ABSTRACT
One part of improving and developing lecturer pedagogical competence and evaluating the substance of the courses taught in the lecture process is to use Lecturer Evaluation Effectiveness by Students (EDOM). The EDOM is a feedback questionnaire that can be used by educators to assess and evaluate pedagogical competencies that they have and aims to make continuous im-provements to various matters related to the teaching process in the classroom. In addition to questions in the form of a Likert scale, EDOM also accommodates open-ended questions. However, this type of question tends not to be considered in the feedback assessment and is not explored more deeply to gain insight. Therefore, in this study text mining was carried out to find out the sentiments given by students, whether positive or negative sentiments using Convolutional Neural Network (CNN). The experimental results showed that CNN was able to achieve accuracy, precision, recall, and F1-Score of 87.95%, 87%, 78%, and 81%, respectively.
Keywords:
Convolutional Neural Network, Deep Learning, Opinion Mining, Sentiment Analysis, Teaching Evaluation.
