An analysis of machine learning models for sentiment analysis of Tamil code-mixed data

Kogilavani Shanmugavadivel, Sai Haritha Sampath, Pramod Nandhakumar, Prasath Mahalingam, Malliga Subramanian, Prasanna Kumar Kumaresan, Ruba Priyadharshini

Abstract: Highlights•To perform sentiment analysis on Tamil code-mixed data learning methods are employed•To improve the accuracy of the models various pre-possessing techniques are explored•To avoid class imbalance problem in the dataset class re-sampling is performed•To capture local and global features hybrid deep learning models are implemented.