I-Corps: Analytic Tool Discovery System for Interdisciplinary Data Analysis

University Of South Dakota Main Campus, Vermillion SD

Investigators

Abstract

The broader impact/commercial potential of this I-Corps project is to significantly improve end-users' data analysis productivity and result accuracy. The Resource Enhancement and Advanced Discovery System (READS) technology streamlines users approaches to locating the most relevant analytic tools for their specific problems through natural language capabilities. While focusing on geoscientists, financial analysts, and biotechnologists as potential users at the inception of this I-Corps project, READS can be customized for any discipline and its specific data analytic tool requirements. The three user examples listed above are only the beginning of the potential applications for which this analytic tool discovery system can be utilized. The ability to query resources with discipline-specific language and increase result accuracy by annotating the queried resources with domain-specific terminology will drastically reduce the amount of time required to locate requisite tools, and ultimately attract users. New users will not need to start from scratch in their analyses, but will be able to build upon the resources discovered by others, therefore maximizing productivity. This broad range of appeal, combined with an accessible user interface not seen in other tools, underscores the potential for significant commercial impact of this technology. This I-Corps project aims to assist users in easily locating the "best" analytic tools for their specific task through an analytic tool discovery system for interdisciplinary data analysis. Data manipulation challenges are emerging and the integration of dataset analytic tools into a comprehensive framework is needed to further knowledge discovery. Abundant tools and methods are available to the research and business communities to analyze datasets across disciplines, but it is often difficult to locate the most relevant tools for a specific task. Adding context to queries greatly enhances a users' ability to retrieve requisite analytic tools. READS integrates analytic tool information into a one-stop-shop retrieval system capable of accessing valuable metadata from multiple repositories. The system allows end-users to retrieve analytic tool information by submitting either keyword-based queries and/or free text-based questions. The system leverages natural language processing, text mining, and an ontology-based metadata annotation system to allow collections of analytic tools and methods to be linked and discovered with high accuracy. This allows end-users to discover, reuse, validate, share, and exchange knowledge related to their chosen data analytic tools. This novel approach facilitates the accessible analysis of cross-domain datasets, greatly enhancing end-users' ability to more efficiently answer challenging questions. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.