- Abstract: User generated content contains opinionated texts not only in dominant languages (like English) but also less dominant languages( like Amharic). However, negation handling techniques that supports for sentiment detection is not developed in such less dominant language(i.e. Amharic). Negation handling is one of the challenging tasks for sentiment classification. Thus, this work builds negation handling schemes which enhances Amharic Sentiment classification. The proposed Negation Handling framework combines the lexicon based approach and character ngram based machine learning model. The performance of framework is evaluated using the annotated Amharic News Comments. The system is outperforming the best of all models and the baselines by an accuracy of 98.0. The result is compared with the baselines (without negation handling and word level ngram model).
- Keywords: Negation Handling Algorithm, Amharic Sentiment Analysis, Amharic Sentiment lexicon, char level, word level ngram, machine learning, hybrid
- TL;DR: This work presents Amharic Negation Handling for efficient Sentiment Classification.