Intent-Classification-and-Sequence-Labelling

Intent classification is the automatic categorization of text data based on customer goals. It is known to be a complex problem in NLP. Sequence Labelling aims to classify each token (word) in a class space C. This project addresses these two problem statements by covering the basic concepts of NLP to advanced ones. For instance, linguistics analysis of a certain corpus, static word embeddings, contextual word embeddings. machine learning, deep learning, transformers and BIO tagging. This project was developed as part of the major project in our NLP coursework for the Data Science Master's degree. All of the work starting from problem statement formulation to project proposal, data collection, preparation. analysis, modelling, feature engineering, presentation, research paper creation were done by two members in the group: myself and Abhisek Panigrahi (https://github.com/Abhisekgit1994).