Contact Information

Anirban Sen

Department of Computer Science
Linkedin handle: Anirban@Linkedin
Phone: +91 11 2659 7300 (Office)
Office: Room 114, ICTD Lab, SIT Building


Current Work

  1. Analysis of Corporate-Political Nexus in India: Like many other economies, cases of nexus formation between various political entities and business houses is prevalent in India. Such nexus usually result in transaction of favours between political entities (like political parties, politicians, people related to politicians, etc.) and corporate entities (companies, top officers of companies, people related to the top officers, etc.) resulting in personal benefits of both parties. Quite often, these nexus indicate economic and political malpractices, which eventually lead to accumulation of wealth among few elites and misuse of political power, and hence corruption. Within such nexus, political entities possess a significant amount of influence on business decisions, and similarly, the big business houses directly or indirectly manipulate decision making for the state. My work concerns analysis of media data (from news archives, RSS feeds, wikis, blogs, and other sources) and developing a metric, which can appropriately red-flag potential cases of suspicious collusion among political and corporate entities in real time, merely by analysing the news snippets. The ultimate goal of this work is to build a search based browsing engine, primarily for target users belonging to journalism, which would list all news articles and media data that point towards suspicious corporat-political collusion between a corporate entity and a political entity.

  2. Analysis of Competition among Indian Industries: Existence of proper competition among industries is indispensable in any economy, to ensure distribution of wealth and effective utilization of resources. Many companies indulge in anti-competitive malpractices like cartelization, which make it extremely difficult for newer and smaller companies to enter the market. This also leads to accumulation of entire profit among hands of a few big companies, leading to deterioration of product/service quality. Such examples of anti-competitive market practices are exemplified in almost all the economies across the globe. My work involves detailed study of the Indian industrial sectors (and firms) to establish whether there exists such anti-competitive practices in the Indian economy. I am primarily focusing on identifying cases of vertical integration among Indian firms, which are often known to adversely impact the competitive environment among industry sectors. Currently, I am working on developing a Graph Kernel based algorithm to achieve this goal.

Previous Work

  1. Predicting Virality of Tweets in Online Social Networks: In Twitter, there are many topics (hashtags) which become quite viral with time, and are retweeted about thousands of times. There are some other topics, which die a natural death within days, hours, or even minutes. This work concerns identifying some structural features include some basic features, user network based features, conductance based features, and some geographical features. Finally, these features are used to build a classifier which, by just looking at the initial few tweets on the hashtag, could accurately predict if a topic will eventually become viral. We have used a Random Forest based classifier, and identified a set of structural features, which lead to a substantially high value of precision.

  2. Extracting Situational Updates from Microblogs during Calamities: Immediately after a man-made or natural disaster, Twitter gets flooded with tweets. Some of these tweets merely state emotions, or sentiments, while some other provide actionable information that can help in knowing the exact whereabout and status of the event. The latter are known as Situational Updates. My work in this project involved analysing tweets corresponding to four disasters namely Uttarakhand Flood (India), Hyderabad Blasts (India), Sandyhook School Shooting (US), and Typhoon Bopha (Phillipines), and extracting a set of tweet and user based features to build a classifier, which would effectively distinguish between situational updates and other non-situational tweets in real time. We have used support vector machine (SVM) for this purpose, and have come up with an interesting set of features, which works with an appreciable accuracy.

  3. Automated Diagram Generation from School Level Geometry Problems: In this project, I worked on creating an NLP based system, which would parse school level geometry problems (stated in English), and generate an intermediate graphical representation. From this representation, information can be extracted into a form which is compatible to any general graphical software (like OpenGL) so as to create an appropriate diagram in support of the problem statement. We tested our system on a set of around 500 geometry problems of varied complexity, and it showed and accuracy of around 70 percent.

Research Interests

  • Data Analytics and Information Retrieval
  • Development of Information and Communication Technologies towards Development (ICTD)

  • Publications

    1. Fine-grained Emotion Detection in Contact Center Chat Utterances
      Shreshtha Mundra (XRCI), Anirban Sen (IIT Delhi), Sandya Mannaswamy (XRCI), Manjira Sinha (XRCI), and Shourya Roy (XRCI)
      21st Pacific Asia Conference on Knowledge Discovery and Data Mining 2017 (PAKDD 2017). Jeju Island, South Korea

    2. Multi-task Representation Learning for Enhanced Emotion Categorization in Short Text
      Anirban Sen (IIT Delhi), Sandya Mannaswamy (XRCI), Manjira Sinha (XRCI), and Shourya Roy (XRCI)
      21st Pacific Asia Conference on Knowledge Discovery and Data Mining 2017 (PAKDD 2017). Jeju Island, South Korea

    3. Embedding Learning of Figurative Phrases for Emotion Classification in Micro-Blog Texts
      Shreshtha Mundra (XRCI), Manjira Sinha (XRCI), Sandya Mannaswamy (XRCI), Anirban Sen (IIT Delhi), and Shourya Roy (XRCI)
      The Fourth ACM IKDD Conferences on Data Sciences (CODS 2017)

    4. On the role of conductance, geography and topology in predicting hashtag virality
      Siddharth Bora, Harvineet Singh, Anirban Sen, Amitabha Bagchi, Parag Singla
      Social Network Analysis and Mining, December 2015

    5. Extracting Situational Awareness from Microblogs during Disaster Events
      Anirban Sen, Koustav Rudra, Saptarshi Ghosh
      Social Networking Workshop, COMSNETS 2015, 7th International Conference on Communication Systems & Networks

    6. Text-to-Diagram Conversion: A Method for Formal Representation of Natural Language Geometry Problems
      Anirban Mukherjee, Sarbartha Sengupta, Dipanjan Chakraborty, Anirban Sen, and Utpal Garain
      IASTED International Conference on Artificial Intelligence and Applications (AIA 2013). Innsbruck, Austria.

    Potential Patents

    • Work on Stance Detection in Healthcare Related Query Responses accepted as an Innovation Document in Xerox Research Centre India (potential patent filing)
    • Work on Fine-grained Emotion Detection in Contact Center Chat Utterances accepted as an Innovation Document in Xerox Research Centre India (potential patent filing)

    Work Experience

    • Selected for internship at Xerox Research Centre India in the Text and Graph Analytics group (June - December, 2016)
    • Teaching Assistant at IIT Delhi for courses Introduction to Computing, Database Management Systems, Computer Networks, and Data Structures
    • Worked at Tata Consultancy Services from December,2010 to July,2012 as Technical Editor.


    • B.Tech (Undergraduate) in Information Technology: RCC Institute of Information Technology, Kolkata (affiliated to West Bengal University of Technology)
    • M.E. (Postgraduate) in Computer Science: Indian Institute of Engineering Science and Technology, West Bengal
    • Ph.D. (Research) in Computer Science: IIT Delhi (currently pursuing)

    Some of the courses taken in IIEST (Postgraduation)

  • Principles of Programming Language
  • Programming Logic and Artificial Intelligence
  • Information Theory
  • Fuzzy Logic and Its Applications
  • Algorithms and Data Structures
  • Image Processing
  • Courses taken in IIT Delhi

  • Analysis of Media (paper reading course)
  • Advanced Data Structures
  • Advanced Data Management
  • Virtualization and Cloud
  • Machine Learning
  • Numerical Algorithms
  • Communication Skills

  • Me and My Other Interests

    I am a simple next door guy (researcher??) who loves to explore almost everything around. To brag about myself, I think my mentionable qualities are that I am a team player and I love making friends. I am also a teetotaler, a health conscious (read boring), and curious-about-stuff kind of guy who likes a lot of social networking. Other than my research, I am interested in reading (almost all kinds of books), programming, watching movies and travelling. Oh, and I am also a bit of a gym-freak, and in my spare time, I like to solve elementary math problems and puzzles. I am a moderator of a Facebook group named 'Elemath'. Last but not the least, I also like to collect information about various forms of Martial Arts (sad that I never properly had the opportunity of practising it myself).