Recovering renewable energy experts through faceted taxonomy and natural language processing techniques

an academic data mining experiment applied by researchers from Bahia state universities

Abstract

Objective: This article proposes a solution for retrieving textual information from an academic database, using natural language processing techniques to identify renewable energy experts. The solution employs a faceted taxonomy and a competency mapping platform.
Methodology: The research follows an experimental approach, structured in the following steps: 1) Problem identification and objective definition; 2) Systematic search and review of articles on renewable energy to form the control vocabulary; 3) Construction of the renewable energy taxonomy using the 101 method; 4) Implementation of the search engine; 5) Analysis of the expert researchers' data. The data were cataloged on the simcc.uesc.br platform, including information such as number of publications, Lattes abstracts, relevance indices, and researchers' institutions.
Results: The development of a search engine and an analytical solution allowed correlating researchers with the renewable energy taxonomy. Applying the faceted taxonomy as a filter resulted in 550 database requests.
Conclusions: The use of faceted taxonomy and the development of the search engine provided a recovery of experts in renewable energy, demonstrating the effectiveness of the proposed approach in the automatic combination of terms to improve the search and analysis of academic information.