-weight: bold; text-decoration: none;">we analyzed the texts in the corpus with a known translator</ins>. <ins style="font-weight: bold; text-decoration: none;">The range for the most frequent words (</ins>100, 200 <ins style="font-weight: bold; text-decoration: none;">or more) can be adjusted </ins>in <ins style="font-weight: bold; text-decoration: none;">most implementations of the method</ins>. <ins style="font-weight: bold; text-decoration: none;">We got the results with the </ins>150 <ins style="font-weight: bold; text-decoration: none;">most frequent words in the texts</ins>. <ins style="font-weight: bold; text-decoration: none;">Each text of the corpus is processed as a vector</ins>, <ins style="font-weight: bold; text-decoration: none;">containin the standarized relative frequencies of these words</ins>. <ins style="font-weight: bold; text-decoration: none;">The distance between these vectors is calculated using </ins>Cosine Delta. <ins style="font-weight: bold; text-decoration: none;">After that, the computer forms groups or clusters based on these distances</ins>, <ins style="font-weight: bold; text-decoration: none;">which can then be visualized </ins>in <ins style="font-weight: bold; text-decoration: none;">a dendrogram. Using this method</ins>, <ins style="font-weight: bold; text-decoration: none;">the computer was indeed able to sort the texts with known translators into groups according to these translators</ins>, <ins style="font-weight: bold; text-decoration: none;">i</ins>.<ins style="font-weight: bold; text-decoration: none;">e</ins>. <ins style="font-weight: bold; text-decoration: none;">one group for the translations by </ins>Dominicus Gundisalvi, <ins style="font-weight: bold; text-decoration: none;">one for </ins>Gerhard <ins style="font-weight: bold; text-decoration: none;">of </ins>Cremona etc. <ins style="font-weight: bold; text-decoration: none;">Once this clustering succeeds, the method is calibratet</ins>, <ins style="font-weight: bold; text-decoration: none;">so to speak</ins>. </div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>In <del style="font-weight: bold; text-decoration: none;">einem zweiten Schritt werden dann die anonymen Übersetzungen dazu gegeben</del>. <del style="font-weight: bold; text-decoration: none;">Das daraus resultierende Dendrogramm muss sorgfältig interpretiert werden</del>: <del style="font-weight: bold; text-decoration: none;">Bleibt die </del>Gundisalvi-<del style="font-weight: bold; text-decoration: none;">Gruppe </del>(<del style="font-weight: bold; text-decoration: none;">oder </del>Gerhard-<del style="font-weight: bold; text-decoration: none;">Gruppe </del>etc.) <del style="font-weight: bold; text-decoration: none;">des kalibrierten Standards stabil und wird nur um die ein oder andere anonyme Übersetzung erweitert</del>, <del style="font-weight: bold; text-decoration: none;">dann ist es sehr wahrscheinlich</del>, <del style="font-weight: bold; text-decoration: none;">dass diese anonymen Übersetzungen tatsächlich von Gundisalvi produziert wurden</del>. <del style="font-weight: bold; text-decoration: none;">Zerfällt aber die Gundisalvi-Gruppe </del>(<del style="font-weight: bold; text-decoration: none;">oder Gerhard-Gruppe etc.</del>) in <del style="font-weight: bold; text-decoration: none;">mehrere Teilgruppen</del>, <del style="font-weight: bold; text-decoration: none;">die im Dendrogramm nicht mehr verbunden sind</del>, <del style="font-weight: bold; text-decoration: none;">gelingt dem Computer offensichtlich die Zuweisung der anonymen Übersetzung nicht</del>. </div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>In <ins style="font-weight: bold; text-decoration: none;">the second step, the anonymous translations are added to the system</ins>. <ins style="font-weight: bold; text-decoration: none;">The resulting dendograms have to be interpreted carefully</ins>: <ins style="font-weight: bold; text-decoration: none;">If the </ins>Gundisalvi-<ins style="font-weight: bold; text-decoration: none;">cluster </ins>(<ins style="font-weight: bold; text-decoration: none;">or the </ins>Gerhard-<ins style="font-weight: bold; text-decoration: none;">cluster </ins>etc.) <ins style="font-weight: bold; text-decoration: none;">remain stable and are merely expanded by additional anonymous translations, these text are likely produced by Gundisalvi. However</ins>, <ins style="font-weight: bold; text-decoration: none;">if the groups disperse</ins>, <ins style="font-weight: bold; text-decoration: none;">the computer is obviously unable to attribute the anonymous translations correctly</ins>.</div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div> </div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">Luckily, the results of method 1 (exclusive words) and 2 </ins>(<ins style="font-weight: bold; text-decoration: none;">MFW</ins>) <ins style="font-weight: bold; text-decoration: none;">mostly matched </ins>in <ins style="font-weight: bold; text-decoration: none;">our attempts</ins>, <ins style="font-weight: bold; text-decoration: none;">at least for the philosophical corpus. However</ins>, <ins style="font-weight: bold; text-decoration: none;">the astronomical/astrological corpus isn´t big enough for method 2 yet</ins>. <ins style="font-weight: bold; text-decoration: none;"> </ins></div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div><del style="font-weight: bold; text-decoration: none;">Bei unseren Versuchen zeigte aber erfreulicherweise, dass die Ergebnisse der Methode 1 (Exklusive Wörter) mit den Ergebnissen der Methode 2 (MFW) weitgehend übereinstimmten, zumindest beim philosophischen Korpus. Das astronomisch-astrologische Korpus ist für die Methode 2 allerdings noch nicht groß genug.</del></div></td><td colspan="2" class="diff-side-added"></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div><headertabs /></div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div><headertabs /></div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>{{Sprachauswahl|Tutorial for identification of anonymous arabic-latin translators (en)|Tutorial_Anonyme_Übersetzer}}</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>{{Sprachauswahl|Tutorial for identification of anonymous arabic-latin translators (en)|Tutorial_Anonyme_Übersetzer}}</div></td></tr>
</table>DeletedUserhttps://kallimachos.de/index.php?title=Tutorial_anonymus_translators_(en)&diff=3911&oldid=prevDeletedUser: /* Analysis */2017-09-05T15:55:17Z<p><span class="autocomment">Analysis</span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 5. September 2017, 15:55 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l40">Zeile 40:</td>
<td colspan="2" class="diff-lineno">Zeile 40:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>Once the texts are finally in an adjusted txt-format, the actual stylometric analysis can begin. The dataset can be devided in two groups, one with known and one with unknown translators. It´s important to keep up with the current state of research in this regard. When in doubt, the translation should rather be marked as "anonymus". For our research, we only accepted unambigous attributions found in the incipits and colophones of the manuscripts as reliable and marked all other texts as anonymous translations. </div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>Once the texts are finally in an adjusted txt-format, the actual stylometric analysis can begin. The dataset can be devided in two groups, one with known and one with unknown translators. It´s important to keep up with the current state of research in this regard. When in doubt, the translation should rather be marked as "anonymus". For our research, we only accepted unambigous attributions found in the incipits and colophones of the manuscripts as reliable and marked all other texts as anonymous translations. </div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div><del style="font-weight: bold; text-decoration: none;">Dieses Korpus lässt sich nun auf </del>(<del style="font-weight: bold; text-decoration: none;">mindestens</del>) <del style="font-weight: bold; text-decoration: none;">zwei verschiedene Weisen stilometrisch analysieren</del>: <del style="font-weight: bold; text-decoration: none;">Erstens im Hinblick auf ausschließlich von einem bekannten Übersetzer verwendete Wörter und zweitens computergestützt im Hinblick auf die </del>(<del style="font-weight: bold; text-decoration: none;">100, 200 o.ä.</del>) <del style="font-weight: bold; text-decoration: none;"> häufigsten Wörter eines Textes</del>. <del style="font-weight: bold; text-decoration: none;">Die erste Methode wurde in </del>Würzburg <del style="font-weight: bold; text-decoration: none;">entwickelt</del>, <del style="font-weight: bold; text-decoration: none;">die zweite basiert auf dem sogenannten </del>''Burrows Delta'' (Burrows 2002)<del style="font-weight: bold; text-decoration: none;">: </del></div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">The corpus can be anylized with </ins>(<ins style="font-weight: bold; text-decoration: none;">at least</ins>) <ins style="font-weight: bold; text-decoration: none;">two different methods</ins>: <ins style="font-weight: bold; text-decoration: none;">By looking at words that are used exclusively by one of the known translators and by computerized analysis of the most frequent words </ins>(<ins style="font-weight: bold; text-decoration: none;">MFW</ins>) <ins style="font-weight: bold; text-decoration: none;">of a text</ins>. <ins style="font-weight: bold; text-decoration: none;">The first method has been developed at </ins>Würzburg <ins style="font-weight: bold; text-decoration: none;">University</ins>, <ins style="font-weight: bold; text-decoration: none;">the second is based on </ins>''Burrows Delta'' (Burrows 2002)<ins style="font-weight: bold; text-decoration: none;">.</ins></div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>===(I) <del style="font-weight: bold; text-decoration: none;">Exklusive und zugleich häufig verwendete Wörter</del>===</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>===(I) <ins style="font-weight: bold; text-decoration: none;">Exclusive Words</ins>===</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div><del style="font-weight: bold; text-decoration: none;">Die Erfahrung hat gezeigt</del>, <del style="font-weight: bold; text-decoration: none;">dass anonyme Übersetzer anhand von häufig verwendeten, und fachunspezifischen Wörtern, die exklusiv von einem einzigen bekannten Übersetzer verwendet werden</del>, <del style="font-weight: bold; text-decoration: none;">identifiziert werden können. </del>Dominicus Gundisalvi <del style="font-weight: bold; text-decoration: none;">ist beispielsweise der einzige Übersetzer, der die Partikeln '</del>'sic ut, vel est, cuius comparatio, opus fuit, id per quod, id autem quod'' <del style="font-weight: bold; text-decoration: none;">und </del>''omnis quod est'' <del style="font-weight: bold; text-decoration: none;">verwendet</del>, <del style="font-weight: bold; text-decoration: none;">die sich auch </del>in <del style="font-weight: bold; text-decoration: none;">der anonymen Übersetzung </del>Alexander <del style="font-weight: bold; text-decoration: none;">von </del>Aphrodisias‘ ''De intellectu'' <del style="font-weight: bold; text-decoration: none;">finden </del>– <del style="font-weight: bold; text-decoration: none;">ein starker Hinweis darauf, dass </del>Gundisalvi <del style="font-weight: bold; text-decoration: none;">Übersetzer dieses Traktats war</del>. <del style="font-weight: bold; text-decoration: none;">Wie kommt man zu diesem Ergebnis? In folgenden zwei Schritten</del>:</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">Experience has shown that anonymous translators can be identified by looking at frequently used words</ins>, <ins style="font-weight: bold; text-decoration: none;">that are used exclusively by a single known translators and that are not dependent on the text´s discipline. As an example</ins>, Dominicus Gundisalvi <ins style="font-weight: bold; text-decoration: none;">is the only translator wo uses </ins>'sic ut, vel est, cuius comparatio, opus fuit, id per quod, id autem quod'' <ins style="font-weight: bold; text-decoration: none;">and </ins>''omnis quod est'', <ins style="font-weight: bold; text-decoration: none;">which can be also found </ins>in <ins style="font-weight: bold; text-decoration: none;">the anonymous translation of </ins>Alexander <ins style="font-weight: bold; text-decoration: none;">of </ins>Aphrodisias‘ ''De intellectu'' – <ins style="font-weight: bold; text-decoration: none;">a strong indication for </ins>Gundisalvi <ins style="font-weight: bold; text-decoration: none;">as the actual translator of the tractate</ins>. <ins style="font-weight: bold; text-decoration: none;">Getting there is a two-step process</ins>:</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>#<del style="font-weight: bold; text-decoration: none;">Der erste Schritt ist die Suche nach häufigen Termini</del>, <del style="font-weight: bold; text-decoration: none;">die exklusiv nur bei einem einzigen Übersetzer auftauchen</del>. <del style="font-weight: bold; text-decoration: none;">Dazu ist die Programmierung eines einfachen Suchprogramms sehr zu empfehlen. Beim Filtern der Wortlisten helfen </del>flexible <del style="font-weight: bold; text-decoration: none;">Parameter, die eine Mindesthäufigkeit der gesuchten Wörter festlegen oder den Anteil der Texte eines Übersetzers bestimmen, </del>in <del style="font-weight: bold; text-decoration: none;">denen die Wörter jeweils mindestens vorkommen müssen</del>. <del style="font-weight: bold; text-decoration: none;">Um auch Wortgruppen zu analysieren</del>, <del style="font-weight: bold; text-decoration: none;">können die Texte in Listen von </del>n-<del style="font-weight: bold; text-decoration: none;">Grammen, d</del>.<del style="font-weight: bold; text-decoration: none;">h</del>. <del style="font-weight: bold; text-decoration: none;">überlappenden Abfolgen mehrerer Wörter aufgeteilt werden</del>. <del style="font-weight: bold; text-decoration: none;">Damit kann die Menge der für einen Autor exklusiven Wörter auf typische und häufig verwendete Wörter reduziert werden</del>, <del style="font-weight: bold; text-decoration: none;">zum Beispiel auf Wörter</del>, <del style="font-weight: bold; text-decoration: none;">die mindestens </del>10 <del style="font-weight: bold; text-decoration: none;">mal </del>in <del style="font-weight: bold; text-decoration: none;">den Werken eines Übersetzers und in mindestens </del>40% <del style="font-weight: bold; text-decoration: none;">seiner Übersetzungen erscheinen</del>. <del style="font-weight: bold; text-decoration: none;">Zum Beispiel erscheint die Wortverbindung </del>''iterum quia'' in 4 <del style="font-weight: bold; text-decoration: none;">der </del>10 <del style="font-weight: bold; text-decoration: none;">Übersetzungen </del>Gerhards <del style="font-weight: bold; text-decoration: none;">von </del>Cremona<del style="font-weight: bold; text-decoration: none;">, die sich </del>in <del style="font-weight: bold; text-decoration: none;">unserem philosophischen Korpus finden</del>, <del style="font-weight: bold; text-decoration: none;">und dort insgesamt </del>56 <del style="font-weight: bold; text-decoration: none;">mal</del>. <del style="font-weight: bold; text-decoration: none;">Es handelt sich also um eine zugleich exklusiv und häufig gebrauchte Wortverbindung bei Gerhard von Cremona</del>. <del style="font-weight: bold; text-decoration: none;">Um dem Verdacht auf mögliche Falschzuschreibungen oder die Zusammenarbeit von Übersetzern nachzugehen, kann zudem ein Parameter eingeführt werden</del>, <del style="font-weight: bold; text-decoration: none;">der eine gewisse Anzahl </del>an <del style="font-weight: bold; text-decoration: none;">Fehlern zulässt</del>, <del style="font-weight: bold; text-decoration: none;">d.h. Wörter</del>, <del style="font-weight: bold; text-decoration: none;">die einige Male eben doch auch von anderen Übersetzern verwendet werden</del>. </div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>#<ins style="font-weight: bold; text-decoration: none;">The first step is searching for frequent terms that are used exclusively by a single translator. To this end</ins>, <ins style="font-weight: bold; text-decoration: none;">programing a simple search enginge is advisable</ins>. <ins style="font-weight: bold; text-decoration: none;">When filtering the word lists, </ins>flexible <ins style="font-weight: bold; text-decoration: none;">parameters can help to set a minimal frequency or the amount of texts that have to contain the word </ins>in <ins style="font-weight: bold; text-decoration: none;">question</ins>. <ins style="font-weight: bold; text-decoration: none;">To analyze word groups</ins>, <ins style="font-weight: bold; text-decoration: none;">the texts can be split into lists of </ins>n-<ins style="font-weight: bold; text-decoration: none;">grams (i</ins>.<ins style="font-weight: bold; text-decoration: none;">e</ins>. <ins style="font-weight: bold; text-decoration: none;">overlapping sequences of multiple words)</ins>. <ins style="font-weight: bold; text-decoration: none;">Thus</ins>, <ins style="font-weight: bold; text-decoration: none;">the list of exclusive words can be reduced to typical and frequently used terms</ins>, <ins style="font-weight: bold; text-decoration: none;">f.i. words that appear at least in </ins>10 <ins style="font-weight: bold; text-decoration: none;">works of the translator and </ins>in 40% <ins style="font-weight: bold; text-decoration: none;">of his translations</ins>. <ins style="font-weight: bold; text-decoration: none;">As an example, the term </ins>''iterum quia'' <ins style="font-weight: bold; text-decoration: none;">appears </ins>in 4 <ins style="font-weight: bold; text-decoration: none;">of the </ins>10 <ins style="font-weight: bold; text-decoration: none;">translations by </ins>Gerhards <ins style="font-weight: bold; text-decoration: none;">of </ins>Cremona in <ins style="font-weight: bold; text-decoration: none;">our philosophical corpus</ins>, <ins style="font-weight: bold; text-decoration: none;">where they are used a total of </ins>56 <ins style="font-weight: bold; text-decoration: none;">times</ins>. <ins style="font-weight: bold; text-decoration: none;">Thus, ''iterum quia'' is both an exclusive and frequently used term in Gerhards work</ins>. <ins style="font-weight: bold; text-decoration: none;">Following a possible suspicion for a a false attribution</ins>, an <ins style="font-weight: bold; text-decoration: none;">additional parameter for error tolerance can be employded</ins>, <ins style="font-weight: bold; text-decoration: none;">admitting also words</ins>, <ins style="font-weight: bold; text-decoration: none;">that are used very rarely by other translators</ins>.</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>#<del style="font-weight: bold; text-decoration: none;">Aus dieser Liste müssen dann in einem zweiten Schritt per Hand diejenigen Wörter ausgesiebt werden</del>, <del style="font-weight: bold; text-decoration: none;">die Inhaltswörter sind</del>, <del style="font-weight: bold; text-decoration: none;">wie z.B. </del>''substantia composita'' oder ''horoscopus''<del style="font-weight: bold; text-decoration: none;">, die typisch für eine bestimmte Fachdisziplin oder Subdisziplin sind wie Metaphysik oder Astrologie</del>. <del style="font-weight: bold; text-decoration: none;">Übrig bleiben stilistische Wörter </del>in <del style="font-weight: bold; text-decoration: none;">einem weiteren Sinn</del>, <del style="font-weight: bold; text-decoration: none;">d</del>.<del style="font-weight: bold; text-decoration: none;">h</del>. <del style="font-weight: bold; text-decoration: none;">solche Wörter, die sich im Prinzip </del>in <del style="font-weight: bold; text-decoration: none;">allen wissenschaftlichen lateinischen Texten des Zeitraums finden lassen könnten</del>, also <del style="font-weight: bold; text-decoration: none;">nicht nur Konjunktionen oder andere Partikeln, sondern auch Wörter und Wortverbindungen wie </del>''examinatio'', ''annullare'' <del style="font-weight: bold; text-decoration: none;"> oder </del>''demonstrare voluimus''<del style="font-weight: bold; text-decoration: none;">, die nicht fachspezifisch sind</del>. <del style="font-weight: bold; text-decoration: none;">Dieser Fokus ist wichtig, weil die Erfahrung zeigt</del>, <del style="font-weight: bold; text-decoration: none;">dass Inhaltswörter leichter von anderen Übersetzern übernommen werden</del>, <del style="font-weight: bold; text-decoration: none;">während stilistische Wörter und Wortverbindungen stabiler nur bei einem Autor erscheinen</del>. </div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>#<ins style="font-weight: bold; text-decoration: none;">In a second step</ins>, <ins style="font-weight: bold; text-decoration: none;">this list has to be filtered for content words specific to the text´s discipline</ins>, <ins style="font-weight: bold; text-decoration: none;">like </ins>''substantia composita'' oder ''horoscopus''. <ins style="font-weight: bold; text-decoration: none;"> The remaining words a stilistic words </ins>in <ins style="font-weight: bold; text-decoration: none;">a more narrow sense</ins>, <ins style="font-weight: bold; text-decoration: none;">i</ins>.<ins style="font-weight: bold; text-decoration: none;">e</ins>. <ins style="font-weight: bold; text-decoration: none;">words that can be used in all scientific latin texts of this perios </ins>in <ins style="font-weight: bold; text-decoration: none;">principle. These may contain not only conjunctions and other particle words</ins>, <ins style="font-weight: bold; text-decoration: none;">but </ins>also <ins style="font-weight: bold; text-decoration: none;">words and phrases like </ins>''examinatio'', ''annullare'' <ins style="font-weight: bold; text-decoration: none;">or </ins>''demonstrare voluimus''. <ins style="font-weight: bold; text-decoration: none;">This focus is important</ins>, <ins style="font-weight: bold; text-decoration: none;">as experience has shown that content words are adopted by other translators more easily</ins>, <ins style="font-weight: bold; text-decoration: none;">whereas stilistic words and phrases appear more stable for one author only</ins>.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div><del style="font-weight: bold; text-decoration: none;">Anschließend kann man für jeden anonym übersetzten Text des Korpus notieren, welche dieser exklusiven und häufig gebrauchten Wörter der Übersetzer in den anonym übersetzten Texten erscheint. Verbindet sich positive und negative Evidenz – wenn sich also (positiv) eine ganze Reihe von solchen exklusiven Wörter eines Übersetzers in einem anonymen Text findet und gleichzeitig keine (negativ) der exklusiven Wörter der anderen Übersetzer –, dann ist die Zuschreibung eines Textes an den bekannten Übersetzer sehr sicher.</del></div></td><td colspan="2" class="diff-side-added"></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div><del style="font-weight: bold; text-decoration: none;">Bei sehr kurzen anonym übersetzten Texten kann es sich lohnen</del>, <del style="font-weight: bold; text-decoration: none;">auch seltenere stilistische Wörter systematisch zu untersuchen</del>, <del style="font-weight: bold; text-decoration: none;">also z</del>.<del style="font-weight: bold; text-decoration: none;">B</del>. <del style="font-weight: bold; text-decoration: none;">solche</del>, <del style="font-weight: bold; text-decoration: none;">die weniger als 10 Mal und </del>in <del style="font-weight: bold; text-decoration: none;">weniger als 40% der Übersetzungen eines Übersetzers erscheinen. Eine solche Analyse muss aber systematisch auch die selteneren Wörter der anderen Übersetzer vergleichen</del>. <del style="font-weight: bold; text-decoration: none;"> Die Erfahrung zeigt</del>, <del style="font-weight: bold; text-decoration: none;">dass nur eine Massierung solcher weniger typischen Wörter und Wortverbindungen </del>in <del style="font-weight: bold; text-decoration: none;">einem anonymen Text wirklich eine Übersetzerattribuierung erlaubt</del>.</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">Subsequently</ins>, <ins style="font-weight: bold; text-decoration: none;">you can note for each anonymously translated text in the corpus which of these words appear in the text. If negative and positive evidence fit – meaning when a bunch of words exclusive to a single translator appear in the text (positive) and at the same time no exclusive words of other translators (negative)</ins>, <ins style="font-weight: bold; text-decoration: none;">the attribution of the text to the known translator is quite certain</ins>.</div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div> </div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">For very short texts, it my be advisibale to expand the analysis to less frequent words</ins>. <ins style="font-weight: bold; text-decoration: none;">However</ins>, in <ins style="font-weight: bold; text-decoration: none;">this case, the less frequent words of other translators have to be compared as well</ins>. <ins style="font-weight: bold; text-decoration: none;">Experience shows</ins>, <ins style="font-weight: bold; text-decoration: none;">that only a huge amassment of these less typical words and phrases </ins>in <ins style="font-weight: bold; text-decoration: none;">an anonymously translated text allow for a credible attribution</ins>.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>===(II) Computergestützte Stilometrie mit ''Burrows Delta''===</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>===(II) Computergestützte Stilometrie mit ''Burrows Delta''===</div></td></tr>
</table>DeletedUserhttps://kallimachos.de/index.php?title=Tutorial_anonymus_translators_(en)&diff=3910&oldid=prevDeletedUser: /* Analysis */2017-09-05T12:54:47Z<p><span class="autocomment">Analysis</span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 5. September 2017, 12:54 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l38">Zeile 38:</td>
<td colspan="2" class="diff-lineno">Zeile 38:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>=Analysis=</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>=Analysis=</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Analysis==</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Analysis==</div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div><del style="font-weight: bold; text-decoration: none;">Liegen die Texte des Korpus </del>in <del style="font-weight: bold; text-decoration: none;">derart bereinigten </del>txt-<del style="font-weight: bold; text-decoration: none;">Formaten vor</del>, <del style="font-weight: bold; text-decoration: none;">kann die eigentliche stilometrische Arbeit beginnen</del>. <del style="font-weight: bold; text-decoration: none;">Die Dateien mit den Übersetzungen lassen sich </del>in <del style="font-weight: bold; text-decoration: none;">verschiedene Gruppen sortieren: diejenigen mit unbekannten und diejenigen mit bekannten Übersetzern</del>. <del style="font-weight: bold; text-decoration: none;">Dabei sollte unbedingt der neueste Forschungsstand berücksichtigt werden</del>. <del style="font-weight: bold; text-decoration: none;">Im Zweifelsfall sollte eine Übersetzung lieber als „anonym“ gekennzeichnet werden</del>. <del style="font-weight: bold; text-decoration: none;">In unseren Studien haben wir nur die eindeutigen Übersetzerzuschreibungen</del>, <del style="font-weight: bold; text-decoration: none;">die sich </del>in <del style="font-weight: bold; text-decoration: none;">den Incipits und Kolophonen der Handschriften finden, als verlässlich akzeptiert und alle anderen Texte als anonyme Übersetzungen gekennzeichnet</del>.</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">Once the texts are finally </ins>in <ins style="font-weight: bold; text-decoration: none;">an adjusted </ins>txt-<ins style="font-weight: bold; text-decoration: none;">format</ins>, <ins style="font-weight: bold; text-decoration: none;">the actual stylometric analysis can begin</ins>. <ins style="font-weight: bold; text-decoration: none;">The dataset can be devided </ins>in <ins style="font-weight: bold; text-decoration: none;">two groups, one with known and one with unknown translators</ins>. <ins style="font-weight: bold; text-decoration: none;">It´s important to keep up with the current state of research in this regard</ins>. <ins style="font-weight: bold; text-decoration: none;">When in doubt, the translation should rather be marked as "anonymus"</ins>. <ins style="font-weight: bold; text-decoration: none;">For our research</ins>, <ins style="font-weight: bold; text-decoration: none;">we only accepted unambigous attributions found </ins>in <ins style="font-weight: bold; text-decoration: none;">the incipits and colophones of the manuscripts as reliable and marked all other texts as anonymous translations</ins>. </div></td></tr>
<tr><td colspan="2" class="diff-side-deleted"></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div> </div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>Dieses Korpus lässt sich nun auf (mindestens) zwei verschiedene Weisen stilometrisch analysieren: Erstens im Hinblick auf ausschließlich von einem bekannten Übersetzer verwendete Wörter und zweitens computergestützt im Hinblick auf die (100, 200 o.ä.) häufigsten Wörter eines Textes. Die erste Methode wurde in Würzburg entwickelt, die zweite basiert auf dem sogenannten ''Burrows Delta'' (Burrows 2002): </div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>Dieses Korpus lässt sich nun auf (mindestens) zwei verschiedene Weisen stilometrisch analysieren: Erstens im Hinblick auf ausschließlich von einem bekannten Übersetzer verwendete Wörter und zweitens computergestützt im Hinblick auf die (100, 200 o.ä.) häufigsten Wörter eines Textes. Die erste Methode wurde in Würzburg entwickelt, die zweite basiert auf dem sogenannten ''Burrows Delta'' (Burrows 2002): </div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
</table>DeletedUserhttps://kallimachos.de/index.php?title=Tutorial_anonymus_translators_(en)&diff=3909&oldid=prevDeletedUser: /* Processing the texts for comparative analysis */2017-09-05T12:48:40Z<p><span class="autocomment">Processing the texts for comparative analysis</span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 5. September 2017, 12:48 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l34">Zeile 34:</td>
<td colspan="2" class="diff-lineno">Zeile 34:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>If the analysis program is configured correctly, it should be able to recogniza large portions of the texts as orthographically correct latin. Unrecognized words can be routinely replaced by their classical counterparts via a progressively adjusted ruleset. Usefull replacement rules are f.i. ci/ti, diff/def, ch/c etc., but also typical OCR mistakes like ic/it, ee/ec, b/h etc. </div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>If the analysis program is configured correctly, it should be able to recogniza large portions of the texts as orthographically correct latin. Unrecognized words can be routinely replaced by their classical counterparts via a progressively adjusted ruleset. Usefull replacement rules are f.i. ci/ti, diff/def, ch/c etc., but also typical OCR mistakes like ic/it, ee/ec, b/h etc. </div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>For a usable stylometric analysis, at least 95% of the words in the processed text should be recognized as correct latin by the reference lexica. However, 100% recognition should be the goal. To help with the correction of the latin texts, it may be advisable to program simple comparison and input masks, allowing the user to directly compare the words in question with the word in the original scan and correct them on the spot. Furthermore, it is advisable to expand the employed <del style="font-weight: bold; text-decoration: none;">lexicons </del>by custom wort lists to cover the specific vocabulary of arabic-latin translations and the corresponding disciplines.</div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>For a usable stylometric analysis, at least 95% of the words in the processed text should be recognized as correct latin by the reference lexica. However, 100% recognition should be the goal. To help with the correction of the latin texts, it may be advisable to program simple comparison and input masks, allowing the user to directly compare the words in question with the word in the original scan and correct them on the spot. Furthermore, it is advisable to expand the employed <ins style="font-weight: bold; text-decoration: none;">dictionariess </ins>by custom wort lists to cover the specific vocabulary of arabic-latin translations and the corresponding disciplines.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>=Analysis=</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>=Analysis=</div></td></tr>
</table>DeletedUserhttps://kallimachos.de/index.php?title=Tutorial_anonymus_translators_(en)&diff=3908&oldid=prevDeletedUser: /* Processing the texts for comparative analysis */2017-09-05T12:48:04Z<p><span class="autocomment">Processing the texts for comparative analysis</span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 5. September 2017, 12:48 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l24">Zeile 24:</td>
<td colspan="2" class="diff-lineno">Zeile 24:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>The citable text (3) isn´t usable for stylometry yet, but can be useful for other scientific tasks. Of course, to be able to compare texts using stylometry, they need to be made comparable beforehand. In the field of medieval editions, punctuation rules and orthography are major obstacles, for the punctuation rules often vary according to the national customs of the editors (german, french, english etc.), while the "signal" of the author ist lost. In turn, the orthography ranges from "classizied" editions (e.g. Avicenna Latinus) to the faithful reproduction of the exact orthography of a single medieval manuscript. These problems can be mitigated by radically removing all punctuation marks, changing all uppercase letters to lowercase letters und finally classizying the orthography. The last step is quite painfull for medievalist, but theres is no better alternative. As a first step, it is f.i. helpfull to replace all v with u and all j with i.</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>The citable text (3) isn´t usable for stylometry yet, but can be useful for other scientific tasks. Of course, to be able to compare texts using stylometry, they need to be made comparable beforehand. In the field of medieval editions, punctuation rules and orthography are major obstacles, for the punctuation rules often vary according to the national customs of the editors (german, french, english etc.), while the "signal" of the author ist lost. In turn, the orthography ranges from "classizied" editions (e.g. Avicenna Latinus) to the faithful reproduction of the exact orthography of a single medieval manuscript. These problems can be mitigated by radically removing all punctuation marks, changing all uppercase letters to lowercase letters und finally classizying the orthography. The last step is quite painfull for medievalist, but theres is no better alternative. As a first step, it is f.i. helpfull to replace all v with u and all j with i.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>This process can be digitally enhanced by asking digital latin reference lexica if they can recognize words in the texts of the corpus. The easiest approach is the comparison with a latin word list. (f.i. '''[https://github.com/cisocrgroup/Resources/tree/master/lexica/latin here]''' or in the word list of the '''[http://extensions.openoffice.org/en/project/latin-spelling-and-hyphenation-dictionaries OpenOffice lexicon]''', which can also be used in a Python script via PyEnchant) or the use of a morphology programm, able to lemmatize and kategorize every word in the text<del style="font-weight: bold; text-decoration: none;">. </del></div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>This process can be digitally enhanced by asking digital latin reference lexica if they can recognize words in the texts of the corpus. The easiest approach is the comparison with a latin word list. (f.i. '''[https://github.com/cisocrgroup/Resources/tree/master/lexica/latin here]''' or in the word list of the '''[http://extensions.openoffice.org/en/project/latin-spelling-and-hyphenation-dictionaries OpenOffice lexicon]''', which can also be used in a Python script via PyEnchant) or the use of a morphology programm, <ins style="font-weight: bold; text-decoration: none;">which is </ins>able to lemmatize and kategorize every word in the text <ins style="font-weight: bold; text-decoration: none;">and look them up in a dictionary </ins></div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>For the latter, there are currently two open-source solutions: </div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>For the latter, there are currently two open-source solutions: </div></td></tr>
</table>DeletedUserhttps://kallimachos.de/index.php?title=Tutorial_anonymus_translators_(en)&diff=3904&oldid=prevDeletedUser: /* Processing the texts for comparative analysis */2017-09-05T12:47:17Z<p><span class="autocomment">Processing the texts for comparative analysis</span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 5. September 2017, 12:47 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l24">Zeile 24:</td>
<td colspan="2" class="diff-lineno">Zeile 24:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>The citable text (3) isn´t usable for stylometry yet, but can be useful for other scientific tasks. Of course, to be able to compare texts using stylometry, they need to be made comparable beforehand. In the field of medieval editions, punctuation rules and orthography are major obstacles, for the punctuation rules often vary according to the national customs of the editors (german, french, english etc.), while the "signal" of the author ist lost. In turn, the orthography ranges from "classizied" editions (e.g. Avicenna Latinus) to the faithful reproduction of the exact orthography of a single medieval manuscript. These problems can be mitigated by radically removing all punctuation marks, changing all uppercase letters to lowercase letters und finally classizying the orthography. The last step is quite painfull for medievalist, but theres is no better alternative. As a first step, it is f.i. helpfull to replace all v with u and all j with i.</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>The citable text (3) isn´t usable for stylometry yet, but can be useful for other scientific tasks. Of course, to be able to compare texts using stylometry, they need to be made comparable beforehand. In the field of medieval editions, punctuation rules and orthography are major obstacles, for the punctuation rules often vary according to the national customs of the editors (german, french, english etc.), while the "signal" of the author ist lost. In turn, the orthography ranges from "classizied" editions (e.g. Avicenna Latinus) to the faithful reproduction of the exact orthography of a single medieval manuscript. These problems can be mitigated by radically removing all punctuation marks, changing all uppercase letters to lowercase letters und finally classizying the orthography. The last step is quite painfull for medievalist, but theres is no better alternative. As a first step, it is f.i. helpfull to replace all v with u and all j with i.</div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>This process can be digitally enhanced by asking digital latin reference lexica if they can recognize words in the texts of the corpus. The easiest approach is the comparison with a latin word list. (f.i. '''[https://github.com/cisocrgroup/Resources/tree/master/lexica/latin here]''' or in the word list of the '''[http://extensions.openoffice.org/en/project/latin-spelling-and-hyphenation-dictionaries OpenOffice<del style="font-weight: bold; text-decoration: none;">-</del>lexicon]''', which can also be used in a Python script via PyEnchant) or the use of a morphology programm, able to lemmatize and kategorize every word in the text. </div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>This process can be digitally enhanced by asking digital latin reference lexica if they can recognize words in the texts of the corpus. The easiest approach is the comparison with a latin word list. (f.i. '''[https://github.com/cisocrgroup/Resources/tree/master/lexica/latin here]''' or in the word list of the '''[http://extensions.openoffice.org/en/project/latin-spelling-and-hyphenation-dictionaries OpenOffice lexicon]''', which can also be used in a Python script via PyEnchant) or the use of a morphology programm, able to lemmatize and kategorize every word in the text. </div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>For the latter, there are currently two open-source solutions: </div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>For the latter, there are currently two open-source solutions: </div></td></tr>
</table>DeletedUserhttps://kallimachos.de/index.php?title=Tutorial_anonymus_translators_(en)&diff=3902&oldid=prevDeletedUser: /* Processing the texts for comparative analysis */2017-09-05T12:47:06Z<p><span class="autocomment">Processing the texts for comparative analysis</span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 5. September 2017, 12:47 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l32">Zeile 32:</td>
<td colspan="2" class="diff-lineno">Zeile 32:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>Both programs are quite complex and may often require some effort to compile correctly, especially if you want to integrate these programms into your own scripts using a wrapper. As an easier alternative, at least for some tests, the according web services ([http://services.perseids.org/bsp/morphologyservice/analysis/word?lang=lat&engine=morpheuslat&word=et example]) can be used as well. </div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>Both programs are quite complex and may often require some effort to compile correctly, especially if you want to integrate these programms into your own scripts using a wrapper. As an easier alternative, at least for some tests, the according web services ([http://services.perseids.org/bsp/morphologyservice/analysis/word?lang=lat&engine=morpheuslat&word=et example]) can be used as well. </div></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>If the analysis program is configured correctly, it should be able to recogniza large portions of the texts as orthographically correct latin. Unrecognized words can be routinely replaced by <del style="font-weight: bold; text-decoration: none;">they </del>classical counterparts via a progressively adjusted ruleset. Usefull replacement rules are f.i. ci/ti, diff/def, ch/c etc., but also typical OCR mistakes like ic/it, ee/ec, b/h etc. </div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>If the analysis program is configured correctly, it should be able to recogniza large portions of the texts as orthographically correct latin. Unrecognized words can be routinely replaced by <ins style="font-weight: bold; text-decoration: none;">their </ins>classical counterparts via a progressively adjusted ruleset. Usefull replacement rules are f.i. ci/ti, diff/def, ch/c etc., but also typical OCR mistakes like ic/it, ee/ec, b/h etc. </div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>For a usable stylometric analysis, at least 95% of the words in the processed text should be recognized as correct latin by the reference lexica. However, 100% recognition should be the goal. To help with the correction of the latin texts, it may be advisable to program simple comparison and input masks, allowing the user to directly compare the words in question with the word in the original scan and correct them on the spot. Furthermore, it is advisable to expand the employed lexicons by custom wort lists to cover the specific vocabulary of arabic-latin translations and the corresponding disciplines.</div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>For a usable stylometric analysis, at least 95% of the words in the processed text should be recognized as correct latin by the reference lexica. However, 100% recognition should be the goal. To help with the correction of the latin texts, it may be advisable to program simple comparison and input masks, allowing the user to directly compare the words in question with the word in the original scan and correct them on the spot. Furthermore, it is advisable to expand the employed lexicons by custom wort lists to cover the specific vocabulary of arabic-latin translations and the corresponding disciplines.</div></td></tr>
</table>DeletedUserhttps://kallimachos.de/index.php?title=Tutorial_anonymus_translators_(en)&diff=3915&oldid=prevDeletedUser: /* Processing the texts for comparative analysis */2017-09-05T12:44:18Z<p><span class="autocomment">Processing the texts for comparative analysis</span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 5. September 2017, 12:44 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l31">Zeile 31:</td>
<td colspan="2" class="diff-lineno">Zeile 31:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>Both programs are quite complex and may often require some effort to compile correctly<del style="font-weight: bold; text-decoration: none;">; even more</del>, if you want to integrate these programms into your own scripts. As an easier alternative, at least for some tests, the according web services ([http://services.perseids.org/bsp/morphologyservice/analysis/word?lang=lat&engine=morpheuslat&word=et example]) can be used as well. </div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>Both programs are quite complex and may often require some effort to compile correctly, <ins style="font-weight: bold; text-decoration: none;">especially </ins>if you want to integrate these programms into your own scripts <ins style="font-weight: bold; text-decoration: none;">using a wrapper</ins>. As an easier alternative, at least for some tests, the according web services ([http://services.perseids.org/bsp/morphologyservice/analysis/word?lang=lat&engine=morpheuslat&word=et example]) can be used as well. </div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>If the analysis program is configured correctly, it should be able to recogniza large portions of the texts as orthographically correct latin. Unrecognized words can be routinely replaced by they classical counterparts via a progressively adjusted ruleset. Usefull replacement rules are f.i. ci/ti, diff/def, ch/c etc., but also typical OCR mistakes like ic/it, ee/ec, b/h etc. </div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>If the analysis program is configured correctly, it should be able to recogniza large portions of the texts as orthographically correct latin. Unrecognized words can be routinely replaced by they classical counterparts via a progressively adjusted ruleset. Usefull replacement rules are f.i. ci/ti, diff/def, ch/c etc., but also typical OCR mistakes like ic/it, ee/ec, b/h etc. </div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
</table>DeletedUserhttps://kallimachos.de/index.php?title=Tutorial_anonymus_translators_(en)&diff=3919&oldid=prevDeletedUser: /* Processing the texts for comparative analysis */2017-09-05T12:43:57Z<p><span class="autocomment">Processing the texts for comparative analysis</span></p>
<table style="background-color: #fff; color: #202122;" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="de">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Nächstältere Version</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Version vom 5. September 2017, 12:43 Uhr</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l31">Zeile 31:</td>
<td colspan="2" class="diff-lineno">Zeile 31:</td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
<tr><td class="diff-marker" data-marker="−"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>Both programs are quite complex and may often require some effort to compile correctly <del style="font-weight: bold; text-decoration: none;">- </del>even more, if you want to integrate these programms into your own scripts. As an easier alternative, at least for some tests, the according web services ([http://services.perseids.org/bsp/morphologyservice/analysis/word?lang=lat&engine=morpheuslat&word=et example]) can be used as well. </div></td><td class="diff-marker" data-marker="+"></td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>Both programs are quite complex and may often require some effort to compile correctly<ins style="font-weight: bold; text-decoration: none;">; </ins>even more, if you want to integrate these programms into your own scripts. As an easier alternative, at least for some tests, the according web services ([http://services.perseids.org/bsp/morphologyservice/analysis/word?lang=lat&engine=morpheuslat&word=et example]) can be used as well. </div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>If the analysis program is configured correctly, it should be able to recogniza large portions of the texts as orthographically correct latin. Unrecognized words can be routinely replaced by they classical counterparts via a progressively adjusted ruleset. Usefull replacement rules are f.i. ci/ti, diff/def, ch/c etc., but also typical OCR mistakes like ic/it, ee/ec, b/h etc. </div></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>If the analysis program is configured correctly, it should be able to recogniza large portions of the texts as orthographically correct latin. Unrecognized words can be routinely replaced by they classical counterparts via a progressively adjusted ruleset. Usefull replacement rules are f.i. ci/ti, diff/def, ch/c etc., but also typical OCR mistakes like ic/it, ee/ec, b/h etc. </div></td></tr>
<tr><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td><td class="diff-marker"></td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><br></td></tr>
</table>DeletedUser