Clustering Nilai Akademik Mahasiswa Teknik Informatika STMIK Widya Cipta Dharma Menggunakan DBSCAN

Ramadhan, Muhammad Arsyad (2022) Clustering Nilai Akademik Mahasiswa Teknik Informatika STMIK Widya Cipta Dharma Menggunakan DBSCAN. S1 Teknik Informatika thesis, STMIK Widya Cipta Dharma.

	Text 1843035-S1-Teknik Informatika.pdf Restricted to Repository staff only Download (3MB) | Request a copy
	Text 1843035-S1-Jurnal.pdf Restricted to Repository staff only Download (708kB) | Request a copy

Abstract

Penelitian ini bertujuan menerapkan metode DBSCAN untuk mengelompokkan mahasiswa TI Wicida dan mengetahui karakteristik cluster. Di samping itu, untuk mengetahui jumlah kelompok cluster yang terbentuk dari data nilai akademik mahasiswa TI STMIK Wicida tahun 2011–2017 terhadap status akhir. Data yang digunakan bersumber dari Puskom dan Prodi TI Wicida, serta web forlap kemendikbud yang di-scraping. Untuk memodelkan clustering Nilai Akademik terhadap Status Akhir dengan DBSCAN, digunakan Spatial Coordinate Way dan DBSCAN-rpr, di mana atribut data sumber dianalisis keterikatannya sedemikian sehingga menjadi variable untuk model, yaitu (X, Y(Z)) dengan X = Nilai Akademik, Y = Status Akhir, dan Z = Cuti. Berdasar model ini, maka Dimensi ; D = 3, minPts = 4, dan ɛpsilon ; ɛ = 1, 2, 3. Setelah eksekusi model clustering dengan Bahasa Pemrograman Python, didapatkan bahwa Silhouette Coefficient ; SC tertinggi = 0.5989 pada ɛ = 2. Dengan demikian, dipilih cluster-cluster ini karena paling baik, kuat, dan berkualitas daripada cluster dari ɛ yang lain. Penelitian menghasilkan 8 cluster dan 1 cluster noise dengan interpretasi khusus sebagai berikut : Cluster 0 dan 4 memberikan simpulan bahwa rata-rata IP Semester 1 - 3 tidak memberikan pola seragam terhadap ke-Lulus-an Tepat Waktu mahasiswa ; Cluster 1, 2, dan 3 merekomendasikan penggalian faktor lain penyebab Status Akhir Mengundurkan Diri. Cluster-cluster ini juga menunjukkan karakter khas STMIK Wicida, bahwa mahasiswa TI Wicida tetap bisa Lulus kuliah meskipun Tidak Tepat Waktu ; Cluster 6 memberikan simpulan bahwa nampaknya mahasiswa Mengundurkan Diri karena tidak berbakat atau berminat terhadap TI. Bagaimana pun, clustering ini menghasilkan 9 kelompok mahasiswa TI Wicida. =========================================================== This research is aimed to apply the DBSCAN method to cluster Wicida Informatics Engineering (IE) students and to determine characteristic of the clusters. In addition, it is also aimed to determine the number of formed clusters from the academic value data of STMIK Wicida IE students in 2011–2017 on to the final status. The data used is sourced from Puskom (Computer Center) and IE Study Program of Wicida, and also from kemendikbud forlap web which was being scrapping data. To modelling the clustering on Academic Value to Final Status with DBSCAN, it used Spatial Coordinate Way and DBSCAN-rpr, where the attributes of the source data were analyzed due to their relationship so that becoming variables for the model. It was (X, Y(Z)) with X = Academic Value, Y = Final Status, and Z = Leave. Based on this model, Dimension ; D = 3, minPts = 4, and ɛpsilon ; ɛ = 1, 2, 3. After running the clustering model with Python programming language, it found that the Silhouette Coefficient ; the highest SC = 0.5989 at ɛ = 2. Thus, these formed clusters were chosen, because they were the best, strong, and qualified than the formed clusters of the other ɛ. The research resulted 8 clusters and 1 noise cluster with the following special interpretations : Cluster 0 and 4 gave the conclusion that the IP Semester (Semester Grade Point ; SGP) 1 - 3 average did not provide a uniform pattern for the students’ On-Time Graduate-ion ; Clusters 1, 2, and 3 recommended to explore the other factors causing the Final Status of Resign-ation. These clusters also pointed out the distinctive character of STMIK Wicida, that Wicida IE students could still be Graduate even Not On Time ; Cluster 6 gave the conclusion that it looked like the students Resign because of they were not talented or interested in IE. Therefore, this clustering resulted 9 formed clusters of Wicida IE students.

Item Type:	Thesis (S1 Teknik Informatika)
Additional Information:	Pembimbing 1 : Eka Arriyanti, S.Pd., M.Kom., I.G. Pembimbing 2 : Pitrasacha Adytia, S.T., M.T.
Uncontrolled Keywords:	cluster, TI, Wicida, mahasiswa, model
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:	Teknik Informatika
Depositing User:	Mr Muhammad Arsyad Ramadhan
Date Deposited:	26 Aug 2022 02:15
Last Modified:	26 Aug 2022 02:15
URI:	http://repository.wicida.ac.id/id/eprint/4643

Actions (login required)

View Item