Suchergebnisse

The system for biannual teachers' evaluation made by students, acquires a large amount of information that becomes difficult to be analyzed by a manager. It is true that delivers a final score, but it is calculated based only on the answers to the questions with several pre-defined answer options. In their questionnaire, there are also open questions in which students can express themselves freely. These answers contain valuable information, but an educational manager needs a long time to read them all individually. In this paper is proposed an automatic module for interpreting the answers. The module will report to the educational manager only the answers that need to be analyzed. The module uses natural language processing techniques, and supervised machine learning for selecting the answers to be analyzed in a certain context.