A New Sentiment Analysis Model for Mixed Language using Contextual Lexicon

Sentiment analysis is one of the most active research areas in Natural Language Processing since early 2000. To date, sentiment analysis has been applied to various domains such as product, movie, sport and political reviews. Previously, the research in sentiment analysis area only concentrated on mining a single language. Nevertheless, due to the growth of multiple language usage in the form of writing and speaking, sentiment analysis activities have become more challenging. Many opinion keywords carry different polarities when they are used in different context, posing huge challenges to this field of research. Furthermore, in a social media environment where users tend to mix up languages, a lot of ambiguous content is present which makes a post difficult to be classified. Thus, this paper is a foray into the sentiment analysis for the context of mixed language.