VietSentiLex: a sentiment dictionary that considers the polarity of ambiguous sentiment words

摘要

The ability to analyze sentiment is a major technology to analyze social media process. The sentiment analysis involves reading and understanding what is being said about a brand as well as advertising campaigns in online services to determine the nature of a product. Because the Vietnamese language has few resources for applying machine learning tasks, use of sentiment dictionaries is required. In this study, a sentiment dictionary called "VietSentiLex" is introduced for the aforementioned task in the Vietnamese language. Most notably, , In this study, instead of applying scores for every word, ambiguous words are considered more carefully as it is periodically positive or negative. Related words such as target nouns or verbs are used as contextual information for a sentiment word. Experiments to compare the performance of our dictionary to others are conducted. We prove that our dictionary has a high potential in predicting the polarity of reviews as compared to other dictionaries. In addition, various challenges and disadvantages of this dictionary are also outlined for future improvement until VietSentiLex can be a commercial product.