Dictionary Based Amharic Sentiment Lexicon Generation

Abstract

Sentiment analysis is a hot research area with several applications including analysis of political opinions, classifying comments, movie reviews, news reviews and product reviews. To employ rule based sentiment analysis, sentiment lexicon is required. However, manual construction of a sentiment lexicon is time consuming and costly for resource-limited languages. To reduce development time and costs, we propose an algorithm for constructing Amharic sentiment lexicons. The proposed approach transfers sentiment labels from a one language (e.g. English) to resource-limited language (e.g. Amharic) relying on Amharic-English dictionary. Using Bilingual/Monolingual dictionaries as a bridge, two Amharic sentiment lexicons are automatically generated the first based on SO-CAL polarity lexicon, the second on SentiWordNet 3.0. For each Amharic word, the algorithm finds the meaning of the corresponding English word(s). For these English words, sentiment information is searched from the aforementioned sentiment lexicon(s). The weighted average of returned sentiment values, part of speech and gloss information is assigned to the Amharic word. Lexicons of 5683 and 13679 words, respectively, are generated automatically and evaluated subsequently.