Open AccessIndonesian sentiment summarization for lecturer learning evaluation by using textrank algorithm
Author(s) -
I Gede Mahendra Darmawiguna,
Gede Aditra Pradnyana,
I B Jyotisananda
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1810/1/012024
Subject(s) - automatic summarization , ranking (information retrieval) , computer science , graph , subject (documents) , indonesian , artificial intelligence , algorithm , natural language processing , information retrieval , machine learning , mathematics education , world wide web , theoretical computer science , mathematics , linguistics , philosophy
Learning evaluation is a very important process to assess lecturer performances. Evaluation is carried out at the end of the semester by students. Student can give comments, critics, suggestions related to the subject taught by the lecturers. To analyse student comments, the lecturer can read the entire comments one by one. The main focus in this research was to develop the system that will be applied in summarizing student comments on the implementation of course learning. The algorithm that was used to perform text summarization was the TextRank algorithm. TextRank is a graph-based ranking algorithm (a graph with a ranking model) for processing text from natural or human language documents. We used 100 data sentiments by students to lecturers who teach one course in one semester. It was directly taken from SIAK (academic information system) by using web services. Based on the evaluation of the summaries by the expert, it was found that the textRank algorithm is quite representative in performing summarization for sentiments with the accuracy of 82%.