Human-Centered Data Discovery & AI

Nina Klimenkova

PhD Student in Data Science

Worcester Polytechnic Institute

I am a PhD student in Data Science at Worcester Polytechnic Institute, advised by Dr. Roee Shraga. My research spans human-centered and algorithmic approaches to data discovery and integration, with a focus on how people, models, and systems judge table unionability. I study metacognitive signals—confidence, decision time, calibration—as a lens for improving evaluation, retrieval, and decision-making across data discovery tasks, including graded and uncertainty-aware evaluation frameworks, cost-aware system design, and emerging agentic approaches to exploring data lakes.

Professional headshot of the site owner.

Human-Centered Data Discovery and Integration

The role of human judgment in finding, understanding, and integrating heterogeneous data.

Table Unionability and Union Search

How people and computational systems determine whether tables can be meaningfully combined.

Human-in-the-Loop and Collaborative Intelligence

Combining human judgment, classical machine learning, and large language models rather than treating any one source as automatically correct.

Human-Centered Exploration of Table Unionability

Proceedings of the VLDB Endowment, vol. 19, no. 9, pp. 2099–2112.

Ten questions concerning Large Language Models (LLMs) for building applications

Building and Environment, vol. 291, article 114260.

Humans, Machine Learning, and Language Models in Union: A Cognitive Study on Table Unionability

HILDA ’25: Workshop on Human-In-the-Loop Data Analytics, pp. 1–7.

At a glance

Academic profile

My work connects data discovery with human judgment and computational support, with an emphasis on reliable collaboration between people and AI.

Program
Data Science Program, Worcester Polytechnic Institute
Adviser
Dr. Roee Shraga
Principal themes
Table unionability, human-in-the-loop systems, and metacognitive signals