Scrape online for Sanskrit Corpus

Assignments and final project of NLP course at La Sapienza

Jointly Learning knowledge graph Embedding, Fine Grain Entity Types and Language Modeling.

Designed a mini-search engine using English language Wikipedia corpus of size 80 GB by creating inverted indices. It gives you top search result related to given query words.

Disaster information extraction using named entity recognition, we have incorporated a machine learning model which predicts entites in a given iput.

This is the benchmark repo for the LLM-IE Python package.

Information Extraction from books, using Natural Language Processing (NLP)

Applying BERT and LSTM on TACRED dataset for relation extraction task.

A Streamlit app to perform relation extraction using OpenIE.

Expert information extraction

Reproducing baseline model from ACL-2022 paper X-GEAR for Zero-shot Cross-Lingual EAE

A Python package for working with the outputs of Information Extraction models and tools such as SPERT and QuickGraph.

A Python application for educational penetration testing and cybersecurity learning.

Attention-based approach to NIL Entity Linking

Extract entities from a research paper.

LLM end to end

Linked Data Knowledge Base Population (KBP) framework built on top of Snorkel. The default configuration uses Wikipedia as text corpus and DBpedia as target.

Extract and aggregate tables of empirical results from computer science papers

Parse Institutiional Investment data from 13F-HR filings into a json file with time stamped data

Generate a score for insider trading activity for securities