A Strategy for Expert Recommendation From Open Data Available on the Lattes Platform

TL;DR

This paper presents a methodology for extracting and processing open data from the Lattes Platform to develop a deep learning-based recommendation system for experts, addressing the challenge of identifying specialists in large datasets.

Contribution

It introduces a novel data extraction and treatment process from the Lattes Platform and applies deep neural networks with autoencoders for expert recommendation.

Findings

Effective data extraction methodology from Lattes Platform

Deep neural network model achieves high recommendation accuracy

Autoencoder improves the quality of expert recommendations

Abstract

With the increasing volume of data and users of curriculum systems, the difficulty of finding specialists is increasing.This work proposes an open data extraction methodology of the Lattes Platform curricula, a treatment for this data and investigates a Recommendation Agent approach based on deep neural networks with autoencoder.