Sulaeman, Achmad Firmansyah (2019) Analisis Sentimen Opini Mahasiswa Terhadap Saran Evaluasi Kinerja Dosen Menggunakan TF-IDF dan Support Vector Machine. Sarjana thesis, Universitas Brawijaya.
Abstract
Dosen khsususnya di perguruan tinggi senantiasa harus meningkatkan pengetahuan dan cara mengajar yang efektif dan efisien guna dapat memberikan ilmu yang dapat di terima dengan baik oleh mahasiswa. Maka dari itu, perlu dilakukannya peningkatan dan proses evaluasi untuk kualitas pengajaran dan standarisasi akademik sehingga tujuan dari pengajaran dapat tercapai sesuai dengan yang diharapkan. Universitas Brawijaya Malang khususnya Fakultas Ilmu Komputer (FILKOM), menerapkan adanya pengisian kuesioner online untuk penilaian kinerja dosen pada setiap pergantian semester. Opini mahasiswa dalam kuesioner merupakan salah satu hal penting yang di jadikan bahan dalam proses evaluasi kinerja. Namun saat ini data opini yang dimiliki Tim UJM Jurusan Sistem Informasi masih berupa tulisan dan belum adanya teknologi pengklasifikasian analisis data yang mampu memberikan informasi mengenai perspektif mahasiswa dalam pengajaran dosen. Untuk dapat melihat perspektif mahasiswa dalam pengajaran yang di lakukan oleh dosen, dapat di lakukan analisis sentimen pada tingkat kalimat. Metode yang digunakan dalam klasifikasi data opini mahasiswa adalah Support Vector Machine dengan menggabungkan Term Frequency Inverse Document Frequency sebagai pembobotan kata. Penelitian ini menggunakan data bahasa Indonesia dalam komentar mahasiswa pada 3 semester yaitu semester 2017 ganjil, semester 2017 genap dan semester 2018 genap. Pengujian terhadap klasifikasi sentimen mendapatkan hasil rata – rata Accuracy sebesar 82%, Precision sebesar 74%, Recall sebesar 72% dan F-1 Score sebesar 73%. Visualisasi Dashborad menjadi keluaran dari penelitian ini berdasarkan hasil klasifikasi dengan menampilkan 3 halaman semester yang berbeda. Uji Usability dengan System Usability Scale terhadap 2 responden menghasilkan rata – rata sebesar 70 yang berarti Tim UJM Jurusan Sistem Informasi menerima dengan baik visualisasi Dashboard.
English Abstract
Lecturers especially in institution should always need to improve their knowledge and effective and efficient teaching methods to be able to provide knowledge that can be well received by students. Therefore, it is necessary to improve and evaluate the quality of teaching methods and academic standardization so that teaching goal can be achieved as it has been expected. Brawijaya University especially Computer Science Faculty implements of filling the online questionnaire due to evaluating lecturers' performance at each semester. Student opinion in the questionnaire is one of the important things that are made into material in the performance evaluation process. However , the opinion data that held by the UJM Team of Information Systems Department is still in the form of writing and still no processed by using data analysis classification technology that is able to provide information about student perspectives in teaching lecturers. To be able to get student's perspective on teaching that is done by the lecturer, could be done by performing sentiment analysis. Support Vector Machine in combining with Term Frequency Inverse Document Frequency as word weighting can be used as a method in classification of student opinion data. This research managed to used student opinion data in bahasa for 3 last semesters, in 2017 odd semester, 2017 even semester and 2018 even semester. The Test that have been conducted on the results of sentiment classification show the average of 82% on Accuracy, 74% on Precision, 72% of Recall and 73% of F-1 Score. the visualization of Dashboard displaying 3 different kind of semester pages, bar chart and pie chart were present in order to show the output of this research based on the classification result. The average result of usability Test using System Usability Scale for 2 respondents show result in 70, which means that the UJM Team of the Information Systems Department receives a good visualization of the Dashboard at it self
Other obstract
-
Item Type: | Thesis (Sarjana) |
---|---|
Identification Number: | SKR/FILKOM/2019/450/051905833 |
Uncontrolled Keywords: | Opini Mahasiswa, Analisis Sentimen, Klasifikasi, Support Vector Machine, Visualisasi Data, Student Opinion, Sentiment Analysis, Classification, Support Vector Machine, Data Visualization |
Subjects: | 000 Computer science, information and general works > 004 Computer science > 004.2 System analysis and design, computer architecture, performance evaluation |
Divisions: | Fakultas Ilmu Komputer > Sistem Informasi |
Depositing User: | Nur Cholis |
Date Deposited: | 30 Jul 2020 06:42 |
Last Modified: | 25 Oct 2021 02:26 |
URI: | http://repository.ub.ac.id/id/eprint/172062 |
Preview |
Text
Achmad Firmansyah Sulaeman (2).pdf Download (4MB) | Preview |
Actions (login required)
![]() |
View Item |