Automatically Creating a Lexicon of Verbal Polarity Shifters: Mono- and Cross-lingual Methods for German

5306

Authors

nan

Abstract

AbstractIn this paper we use methods for creating a large lexicon of verbal polarity shifters and apply them to German. Polarity shifters are content words that can move the polarity of a phrase towards its opposite, such as the verb “abandon” in “abandon all hope”. This is similar to how negation words like “not” can influence polarity. Both shifters and negation are required for high precision sentiment analysis. Lists of negation words are available for many languages, but the only language for which a sizable lexicon of verbal polarity shifters exists is English. This lexicon was created by bootstrapping a sample of annotated verbs with a supervised classifier that uses a set of data- and resource-driven features. We reproduce and adapt this approach to create a German lexicon of verbal polarity shifters. Thereby, we confirm that the approach works for multiple languages. We further improve classification by leveraging cross-lingual information from the English shifter lexicon. Using this improved approach, we bootstrap a large number of German verbal polarity shifters, reducing the annotation effort drastically. The resulting German lexicon of verbal polarity shifters is made publicly available.

Keywords

nan

Full Content

Proceedings of the 27th International Conference on Computational Linguistics, pages 2516–2528 Santa Fe, New Mexico, USA, August 20-26, 2018. 2516 Automatically Creating a Lexicon of Verbal Polarity Shifters: Mono- and Cross-lingual Methods for German Marc Schulder, Michael Wiegand Spoken Language Systems Saarland University Germany marc.schulder@lsv.uni-saarland.de michael.wiegand@lsv.uni-saarland.de Josef Ruppenhofer Institute for German Language Mannheim Germany ruppenhofer@ids-mannheim.de Abstract In this paper we use methods for creating a large lexicon of verbal polarity shifters and apply them to German. Polarity shifters are content words that can move the polarity of a phrase towards its opposite, such as the verb “abandon” in “abandon all hope”. This is similar to how negation words like “not” can influence polarity. Both shifters and negation are required for high precision sentiment analysis. Lists of negation words are available for many languages, but the only language for which a sizable lexicon of verbal polarity shifters exists is English. This lexicon was created by bootstrapping a sample of annotated verbs with a supervised classifier that uses a set of data- and resource-driven features. We reproduce and adapt this approach to create a German lexicon of verbal polarity shifters. Thereby, we confirm that the approach works for multiple languages. We further improve classification by leveraging cross-lingual information from the English shifter lexicon. Using this improved approach, we bootstrap a large number of German verbal polarity shifters, reducing the annotation effort drastically. The resulting German lexicon of verbal polarity shifters is made publicly available. Title and Abstract in German Die automatische Erstellung eines Lexikons polaritätsverschiebender Verben: Einsprachige und sprachübergreifende Methoden für das Deutsche In dieser Arbeit untersuchen wir Methoden zur Erstellung eines deutschsprachigen Lexikons po- laritätsverschiebender Verben. Diese Verben, die vielfach auch Polaritätsshifter genannt werden, sind Inhaltswörter, die die Polarität einer Phrase zu ihrem entgegengesetzten Wert verschieben, wie z.B. das Verb „aufgeben“ in der Verbalphrase „alle Hoffnung aufgeben“. Das Verhalten von Polaritätsshiftern ähnelt somit dem von Negationswörtern wie „nicht“. Für robuste Sentimen- tanalyse werden sowohl Negationswörter als auch Polaritätsshifter benötigt. Während Listen von Negationswörtern in vielen Sprachen verfügbar sind, existiert jedoch ein Polaritätsshifter- Lexikon hinreichender Größe nur für das Englische. Jene Ressource wurde mittels Bootstrapping erzeugt, indem ein überwachter Klassifikator auf einer kleinen Stichprobe von Verben trainiert wurde. Dieser Klassifikator nutzt Daten- und Ressourcen-getriebene Merkmale. Wir repro- duzieren diesen Ansatz und passen ihn soweit notwendig für das Deutsche an. Somit weisen wir die Übertragbarkeit dieses Ansatzes auf andere Sprachen nach. Wir verbessern die Qualität der Klassifikation zudem weiterhin, indem wir Informationen aus dem existierenden englischen Polaritätsshifter-Lexikon nutzten. Mittels dieses verbesserten Ansatzes finden wir per Bootstrap- ping eine große Anzahl deutscher Polaritätsshifter und verringern somit deutlich den manuellen Annotationsaufwand. Das resultierende deutsche Lexikon polaritätsverschiebender Verben ist frei verfügbar. This work is licensed under a Creative Commons Attribution 4.0 International License. License details: http://creativecommons.org/licenses/by/4.0/ 2517 1 Introduction Polarity shifters are content words such as verbs, nouns or adjectives that influence the sentiment polarity of an expression in ways similar to negation words. For example, the negated statement in (1) that uses the negation word …

Content truncated to 500 words

Labeled Text

AbstractIn this paper we use methods for creating a large lexicon of verbal polarity shifters and apply them to German. Polarity shifters are content words that can move the polarity of a phrase towards its opposite, such as the verb “abandon” in “abandon all hope”. This is similar to how negation words like “not” can influence polarity. Both shifters and negation are required for high precision sentiment analysis. Lists of negation words are available for many languages, but the only language for which a sizable lexicon of verbal polarity shifters exists is English. This lexicon was created by bootstrapping a sample of annotated verbs with a supervised classifier that uses a set of data- and resource-driven features. We reproduce and adapt this approach to create a German lexicon of verbal polarity shifters. Thereby, we confirm that the approach works for multiple languages. We further improve classification by leveraging cross-lingual information from the English shifter lexicon. Using this improved approach, we bootstrap a large number of German verbal polarity shifters, reducing the annotation effort drastically. The resulting German lexicon of verbal polarity shifters is made publicly available.

Publication Details

Paper ID:
5306
Year:
2018
Publication:
TIPSTER
Publication Type:
conf
DOI:
nan