Publication

Abstract : This paper presents the work done for the shared task on Me- OffendEs@IberLEF 2021 Non-contextual binary classification for Mex- ican Spanish. We implemented two deep neural network architectures such as a network containing a Bi-LSTM, LSTM, fully connected layer and another with a Bi-LSTM and LSTM stack. In addition to that we also implemented a BERT classifier. Among the three models the BERT exhibited better training performance, and we submitted the predictions based on the same. BERT performed well compared to other languages as it has pretrained embeddings that are trained on huge corpus of multiple languages.