MS Computer Science, 2020
![]()
Adithya Bandi worked on the Knowledge Representation of Unstructured Data (KRUD) project, looking at automated relation extraction from sparsely labeled datasets. He also interned at GE Research in summer 2019.
Adithya is currently working for the Investment Banking industry. Prior to UMBC, Adithya worked as Software Engineer in Oracle India and Amagi Media Labs for about three and half years, where he developed highly scalable products deployed on Cloud.
Publications
- Adithya Bandi, Karuna P. Joshi and Vaish Mulwad, “Affinity Propagation Initialisation Based Proximity Clustering For Labeling in Natural Language Based Big Data Systems“, In Proceedings of 6th IEEE International Conference on Big Data Security on Cloud (BigDataSecurity 2020), May 2020.
Adithya successfully defended his Master’s thesis in June 2020.
MS Thesis title: Affinity Propagation Initialization Based Proximity Clustering for Labeling
Committee: Dr. Karuna P Joshi (Chair), Dr. Varish Mulwad (co-chair, GE Research), Dr. Tim Finin, Dr. Frank Ferraro
Abstract:
Modern state of art relation extraction systems require large amounts of labeled data. However, obtaining such huge amounts of labeled data is a costly task and is almost impossible in industrial settings due to the time constraints of subject matter experts. Techniques like distant supervision have been used to provide noisy annotations but this requires the availability of a related knowledge base which is rarely possible. We propose a novel method where we obtain labeled data based on techniques inspired from Active Learning and Clustering. Our approach differs from Active Learning as we operate under weak supervision, where all the instances provided for training are not manually labeled. Secondly, This differs from any prevailing clustering algorithms as we adopt a whole new approach. Due to the extrapolation of the labeling efforts, it will make it easier to adopt deep learning approaches with minimal manual effort.