VALUE: Visual Analytics driven Linked data Utility Evaluation

Document Type

Conference Proceeding

Publication Date

6-18-2023

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.

Identifier

85167947255 (Scopus)

ISBN

[9798400702167]

Publication Title

Hilda 2023 Workshop on Human in the Loop Data Analytics Co Located with SIGMOD 2023

External Full Text Location

https://doi.org/10.1145/3597465.3605225

This document is currently not available here.

Share

COinS