A Review on Sentiment Analysis of Resource-Scarce Languages and Code-Mixed Social Media Text

With the phenomenal increase of social media usage today, organizations and governments have more aggressively turned to analysing user-generated content freely available on social media platforms in order to gain insights and sentiments on products, brands and policies. One challenge faced by social media analytics researchers in this respect is the habit of users frequently expressing their opinions using code-mixed text in their posts and comments. Code-mixing refers to the mixing of words from multiple languages in a single sentence. The languages used may include both formal and informal languages, with at least one of the languages being a resource-scarce language. This paper presents a review on methods that have been proposed and used for sentiment analysis of code-mixed social media text. Keywords: code-mixed, resource-scarce, sentiment analysis, social media