ml

Machine Learning Research, CNeRG

Department of Computer Science and Engineering
Indian Institute of Technology Kharagpur

Students

• Avirup Saha

  • Supervisor: Niloy Ganguly

  • Projects:

    • Learning hashtag/e-commerce dynamics: Here we investigate hashtag/product popularity using temporal point processes

    • Leaning network traffic dynamics: Here we investigate network traffic flow using temporal point processes

    • Fake news detection: Here we develop a system for automatic fact-checking to identify fake news

  • Publications:

    • Saha, Avirup, Bidisha Samanta, Niloy Ganguly, and Abir De. “CRPP: Competing Recurrent Point Process for Modeling Visibility Dynamics in Information Diffusion.” In Proceedings of the 27th ACM International Conference on Information and Knowledge Management, pp. 537-546. ACM, 2018.
    • Saha, Avirup, Niloy Ganguly, Sandip Chakraborty, and Abir De. “Learning Network Traffic Dynamics Using Temporal Point Process.” In IEEE INFOCOM 2019-IEEE Conference on Computer Communications, pp. 1927-1935. IEEE, 2019.
    • Santosh, T. Y. S. S., Srijan Bansal, and Avirup Saha. “Can Siamese Networks help in stance detection?.” Proceedings of the ACM India Joint International Conference on Data Science and Management of Data. ACM, 2019.
    • Tokala, Santosh, G. Vishal, Avirup Saha, and Niloy Ganguly. “AttentiveChecker: A Bi-Directional Attention Flow Mechanism for Fact Verication.” In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pp. 2218-2222. 2019.

• Aakash Verma

  • Supervisor: Niloy Ganguly

  • Projects:

    • Data-driven approach for optimizing ML inference time- : In this project, we are trying to build some hashing mechanism to skip some layers in the computational graph for some clustered data-points to reduce the inference time.

  • Publications:

    • None

• Soumyadeep Roy

  • Supervisor: Prof. Niloy Ganguly, Prof. Shamik Sural

  • Projects:

    • Computational Approaches for Brand Positioning - A collaraboration with Adobe Big Data Experience Lab, Bangalore, India

    • Development of a remote healthcare delivery system: Early diagnosis, therapy, follow-up and preventive care for non-communicable diseases (cardio-pulmonary) - IMPRINT-1 Project (Code: RCO)

  • Publications:

    • Soumyadeep Roy, Koustav Rudra, Nikhil Agrawal, Shamik Sural, Niloy Ganguly, Towards an Aspect-based Ranking Model for Clinical Trial Search, in the 8th International Conference on Computational Data and Social Networks (CSoNet 2019), Novemer 18 – 20, 2019, Ho Chi Minh City, Vietnam (Conference Full paper) Link : https://github.com/nikhil741/COCTR_multidimensional_ranking/

    • Soumyadeep Roy, Niloy Ganguly, Shamik Sural, Niyati Chhaya, and Anandhavelu Natarajan, Understanding Brand Consistency from Web Content, in Proceedings of the 10th ACM Conference on Web Science, WebSci 19, (Boston, MA, USA), pp. 245–253, ACM, 2019. (Conference full paper) Link : https://zenodo.org/record/3565079

• Abhisek Dash

  • Supervisor: Saptarshi Ghosh, Animesh Mukherjee

  • Projects:

    • A Network-centric framework for Auditing Recommendation System– In this project we proposed a network centric framework and few methodologies for third party auditing of recommendation systems prevailing on different online movie recommendation systems.

    • Fairness in text summarization– In this project we proposed methodologies to generate demographically fair summaries of social media microblog posts (tweets) generated by real users during different events.

    • Auditing Amazon Marketplace– Currently, we are working on a project where we are investigating different information filtering algorithms e.g., recommendation systems, search systems etc. on Amazon website and its repercussions on exposure of different sellers and producers on Amazon.

  • Publications:

    • A Network-centric Framework for Auditing Recommendation Systems– IEEE INFOCOM 2019

    • Summarizing User-generated Textual Content: Motivation and Methods for Fairness in Algorithmic Summaries– PACM HCI (CSCW) 2019

• Bishal Santra

  • Supervisor: Pawan Goyal

  • Projects:

    • Dialogue Systems, Knowledge Graph Completion, Recommendation Systems

  • Publications:

    • Soumya Sharma, Bishal Santra, Abhik Jana, T Y S S Santosh, Niloy Ganguly and Pawan Goyal. Incorporating Domain Knowledge into Medical NLI using Knowledge Graphs. EMNLP-IJCNLP 2019, November 3-7, Hong Kong (short paper). https://github.com/soummyaah/KGMedNLI/

    • Amrith Krishna, Vishnu Dutt Sharma, Bishal Santra, Pavan Kumar Satuluri and Pawan Goyal (2019). Poetry to Prose Conversion in Sanskrit as a Linearisation Task: A case for Low-Resource Languages. ACL, July 28th - August 3rd, Florence, Italy (short paper). https://www.aclweb.org/anthology/P19-1111/

- Abhinav Kuma, Aishwarya Gupta, Bishal Santra, Lalitha Srinivasan, Manasa Kolla, Mayank Gupta and Rishabh Singh, VPDS: An AI-based Automated Vehicle Occupancy and Violation Detection System, IAAI (application conference of AAAI), 2019 https://www.aaai.org/ojs/index.php/AAAI/article/view/5005/

	- Amrith Krishna, Bishal Santra, Sasi Prasanth Bandaru, Sahu, Gaurav, Sharma, Vishnu Dutt, Satuluri, Pavan Kumar and Pawan Goyal (2018). Free as in Free Word Order: An Energy Based Model for Word Segmentation and Morphological Tagging in Sanskrit. EMNLP, Brussels, Belgium, October 31-November 4, 2018. https://github.com/bsantraigi/Sanskrit-Segmentation-Extended
	
	- Amrith Krishna, Bishal Santra, Pavan Kumar Satuluri, Sasi Prasanth Bandaru, Bhumi Faldu, Yajuvendra Singh and Pawan Goyal (2016). Word Segmentation in Sanskrit Using Path Constrained Random Walk. 26th International Conference on Computational Linguistics (Coling), Osaka, Japan, December 11-16, 2016 (Poster). https://github.com/bsantraigi/Sanskrit-Segmentation 

• Abhilash Nandy

  • Supervisor: Prof. Niloy Ganguly

  • Projects:

    • • Question Answering on E-Manuals (in collaboration with Samsung)

  • Publications:

    • Identification of Cervical Pathology using Adversarial Neural Networks.

    • An Adaptive Anaphylaxis Detection and Emergency Response System.

    • Kinship Verification using Deep Siamese Convolutional Neural Network. Dataset - https://web.northeastern.edu/smilelab/fiw/

    • A Densenet Based Robust Face Detection Framework. Dataset - https://www.tensorflow.org/datasets/catalog/wider_face

    • Understanding Community Rivalry on Social Media: A Case Study of Two Footballing Giants.

    • Low-cost Brain Controlled Orthotic Exoskeleton Arm for Monoplegic Paralyzed Individuals

• Punyajoy Saha

  • Supervisor: Animesh Mukherjee

  • Projects:

    • Fear speech detection and analysis, In this project we try to build a dataset for fear speech (fear against a target community) and model to detect and analyse them

    • Hate speech explanation dataset, In this project we try to build a dataset and model for explaining hate speech detection.

    • Temporal effects of Unmoderated Hate speech in Gab, In this project we try to analyse the temporal characteristics of hate in an unmoderated community

    • Deep Learning Models for Multilingual Hate Speech Detection, In this project, we build a benchmark for hate speech detection in 9 languages across 16 datasets

    • Thou shalt not hate: Countering Online Hate Speech, In this project, we tried to analyze counter speech on YouTube by curating a dataset and building several models

  • Publications:

    • Deep Learning Models for Multilingual Hate Speech Detection (accepted in ECML PKDD 2020) publication– https://arxiv.org/abs/2004.06465 code–https://github.com/punyajoy/DE-LIMIT
    • Thou shalt not hate: Countering Online Hate Speech (accepted at ICWSM 2019) publication– https://www.aaai.org/ojs/index.php/ICWSM/article/view/3237 dataset and code–https://github.com/binny-mathew/Countering_Hate_Speech_ICWSM2019

• Sayantan Adak

  • Supervisor: Animesh Mukherjee

  • Projects:

    • Gandhipedia

    • A one-stop AI-Enabled Portal for Browsing Gandhian Literature, Life-events and His Social Network

  • Publications:

    • JCDL2020 demo/poster: https://github.com/gandhipedia/

• Souvic Chakraborty

  • Supervisor: Animesh Mukherjee, Pawan Goyal

  • Projects:

    • Aspect based Sentiment Analysis of Scientific Reviews

    • We used publicly available dataset and with the help of annotations of aspect sentiments, we established the importance of each aspect like Clarity, the novelty of ideas on paper acceptance. We also studied the impact of disagreement between reviewers on Chair’s decision, how aspect based disagreements have strong correlations with overall disagreements, and how each aspect sentiment influenced the degree of disagreement.

  • Publications:

    • Aspect based Sentiment Analysis of Scientific Reviews, JCDL 2020

• SK Mainul Islam

  • Supervisor: Sourangshu Bhattacharya

  • Projects:

    • Aspect Based Sentiment Analysis

    • In this work we are trying to improve the sentiment prediction of the sentence wrt aspect by incorporating knowledge graph embedding.

  • Publications:

    • Under review at EMNLP 2020