Author(s)

M. C. S. Lopes, M. Onoda, V. M. Bastos & N. F. F. Ebecken

Abstract

This work presents the study related to a text database with employees’ opinions and complaints about general situations that occurred in the working environment in a large company. The study intends to discover possible conflicts, recurrent problems inside the company and so on. In order to find these points out, a text mining commercial tool was used to make explicit the hidden knowledge. Thus, we are able to evaluate how text data is organized and the relevant information to be extracted. The results obtained can define possible actions to improve the relationship between employees and the quality of the working environment. Keywords: text mining, clustering. 1 Introduction The objective of this work is to show that a text mining tool, Insight Discoverer Clusterer by Temis [1], can assist the company in the solution of typical problems of the employees’ work environment, identifying problems that can indicate or cause the low productivity of one or more sectors of the company, or presenting practical and efficient solutions, capable to prevent problems that could appear in a short time. From messages sent to a centralizer department, the employees indicate doubts, suggestions and claims, or until they praise the way how the problems are solved. These messages are analyzed by the text-mining tool, presenting the argued subjects. In general the analyses in this type of data are made through auditorship, where the approach is the process of treatment of the messages, and not its content.

Keywords

text mining, clustering.