@inproceedings{balouchzahi-etal-2021-mucs-lt,
    title = "{MUCS}@{LT}-{EDI}-{EACL}2021:{C}o{H}ope-Hope Speech Detection for Equality, Diversity, and Inclusion in Code-Mixed Texts",
    author = "Balouchzahi, Fazlourrahman  and
      B K, Aparna  and
      Shashirekha, H L",
    editor = "Chakravarthi, Bharathi Raja  and
      McCrae, John P.  and
      Zarrouk, Manel  and
      Bali, Kalika  and
      Buitelaar, Paul",
    booktitle = "Proceedings of the First Workshop on Language Technology for Equality, Diversity and Inclusion",
    month = apr,
    year = "2021",
    address = "Kyiv",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/add-emnlp-2024-awards/2021.ltedi-1.27/",
    pages = "180--187",
    abstract = "This paper describes the models submitted by the team MUCS for {\textquotedblleft}Hope Speech Detection for Equality, Diversity, and Inclusion-EACL 2021{\textquotedblright} shared task that aims at classifying a comment / post in English and code-mixed texts in two language pairs, namely, Tamil-English (Ta-En) and Malayalam-English (Ma-En) into one of the three predefined categories, namely, {\textquotedblleft}Hope{\_}speech{\textquotedblright}, {\textquotedblleft}Non{\_}hope{\_}speech{\textquotedblright}, and {\textquotedblleft}other{\_}languages{\textquotedblright}. Three models namely, CoHope-ML, CoHope-NN, and CoHope-TL based on Ensemble of classifiers, Keras Neural Network (NN) and BiLSTM with Conv1d model respectively are proposed for the shared task. CoHope-ML, CoHope-NN models are trained on a feature set comprised of char sequences extracted from sentences combined with words for Ma-En and Ta-En code-mixed texts and a combination of word and char ngrams along with syntactic word ngrams for English text. CoHope-TL model consists of three major parts: training tokenizer, BERT Language Model (LM) training and then using pre-trained BERT LM as weights in BiLSTM-Conv1d model. Out of three proposed models, CoHope-ML model (best among our models) obtained 1st, 2nd, and 3rd ranks with weighted F1-scores of 0.85, 0.92, and 0.59 for Ma-En, English and Ta-En texts respectively."
}