Salam, Nur Shafiya Nabilah (2019) Analisis Sentimen Opini Mahasiswa Terhadap Saran Kuesioner Penilaian Kinerja Dosen dengan Menggunakan Tf-Idf dan K-Nearest Neighbor. Sarjana thesis, Universitas Brawijaya.
Abstract
Fakultas Ilmu Komputer Universitas Brawijaya (FILKOM UB) mengevaluasi kinerja dosen setiap semester yang dilakukan oleh Tim Unit Jaminan Mutu (UJM). Evaluasi dilakukan dengan penyebaran kuesioner pada Sistem Informasi Akademik Mahasiswa (SIAM). Ada dua bagian pada kuesioner yaitu kolom pilihan ganda dan kolom saran. Setelah data kuesioner terkumpul, Tim UJM mengolah data dari kolom saran secara manual. Data pada kolom saran tentang perspektif mahasiswa terhadap kinerja dosen, hingga saat ini belum diolah secara mendalam untuk dijadikan bahan evaluasi. Untuk itu penelitian ini dilakukan agar dapat membantu Tim UJM mengolah data dari kolom saran yaitu menggunakan Analisis Sentimen pada tingkat kalimat, dengan metode klasifikasi K-Nearest Neighbor. Dari 2210 data opini mahasiswa Program Studi Teknologi Informasi yang dianalisa dalam tiga semester yaitu Semester Genap 2016/2017, Ganjil 2017/2018, dan Genap 2017/2018, diperoleh hasil klasifikasi dengan rata-rata Accuracy 81%. Hasil penelitian ini berupa daftar data opini dan grafik frekuensi data opini yang terklasifikasi dan divisualisasikan menjadi tampilan dashboard, serta dapat ditampilkan dengan fungsi filter berdasarkan nama dosen dan mata kuliah.
English Abstract
Faculty of Computer Science on the University of Brawijaya evaluates lecturer performance every semester and the evaluation program was carried out by the Quality Assurance Unit (UJM) Team. UJM Team does the evaluation by distributing questionnare within the Student Academic Information System (SIAM). There were two section on the questionnaire, multiple choice section and the comment section. After collecting data, UJM Team processed the data from the comment section manually. The comment section contained student's perspective of the lecturer performance, which has never been processed to date to be used as an evaluation material. Therefore, this study is performed to assist UJM Team to process the data from the comment section using Sentiment Analysis on sentence level with KNearest Neighbor classification method. Out of 2210 opinion data from students of the Information Technology Study Program analyzed from three semester period in the Even Semester 2016/2017, Odd Semester 2017/2018, and Even Semester 2017/2018, the classification result in an average accuracy of 81%. The result of this study is a list of opinions and bar chart showing the frequency of opinions that have been classified and will be displayed as a dashboard. This data opinions can also be filterred by lecturer’s subjects and his/her name.
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | SKR/FILKOM/2019/365/051905683 |
| Uncontrolled Keywords: | Lecturer Evaluation, Sentiment Analysis, Classification, K-Nearest Neighbor |
| 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: | Endang Susworini |
| Date Deposited: | 30 Jul 2020 05:51 |
| Last Modified: | 30 Jul 2020 05:51 |
| URI: | http://repository.ub.ac.id/id/eprint/171656 |
Actions (login required)
 |
View Item |