Shubhranshu Shekhar

Assistant Professor of Data Science at Brandeis International Business School.
Email: sshekhar [at] brandeis [dot] edu

prof_pic.jpeg

About

I am an Assistant Professor in the Brandeis International Business School at the Brandeis University.

I work, broadly, in Machine Learning. My research interests are in unsupervised and explainable methods for data-driven decision support in high-stakes domains like healthcare and finance. Nowadays, my focus is on empowering human decision-making by building intelligent systems that are unsupervised, explainable, and equitable.

Prior to Brandeis: I obatined my PhD in Machine Learning and Public Policy, and MS in Machine Learning Research from Carnegie Mellon University, where I worked with Prof. Leman Akoglu and Prof. Christos Faloutsos.

My CV is available here.

Publications

  1. PAKDD
    NETEFFECT: Discovery and Exploitation of Generalized Network Effects
    Meng-Chieh Lee, Shubhranshu Shekhar, Jaemin Yoo, and Christos Faloutsos
    In PAKDD 2024
    PDF Code
  2. PAKDD
    DIFFFIND: Discovering Differential Equations from Time Series
    Lalithsai Posam, Shubhranshu Shekhar, Meng-Chieh Lee, and Christos Faloutsos
    In PAKDD 2024
    PDF Code
  1. KDD
    Less is More: SlimG for Accurate, Robust, and Interpretable Graph Mining
    Jaemin Yoo*, Meng-Chieh Lee*, Shubhranshu Shekhar, and Christos Faloutsos
    In ACM SIGKDD Conference on Knowledge Discovery and Data Mining 2023
    PDF Code
  2. NBER
    Unsupervised Machine Learning for Explainable Health Care Fraud Detection
    Job market paper
    Accepted at ASHEcon 2023
    Shubhranshu Shekhar, Jetson Leder-Luis, and Leman Akoglu
    NBER Working Paper #30946 Feb 2023
    PDF
  3. JBI
    Benefit-aware Early Prediction of Health Outcomes on Multivariate EEG Time Series
    :trophy: George Duncan Award for PhD 2nd Paper at Heinz College
    Shubhranshu Shekhar, Dhivya Eswaran, Bryan Hooi, Jonathan Elmer, Christos Faloutsos, and Leman Akoglu
    Journal of Biomedical Informatics Feb 2023
    PDF Code
  4. ArXiv
    UltraProp: Principled and Explainable Propagation on Large Graphs
    Meng-Chieh Lee, Shubhranshu Shekhar, Jaemin Yoo, and Christos Faloutsos
    In arXiv preprint arXiv:2301.00270 2023
    PDF
    1. IEEE Big Data
      GEN2OUT: Detecting and Ranking Generalized Anomalies
      Meng-Chieh Lee*, Shubhranshu Shekhar*, Christos Faloutsos, T Noah Hutson, and Leon Iasemidis
      In IEEE International Conference on Big Data (Big Data) 2021
      PDF Code
    2. AAAI/ACM AIES
      FAIROD: Fairness-aware outlier detection
      Shubhranshu Shekhar, Neil Shah, and Leman Akoglu
      In Proceedings of the 2021 AAAI/ACM Conference on AI, Ethics, and Society 2021
      PDF Code
    1. TheWebConf
      Entity resolution in dynamic heterogeneous networks
      Shubhranshu Shekhar, Deepak Pai, and Sriram Ravindran
      In Workshop on Deep Learning for Graphs at The Web Conference (WWW) 2020
      PDF
      1. ECML-PKDD
        Incorporating privileged information to unsupervised anomaly detection
        :trophy: Best Student Machine Learning Paper Runner-up Award
        Shubhranshu Shekhar, and Leman Akoglu
        In Joint European Conference on Machine Learning and Knowledge Discovery in Databases 2018
        PDF Code
      1. MIKE
        Spreading Activation Way of Knowledge Integration
        Shubhranshu Shekhar, Sutanu Chakraborti, and Deepak Khemani
        In International Conference on Mining Intelligence and Knowledge Exploration 2015
        HTML
      1. ICCBR
        Linking cases up: An extension to the case retrieval network
        Shubhranshu Shekhar, Sutanu Chakraborti, and Deepak Khemani
        In International Conference on Case-Based Reasoning 2014
        HTML
      2. RecSys
        How popular are your tweets?
        Avijit Saha, Janarthanan Rajendran, Shubhranshu Shekhar, and Balaraman Ravindran
        In Proceedings of the 2014 Recommender Systems Challenge 2014
        HTML
      3. I-CARE
        Business Networking in Social Networks
        Dhanvin Mehta, Shubhranshu Shekhar, and Balaraman Ravindran
        In 6th IBM Collaborative Academia Research Exchange (I-CARE), non-archival 2014

      Patents

      1. Patent
        Utilizing a time-dependent graph convolutional neural network for fraudulent transaction identification
        Shubhranshu Shekhar, Deepak Pai, and Sriram Ravindran
        US Patent 11,403,643 2022
        HTML
      2. Patent
        Machine learning based on post-transaction data
        Moein Saleh, Xing Ji, and Shubhranshu Shekhar
        US Patent 11,321,632 2022
        HTML

      Teaching

      Instructor, Carnegie Mellon University
      Teaching Assistant, Carnegie Mellon University
      • Fall 2018
        10-701 Introduction to Machine Learning, Machine Learning Department
        • Most popular graduate ML course at the university
      • Spring 2019
        95-828 Machine Learning for Problem Solving, Heinz College
        • Most popular graduate ML course at Heinz College
      • Fall 2019
        95-796 Statistics for IT Managers & 90-777 Intermediate Statistics, Heinz College
        • Most popular graduate statistics course at Heinz College
      Teaching Assistant, Department of Computer Science and Engineering, IIT Madras
      • Fall 2013, 2014
        CS6370 Natural Language Processing (NLP)
        • Delivered a lecture on Machine Translation way of solving NLP problems
      • Spring 2014, 2015
        CS6250 Memory Based Reasoning (MBR) in AI
        • Gave a lecture on Search in Large Metric Space