Tensorflow and Keras implementation of the state of the art researches in Dialog System NLU

This is NLP project of text multi-classification. My pre-trained model classify speech into three different categories offensive, hateful and neither.

This project used in my dissertation for detecting sarcasm in twitter dataset using Sate-of-the-Art Transformer

Question Answering Chatbot with DistilRoBERTa Sentence Embeddings, Dialogflow and Ngrok

Using Task Specific Knowledge Distillation to obtain DistilRoBERTa model fine-tuned on SST-2 part of the GLUE dataset for sentiment analysis.

Data and models to extract toponyms and spatio-temporal entities from text data.

A sophisticated multilabel text classification model to seamlessly categorize diverse quote tags.