U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

VEMOS: A GUI for Evaluation of Similarity Metrics on Complex Data Sets

NCJ Number
303990
Author(s)
Eve Fleisig; Gunay Dogan
Date Published
June 2021
Length
34 pages
Annotation

This report presents the features and describes the functions of VEMOS (Visual Explorer for Metrics of Similarity), which is a general-use software tool that can assist researchers and practitioners in evaluating multiple comparison metrics on rich, diverse data sets for a variety of data analysis tasks.

Abstract

VEMOS enables its users to load a set of data files and matrices of similarity/dissimilarity scores from one or more metrics. Existing python packages for data analysis provide some capabilities that can be used for analyses of comparison metrics, but without the ability to easily examine individual items or groups in a data set or to do so alongside analyses of metrics’ performance on the whole data set. VEMOS, on the other hand, is built around an easy, user-friendly graphical interface (GUI) that seamlessly integrates both data set exploration and metric evaluation, as well as experimentation. The goal with VEMOS is to provide a GUI tool that is ready to use, so that both experienced researchers and novices can load the data and directly begin their analyses. This report describes VEMOS capabilities in relation to data types and relations; data organization, loading, and storage; data record browsing; and a visual metric analyzer. 14 figures and 16 references