Aditya Srivastava, V. Harsha Vardhan
This repository holds the code that accompanies our system description paper of the same name as the title.
The preprint can be viewed on arXiv.
The task description and the accompanying data can be found here.
Problems involving code-mixed language are often plagued by a lack of resources and an absence of materials to perform sophisticated transfer learning with. In this paper we describe our submission to the Sentimix Hindi-English task involving sentiment classification of code-mixed texts, and with an F1 score of 67.1% (which we further improve upon), we demonstrate that simple convolution and attention may well produce reasonable results.
The code has been written in Python 3.6.9. Other requirements have been listed in requirements.txt.