Mining of Conclusions of Student Texts for Automatic Assessment

Writing a thesis involves complying with certain requirements and rules established by institutional guides. So, students have guidelines to follow when developing their first draft. Generally this draft presents deficiencies, which has to be polished with the help of the academic advisor to reach an acceptable document. However, this task is repeated every time a student prepares his thesis, becoming extra time spent by the student and the advisor. Our work aims to help the student improve the writing, based on natural language processing techniques. For the current study, we focus primarily on the conclusions section of a thesis, a central element when concluding the research. In this paper we present a Mining Component that includes three models: Coverage, Opinion and Speculation. Our system seeks to assess a conclusion taking into account the general objective, the evidence of value judgments, and the presence of future work as a result of reflection of the student. We provide initial models and their evaluations.