SENTIMENT ANALYSIS FOR VIETNAMESE

Abstract

With the development of technology and the Internet, different types of social media such as social networks and forums have allowed people to not only share information but also to express their opinions and attitudes on products, services and other social issues. The Internet becomes a very valuable and important source. People nowadays use it as a reference to make their decisions on buying a product or using a service. Moreover, this kind of information also let the manufacturers and service providers receive feedback about limitations of their products. Therefore, that information may help them to improve their products in order to better meet the customer needs. Public opinions can also help authorities know the attitudes and opinions of their residents on social events so that they can make appropriate adjustments. The large amount of data from the Internet increases the demand to develop automatic system that can identify and classify sentiments in an opinion. Since early 2000s, opinion mining and sentiment analysis have become a new and active research topic in Natural language processing and Data mining. However, research in sentiment analysis for Vietnamese is relatively new and there is no systematic comparison between the performance2 of Vietnamese sentiment analysis systems. In this thesis, I will present some effective models for classifying Vietnamese sentences based on Support Vector Machine (SVM), an advanced Supervised Learning technique. The scope is to evaluate the ability of classifying Vietnamese reviews into one of three categories: “positive”, “negative”, or “neutral”. Results from different feature sets and classifiers are reported in term of accuracy, which shows the effectiveness of each method