Data Mining on LinkedIn Data to Define Professional Profile via MineraSkill Methodology

• Autor(a) principal: Caldeira, Dayane C. M. F. [UNESP]
• Data de Publicação: 2017
• Outros Autores: Correia, Ronaldo C. M. [UNESP], Spadon, Gabriel [UNESP], Eler, Danilo M. [UNESP], Olivete-, Celso [UNESP], Garcia, Rogerio E. [UNESP], IEEE
• Tipo de documento: Artigo de conferência
• Idioma: por
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://hdl.handle.net/11449/160146
• topic: Data Mining; Professional Profile; Natural Language Processing; MineraSkill

Social networks are of significant analytical interest. This is because their data are generated in great quantity, and intermittently, besides that, the data are from a wide variety, and it is widely available to users. Through such data, it is desired to extract knowledge or information that can be used in decision-making activities. In this context, we have identified the lack of methods that apply data mining techniques to the task of analyzing the professional profile of employees. The aim of such analyses is to detect competencies that are of greater interest by being more required and also, to identify their associative relations. Thus, this work introduces MineraSkill methodology that deals with methods to infer the desired profile of a candidate for a job vacancy. In order to do so, we use keyword detection via natural language processing techniques; which are related to others by inferring their association rules. The results are presented in the form of a case study, which analyzed data from LinkedIn, demonstrating the potential of the methodology in indicating trending competencies that are required together.