Paper Abstract Access BibTex

VALUE: Visual Analytics driven Linked data Utility Evaluation

This paper has been published in proceedings of the Workshop on Human-In-the-Loop Data Analytics (HILDA 2023). The authors of this paper are:

  • Kaustav Bhattacharjee       PhD Candidate, Department of Data Science, New Jersey Institute of Technology, USA
  • Aritra Dasgupta       Assistant Professor, Department of Data Science, New Jersey Institute of Technology, USA

Abstract

The widespread adoption of open datasets across various domains has emphasized the significance of joining and computing their utility. However, the interplay between computation and human interaction is vital for informed decision-making. To address this issue, we first propose a utility metric to calibrate the usefulness of open datasets when joined with other such datasets. Further, we distill this utility metric through a visual analytic framework called VALUE, which empowers the researchers to identify joinable datasets, prioritize them based on their utility, and inspect the joined dataset. This transparent evaluation of the utility of the joined datasets is implemented through a human-in-the-loop approach where the researchers can adapt and refine the selection criteria according to their mental model of utility. Finally, we demonstrate the effectiveness of our approach through a usage scenario using real-world open datasets.


Fig: Inspecting utility of joining real world open datasets through the VALUE interface.

Online Access

BibTeX Reference

        @inproceedings{bhattacharjee2023value,
          title={VALUE: Visual Analytics driven Linked data Utility Evaluation},
          author={Bhattacharjee, Kaustav and Dasgupta, Aritra},
          booktitle={Proceedings of the Workshop on Human-In-the-Loop Data Analytics},
          pages={1--7},
          year={2023}
        }