Advanced search
Start date
Betweenand

A personalized Recommender system for e-Recruitment

Grant number: 16/08352-1
Support type:Research Grants - Innovative Research in Small Business - PIPE
Duration: April 01, 2018 - September 30, 2020
Field of knowledge:Interdisciplinary Subjects
Principal researcher:André Felipe Pires Sonnenburg
Grantee:André Felipe Pires Sonnenburg
Company:Reachr Soluções Inovadoras em RH Ltda
CNAE: Consultoria em tecnologia da informação
Tratamento de dados, provedores de serviços de aplicação e serviços de hospedagem na internet
City: São Paulo
Pesquisadores principais:
Leandro Nunes de Castro Silva
Associated scholarship(s):19/23195-8 - A personalized recommender system for e-Recruitment, BP.TT
18/26628-0 - A personalized recommender system for e-recruitment, BP.TT
18/16899-6 - A personalized recommender system for e-recruitment, BP.TT

Abstract

Online recruitment platforms, called e-recruitment or e-recruitment, have become the main recruitment channel for most companies and the main source of job search for most candidates. On the one hand, these platforms reduce the time to fill vacancies, but on the other, they have the challenge of providing efficient selection mechanisms, including bilateral recommendation systems, that is, vacancies for candidates and candidates for vacancies. Recommender systems have been developed to help users find items that are of interest to them in a virtual environment, for example, products in an e-commerce store, travel, services, among others. In the last decade, also motivated by the explosion in data generation and storage, studies have begun to emerge associated with the latent need to make e-recruitment platforms more efficient and assertive. In this sense, recommender systems have been the most researched and employed tool in the world, and REACHR is in search of this competitive differential in the Brazilian market. The objective of this research project is to develop a recommendation engine for the REACHR that is capable of identifying the main elements and characteristics of the vacancies and the candidates so that the best matching can be done between them. Success in this task will bring the competitive edge to the company and allow professionals to find positions within companies that are more aligned with their profile and expectations and, reciprocally, companies find professionals more aligned with the needs and even the corporate culture. This model will bring a more adequate selection, providing a reduction of cost and increase of quality in the hirings. To achieve these objectives, REACHR has established a partnership with the Natural Computing and Machine Learning Laboratory (LCoN) of Universidade Presbiteriana Mackenzie, which has, within its research team, professionals who were responsible for the design and development of one of the first e-commerce recommender systems in Brazil, as well as publications and dissertations presented on the subject. A research and development methodology will be used that is valid both from a scientific and a commercial point of view. This methodology will combine the main content-based, collaborative and knowledge-based approaches to build an innovative recommender engine for REACHR. The solution to be investigated and proposed will be evaluated using traditional performance metrics in recommender systems, such as error, precision and recall measures, as well as commercial evaluation measures, such as the conversion rate of recommendations. Among the expected results are: increasing the competitive differential of REACHR in relation to its competitors; Increase in the number of professional relocations; and reduced turn-over of companies, due to greater assertiveness in hiring. (AU)

Articles published in Agência FAPESP Newsletter about the research grant:
Articles published in other media outlets (0 total):
More itemsLess items
VEICULO: TITULO (DATA)
VEICULO: TITULO (DATA)

Scientific publications
(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)
DUARTE, FLAVIO GABRIEL; DE CASTRO, LEANDRO NUNES. A Framework to Perform Asset Allocation Based on Partitional Clustering. IEEE ACCESS, v. 8, p. 110775-110788, 2020. Web of Science Citations: 0.

Please report errors in scientific publications list by writing to: cdi@fapesp.br.