Extending a sentiment lexicon with synonym–antonym datasets: SWNetTR++

Abstract

In our previous studies on developing a general-purpose Turkish sentiment lexicon, we constructed SWNetTR-PLUS, a sentiment lexicon of 37K words. In this paper, we show how to use Turkish synonym and antonym word pairs to extend SWNetTR-PLUS by almost 33% to obtain SWNetTR++, a Turkish sentiment lexicon of 49K words. The extension was done by transferring the problem into the graph domain, where nodes are words, and edges are synonymantonym relations between words, and propagating the existing tone and polarity scores to the newly added words using an algorithm we have developed. We tested the existing and new lexicons using a manually labeled Turkish news media corpus of 500 news texts. The results show that our method yielded a significantly more accurate lexicon than SWNetTR-PLUS, resulting in an accuracy increase from 72.2% to 80.4%. At this level, we have now maximized the accuracy rates of translation-based sentiment analysis approaches, which first translate a Turkish text to English and then do the analysis using English sentiment lexicons.

Investigation of the Perception toward Health Workers and Teachers on Twitter Data via Sentiment Analysis in the Covid-19 Pandemic

“…Sözlük tabanlı duygu analizi, 2019 yılında hazırlanan SWNetTR++ sözlüğü kullanılarak gerçekleştirilmiştir (Sağlam, Sever, & Genç, 2019, s. 1806. Sözlük, yaklaşık olarak 49 bin Türkçe kelime ve kelime gruplarına ait yönelimleri gösteren kutup (polarity) ve yoğunluk (tone) değerlerini içermektedir.…”

Turkish Sentiment Analysis for Open and Distance Education Systems

“…There is no such dictionary in Turkish, while there are comprehensive predefined polarity lexicons in English, Dutch, Spanish, and Spanish. Turkish language's distinctive characteristics, such as agglutinative, negation suffixes, make the Lexicon-based approach difficult; therefore, Lexicon-based SA studies in Turkish are based on translation-dependent lexicons or have relatively narrow focuses (Saglam et al, 2019). These studies are SentiTurkNet (Dehkharghani et al, 2016) by using Turkish WordNet (Bilgin et al, 2004) and SWNetTR (Ucan, 2014).…”