big-data

Query Biased Multi-Document Abstractive Summarisation

Implemented multiple modifications and techniques for improving the state-of-the-art TPMS system. Proposed and implemented an alternating optimization approach for completing the matrix of relevance scores between paper vectors and author vectors. Still in progress.

Cross Modal Media Retrieval

Aimed at utilising the emotional information present in images and audios to perform cross modal media retrieval. We propsed a hypothesis, and two methods, implemented them, and performed experiments to verify our claims.

Language Independent Text-to-Emotion Classification

The project aimed at developing a machine learning model to assign an emotion to any chat message provided as input, in any of the top five languages common in Indian chat scenarios, although in the Latin Script.

Machine Learning for large scale logistics platform

Implemented the Document Distance problem using the Word Mover's Distance, and the reverse k-NN problem, using the SLICE algorithm.