Saurabh Mathur

I’m a 2nd year computer science masters student at Indiana University, advised by Prof. David Crandall.

I did my bachelors at Vellore Institute of Technology, where I was advised by Prof. Daphne Lopez. My capstone project was on deep models for image caption generation. I’ve interned at Microsoft, Bangalore and Synopsys, Mountain View.

Research

My areas of interest are Artificial Intelligence and Machine Learning. Specifically, a lot of my work is on applying machine learning techniques in an interpretable way. Some of my Notes and Presentations can be found on GitHub.

Melanoma Detection using Capsule Networks
Melanoma Detection using Capsule Networks

Saurabh Mathur, Sumangali K.
ICNTET, 18 (To be published)
Slides

Automating early detection of one of the deadliest forms of cancer.

A scaled-down neural conversational model for chatbots
A scaled-down neural conversational model for chatbots

Mathur S, Lopez D.
Concurrency and Computation: Practice and Experience, 2018
Slides · Code

A neural conversational agent that requires fewer training examples.

Classification of text documents using association rule mining with critical relative support based pruning
Classification of text documents using association rule mining with critical relative support based pruning

P Karthik, M Saurabh, U Chandrasekhar
ICACCI, 2016
Slides

An interpretable text classifier.

Decision Making Using Fuzzy Soft Set Inference System
Decision Making Using Fuzzy Soft Set Inference System

U Chandrasekhar, S Mathur
ISBCC, 2016
Slides

A generalized framework for decision making systems to parameterize uncertainty.

Projects

I like to develop tools that cut repetitive work. The object of my projects ranges from Detecting Click-Bait to developing Portals for the University's Annual Festivals.

Automated Image Captioning System
Automated Image Captioning System

GitHub, 2018

A deep learning based image caption generator to improve accessibility on the web.

VITacademics
VITacademics

GitHub, 2018
Android · iOS

A web-server for mobile applications that streamline access to academic information for students at VIT.

Clickbait Detector
Clickbait Detector

GitHub, 2018
Chrome Extension

Detecting clickbait headlines in the wild with 90% accuracy.

Movie Recommendation Systems
Movie Recommendation Systems

GitHub, 2016
Blog Post

A comparitive study of collaborative filtering algorithms.

Happy and you know it
Happy and you know it

GitHub, 2016
Demo

Facial Emotion Recognition using Microsoft's Deep Residual Network.

Handwritten Digit Classification
Handwritten Digit Classification

GitHub, 2016
Report · Demo

A comparitive study of handwrittern digit recognition algorithms.

graVITas 2016
graVITas 2016

GitHub, 2016

Web-portals for VIT University’s annual technical festival.

References