Fernando, Edward and Dr. Eng. Panca Mudjirahardjo, S.T., M.T. and Dr. Ir. Muhammad Aswin, M.T. (2022) Sistem Rekomendasi Mata Kuliah dengan Pendekatan Collaborative Filtering dan K-Means Clustering. Magister thesis, Universitas Brawijaya.
Abstract
Dalam perkuliahan, mahasiswa akan menjalani beberapa rangkaian mata kuliah, secara umum prodi akan menyiapkan daftar mata kuliah baik yang bersifat wajib dan pilihan. Prodi telah menyiapkan kurikulum yang berlaku dalam bentuk silabus, namun tidak jarang mahasiswa ragu dalam mengambil keputusan mata kuliah pilihan apa yang akan mereka ambil, terlebih lagi pada awal menentukan konsentrasi keminatan prodi. Hal ini dapat berakibat kurang maksimalnya perolehan nilai mata kuliah mahasiswa hingga mundurnya masa studi. Penelitian terdahulu telah melakukan pendekatan sistem pakar dan data mining untuk menentukan rekomendasi mata kuliah sesuai bakat dan kemampuan individu. Namun dalam implementasinya, tidak jarang mata kuliah yang direkomendasikan terlalu luas dan kurang sesuai dengan profil pengguna, dengan demikian perlu adanya metode optimalisasi rekomendasi untuk meningkatkan akurasi rekomendasi dengan meminimalisir keluarnya rekomendasi mata kuliah yang kurang sesuai dengan profil data uji. Dalam penelitian ini, sistem rekomendasi akan dibangun dengan pendekatan collaborative filtering menggunakan metode jaccard, euclidean distance, cosine similiarity, dan pearson correlation. Setelah itu, k-means clustering digunakan untuk mengoptimalisasi nilai akurasi dari sistem rekomendasi dengan . Dari penelitian yang dilakukan, pendekatan collaborative filtering menggunakan metode cosine similiarity menghasilkan nilai akurasi 64.76%, euclidean distance menghasilkan nilai akurasi 62.04%, jaccard menghasilkan nilai akurasi 53.66%, dan pearson correlation menghasilkan nilai akurasi 43.71%. Optimalisasi akurasi rekomendasi menggunakan k- means clustering juga berhasil meningkatkan performa nilai akurasi hingga 9.92%.
English Abstract
In lectures, students will undergo several series of courses, in general the study program will prepare a list of both mandatory and optional courses. The study program has prepared an applicable curriculum in the form of a syllabus, but it is not uncommon for students to hesitate in making decisions about which elective courses they will take, especially at the beginning of determining the concentration of study program interests. This can result in less than the maximum value of student course scores until the study period is delayed. Previous research has carried out an expert system approach and data mining to determine course recommendations according to individual talents and abilities. However, in its implementation, it is not uncommon for the recommended courses to be too broad and not in accordance with the user profile, thus it is necessary to have a recommendation optimization method to increase the accuracy of the recommendations by minimizing the issuance of course recommendations that are not in accordance with the test data profile. In this study, a recommendation system will be built using a collaborative filtering approach using the jaccard, euclidean distance, cosine similiarity, and pearson correlation methods. After that, k-means clustering is used to optimize the accuracy value of the recommendation system. From the research conducted, the collaborative filtering approach using the cosine similiarity method produces an accuracy value of 64.76%, euclidean distance produces an accuracy value of 62.04%, jaccard produces an accuracy value of 53.66%, and pearson correlation produces an accuracy value of 43.71%. Optimization of recommendation accuracy using k-means clustering also succeeded in increasing the performance of the accuracy value up to 9.92%.
| Item Type: | Thesis (Magister) |
|---|---|
| Identification Number: | 0422070007 |
| Uncontrolled Keywords: | collaborative filtering, k-means clustering, mata kuliah, sistem rekomendasi, collaborative filtering, course, k-means clustering, recommendation system. |
| Subjects: | 600 Technology (Applied sciences) > 621 Applied physics > 621.3 Electrical, magnetic, optical, communications, computer engineering; electronics, lighting > 621.38 Electronics, communications engineering > 621.381 Electronics |
| Divisions: | S2/S3 > Magister Teknik Elektro, Fakultas Teknik |
| Depositing User: | Zainul Mustofa |
| Date Deposited: | 10 Oct 2022 02:58 |
| Last Modified: | 10 Oct 2022 03:01 |
| URI: | http://repository.ub.ac.id/id/eprint/195475 |
![[thumbnail of DALAM MASA EMBARGO]](http://repository.ub.ac.id/style/images/fileicons/text.png) |
Text (DALAM MASA EMBARGO)
Edward Fernando.pdf Restricted to Registered users only until 31 December 2024. Download (5MB) |
Actions (login required)
 |
View Item |