作者: Rahim Dehkharghani , Yucel Saygin , Berrin Yanikoglu , Kemal Oflazer
DOI: 10.1007/S10579-015-9307-6
关键词:
摘要: Sentiment analysis aims to extract the sentiment polarity of given segment text. Polarity resources that indicate words are commonly used in different approaches. While English is richest language regard having such resources, majority other languages, including Turkish, lack resources. In this work we present first comprehensive Turkish resource, SentiTurkNet, where three scores assigned each synset WordNet, indicating its positivity, negativity, and objectivity (neutrality) levels. Our method general applicable languages. Evaluation results for show obtained through more accurate compared those direct translation (mapping) from SentiWordNet.