Experience

  • May 2021- July 2021 Software engineer intern at MathWorks
  • May 2019-Present Undergraduate researcher at Precog
  • October 2019- November 2020 Backend Python developer at Wellowise Precog
  • May 2019- January 2020 Opensource ML contributor at OpenWorm Foundation Precog

Notable Work

  • Characterization of Koo
    Analyzed 4.5 million users and their follower-followee network to characterize the emerging multilingual Indian social media Koo. Featured in :Precog Precog Precog Precog

  • Unmasked COVID-19
    Analyzed the non-pharmaceutical interventions including mask wearing and social distancing in the six different cities in the USA and during the BLM protests using publicly available 2.04 million Instagram posts. Looked at difference in adherence of these NPI’s between Democrats and Republicans. Featured inPrecog

  • Hashtags are not judgemental
    Performed a large-scale empirical analysis of elections using only the hashtags shared on Twitter during the 2019 Lok Sabha elections in India. We study the trends and events unfolded on the ground, the latent topics to uncover representative hashtags and semantic similarity to discover sentiments during elections. Work accepted an published at Precog

Research interests

Social Computing, Deep Learning, Machine Learning.

Tools

Python, Elasticsearch, Kibana, Flask, Pandas, Pytorch, sklearn.

Notable Coursework

  • Deep Learning
  • Natural Language Processing
  • Reinfocement Learning
  • Privacy and Security in Online Soical Media
  • Graph Theory
  • Econometrics
  • Theory of Computation

Education

  • 2018-Present Btech CSE at IIITD , 9.53

Blogs

Is a Democrat more likely to wear a mask than a Republican??

Unmasking the partisan combat, an analysis through Instagram

Introduction The COVID-19 pandemic has drastically changed life in the world including, United States, with the country recording over 32.3 million cases and 574,000 deaths as of April, 2020. With the ongoing widespread vaccine rollout, the adoption of non-pharmaceutical interventions like mask-wearing mandates and stay-at-home laws have proved to be... [Read More]
Tags: Analysis, pandas, Deep Learning, Instagram, OSN, Public Health, Covid-19

bRead

Changing the way you consume content

What is the problem? A Filter Bubble is a state of intellectual isolation that can result from personalized searches when a social media algorithm selectively guesses what information a user would like to see based on their interaction history and other data which platform knows about them. It is difficult... [Read More]
Tags: Social Media, Privacy, Flask, React, python, Twitter

Knowledge Discovery using NLP

Retrieving hidden inferences using linguistic regularities.

Introduction Finding inferences from multiple resources at once has always been a difficult task. Majority of scientific knowledge is published as text, in the form of research papers. This makes it challenging to analyse it by traditional machine learning methods. In this project, we showed that biological research abstracts could... [Read More]
Tags: NLP, Linguistic regularites, Sklearn, pubmed, Covid-19

Instagram Cognition

Social Media Analysis for tracking location wise masks and collection of people.

Introduction The adoption of non-pharmaceutical interventions and their surveillance is critical for detecting and stopping possible transmission routes of COVID-19. A study of the effects of these inter-ventions in terms of adoption can help shape public health decisions. Social media analytics can offer crucial public health insights. The methodology used... [Read More]
Tags: Analysis, kibbana, elastic, Instagram, OSN, Public Health, Covid-19

Elections 2019

Through the lenses of twitter..

Hashtags in online social media have become a way for users to build communities around topics, promote opinions, and categorize messages. In the political context, hashtags on Twitter are used by users to campaign for their parties, spread news, or to get followers and get a general idea by following... [Read More]
Tags: Paper

GETFLIX

Not-Netflix :p ... A movie recommendation system

GETFLIX This is a :movie_camera: movie recommendation system based on collaborative learning (user-user and item-item) which was applied on the IMDB database.The recommendation system asks for ratings on 4 or more movies out of approx. 500 movies and returns 4 recommendations each for User-User and Item-Item collaborative learning algorithms. [Read More]
Tags: Recommendation system, Python, Mongo, Flask