IIT Gandhinagar at SemEval-2020 Task 9: Code-Mixed Sentiment Classification Using Candidate Sentence Generation and Selection

Code-mixing is the phenomenon of using multiple languages in the same\nutterance of a text or speech. It is a frequently used pattern of communication\non various platforms such as social media sites, online gaming, product\nreviews, etc. Sentiment analysis of the monolingual text is a well-studied\ntask. Code-mixing adds to the challenge of analyzing the sentiment of the text\ndue to the non-standard writing style. We present a candidate sentence\ngeneration and selection based approach on top of the Bi-LSTM based neural\nclassifier to classify the Hinglish code-mixed text into one of the three\nsentiment classes positive, negative, or neutral. The proposed approach shows\nan improvement in the system performance as compared to the Bi-LSTM based\nneural classifier. The results present an opportunity to understand various\nother nuances of code-mixing in the textual data, such as humor-detection,\nintent classification, etc.\n