I am a divergent thinker and an inquisitive machine learning researcher. My focus is knowledge-based systems that automate knowledge acquisition, reasoning and representation. I am also passionate about applications of computer intelligence and technology in education and social innovation.

My PhD research addresses unsupervised thematic phrase extraction from text. I am a member of the Cognition, Robotics and Learning (CoRaL) Lab at University of Maryland, Baltimore County. I have been actively involved in machine learning research in the reinforcement learning and cyber security domains.

In the past, I have developed various proof-of-concept automation systems, a continuous learning fraud detection method and a patented peer-to-peer lending risk assessment approach at PayPal Inc.

  • Knowledge Representation
  • Information Retrieval
  • Education Technology
  • Financial Technology
  1. Extending Signature-based Intrusion Detection Systems With Bayesian Abductive Reasoning LINK
    2018 | ACM Dynamic and Novel Advances in Machine Learning and Intelligent Cyber Security (DYNAMICS) Workshop | Ganesan A., Parameshwarappa P., Peshave A., Chen Z., Oates T.
  2. Thematic Hierarchies for Knowledge Discovery in Text
    Mar 2018 | US Semantic Technologies Symposium (Poster) | Peshave A., Oates T.
  3. Baltimore Housing Prices Disparity for Comparable Neighborhoods LINK
    Sep 2017 | Data for Policy | Peshave A., Memon S., Chavan V., Oates T.
  4. Learning Hierarchical Workflows Using Community Detection LINK
    May 2014 | Masters Thesis | Adviser - Dr. Tim Oates
  1. Trust Score Determination Using Peer-to-Peer Interactions LINK
    Nov 2018 | PayPal Inc. | Patent Granted
  2. Trust Score Investigation LINK
    Feb 2019 | PayPal Inc. | Patent Granted