Assessing Thesis Conclusions by their Connectedness with Goal, Judgment and Speculation

Writing a thesis involves complying with certain rules and requirements established by institutional guides of universities. Students, often being too inexperienced to create good written documents, have guidelines to follow when developing their first drafts. This study seeks to help students improve their first writings, based on natural language processing techniques. We focus primarily on the conclusion section of a thesis, a central element when completing a research project. In this paper, a conclusion analyzer that includes three models: goal connectedness, judgment and speculation is presented. Such subsystems try to evaluate the main expected features in conclusions, specifically the connectedness with the general objective, the evidence of value judgments, and the presence of future work as a result of the student’s reflection. In the study, we provide initial models, internal exploration of conclusions, and evaluations of our approach. We found across the three features evaluated that graduate level student texts outperformed those of undergraduate level. The behavior provides evidence, that students with more practice in writing a scientific paper or thesis (at the graduate level), have better writing skills.