Research & Development Project, IIT Bombay

Natural Language Generation from RDF Triples

Mentor: Prof. Pushpak Bhattachryya | Course: CS 691 | Session: Spring, 2023
Related to M. Tech, CSE, IIT Bombay

Developed a state-of-the-art model to generate textual sentences from multiple RDF triples covering all the information. A pipeline approach with T5 and few-shot prompting generates the output.

Seminar Thesis, IIT Bombay

Text Generation

Mentor: Prof. Pushpak Bhattachryya | Course: CS 694 | Session: Spring, 2023
Related to M. Tech, CSE, IIT Bombay

Studied various aspects and objectives of text generation with special emphasis on movie script generation. Analysed and compared the state-of-the-art models in this field and their applications.

Course Projects, IIT Bombay

Argument Generation with Prompting

Mentor: Prof. Pushpak Bhattachryya | Course: CS 772 | Session: Spring, 2023
Related to M. Tech, CSE, IIT Bombay

Performed argument generation on CMV dataset using multi-task prompting on T5 model. The model was trained to optimize all the losses of different objectives.

Picture-based Word Sense Disambiguation

Mentor: Prof. Pushpak Bhattachryya | Course: CS 772 | Session: Spring, 2023
Related to M. Tech, CSE, IIT Bombay

Implemented a model to identify correct meaning of an ambiguous word through picture input. Created a novel sample dataset to test the results.

POS Tagging Using Encoder-Decoder and FFNN

Mentor: Prof. Pushpak Bhattachryya | Course: CS 772 | Session: Spring, 2023
Related to M. Tech, CSE, IIT Bombay

Developed two seperate models for POS Tagging on the universal tagset using Encoder-Decoder architecture and Feed-forward Neural Network.

Dynamics of Challenges Faced in Building a Startup

Mentor: Prof. Om Damani | Course: CS 752 | Session: Spring, 2023
Related to M. Tech, CSE, IIT Bombay

Analysed various social, economic and dynamic challenges faced by a startup system. Modeled them and simulated to find right situation for a startup growth.

Mental Health Prediction Using Machine Learning Models

Mentor: Prof. Preethi Jyothi | Course: CS 725 | Session: Autumn, 2022
Related to M. Tech, CSE, IIT Bombay

A dataset of personal life-based answers of tech people is used to predict mental health. Various models are used and the best of them are mentioned.

Ghar Dhundo

Mentor: Prof. Bhaskaran Raman | Course: CS 699 | Session: Autumn, 2022
Related to M. Tech, CSE, IIT Bombay

Building a website to find various homes or apartments for buying, renting or leasing based on different demographics, features and facilities.

Multi-threaded Web Server

Mentor: Prof. Mythili Vutukuru | Course: CS 744 | Session: Autumn, 2022
Related to M. Tech, CSE, IIT Bombay

Built a webserver with multiple threads, capable of handling many clients at the same time over TCP sockets. HTTP requests are served successfully and also the performance bottlenecks are optimized.

Linux Shell Functions Implementation

Mentor: Prof. Mythili Vutukuru | Course: CS 744 | Session: Autumn, 2022
Related to M. Tech, CSE, IIT Bombay

Implemented functional properties of a linux shell. Background process, foreground process, forking, reaping, process memory management and signal handling have been implemented along with efficient testing.

Other Projects

Hospital Management System

Mentor: Prof. Preetam Sur | B. Tech Project
Related to B. Tech, CSE, Govt. College of Engineering and Textile Technology

Our Proposed System will not only automate these processes with better flexibility and accuracy, but also provide a new way to track activities. We will integrate barcode application in our System. These barcode labels will help in keeping Patient data safe, Inventory control like easy accessibility of beds, tracking patient details in a faster, reliable and easier to use way. Also in the current pandemic situation, wristband efficient barcode labels will provide a way to track patients by the guardian following all the social distancing norms. Apart from these utilities we will be making the system user friendly with rich user interfaces and model the system in such a way that it becomes expandable. In addition to all the above advantages, there comes some issues like extra cost for cloud storage, Technology Integration Cost which may be a little disadvantageous. But with a little bit of one-time expense and periodical maintenance a lot of current problems can be solved using this software.

𝒒𝑹 Characterization and Polar Plot

Mentor: Late Dr. Ashesh Nandy, Prof. Subhas C Basak, Dr. Smarajit Manna
Related to Junior Researcher, C.I.R.E.

Our Recent Accomplishments include proposing peptide vaccines for various viral organisms along with characterising nucleotide or protein sequences with mathematical descriptors and graphical representations. This 2D Polar plot is one of the fruits of such theoritical approaches. In this method we assign certain angles to each amino acid upon their various biochemical properties and while plotting the graph we move through a vector by unit magnitude to the assigned direction. In this way we get a graph and calculating the centre of mass of this graph we get the qR value. Our experiments and analysis indicates this qR value to be characteristic property for a sequence.

Stock Market Prediction Model

A simple model using LSTM to fit a curve in stock prices of previous recordings.

Random Walk Model

Random walk is a popular model of statistics which represents the possible path distance with randomisation with a given time quanta.