CoSaD-Code-Mixed Sentiments Analysis for Dravidian Languages

Analyzing sentiments or opinions in code-mixed languages is gaining importance due to increase in the use of social media and online platforms especially during the Covid-19 pandemic. In a multilingual society like India, code-mixing and script mixing is quite common as people especially the younger generation are quite familiar in using more than one language. In view of this, the current paper describes the models submitted by our team MUCIC for the shared task in’Sentiments Analysis (SA) for Dravidian Languages in Code-Mixed Text’. The objective of this shared task is to develop and evaluate models for code-mixed datasets in three Dravidian languages, namely: Kannada, Malayalam, and Tamil mixed with English language resulting in Kannada-English (Ka-En), Malayalam-English (Ma-En), and Tamil-English (Ta-En) language pairs. N-grams of char, char sequences, and syllables features are transformed into feature vectors and are used to train three Machine Learning (ML) classifiers with majority voting. The predictions on the Test set obtained average weighted F1-scores of 0.628, 0.726, and 0.619 securing 2nd, 4th, and 5th ranks for Ka-En, Ma-En, and Ta-En language pairs respectively. © 2021 Copyright for this paper by its authors.

Hope Speech Detection for Dravidian Languages Using Cross-Lingual Embeddings with Stacked Encoder Architecture.

The task of hope speech detection has gained traction in the natural language processing field owing to the need for an increase in positive reinforcement online during the COVID-19 pandemic. Hope speech detection focuses on identifying texts among social media comments that could invoke positive emotions in people. Students and working adults alike posit that they experience a lot of work-induced stress further proving that there exists a need for external inspiration which in this current scenario, is mostly found online. In this paper, we propose a multilingual model, with main emphasis on Dravidian languages, to automatically detect hope speech. We have employed a stacked encoder architecture which makes use of language agnostic cross-lingual word embeddings as the dataset consists of code-mixed YouTube comments. Additionally, we have carried out an empirical analysis and tested our architecture against various traditional, transformer, and transfer learning methods. Furthermore a k-fold paired t test was conducted which corroborates that our model outperforms the other approaches. Our methodology achieved an F1-score of 0.61 and 0.85 for Tamil and Malayalam, respectively. Our methodology is quite competitive to the state-of-the-art methods. The code for our work can be found in our GitHub repository (https://github.com/arunimasundar/Hope-Speech-LT-EDI).