Open AccessImplementasi Fuzzy C-Means dan Possibilistik C-Means Pada Data Performance Mahasiswa
Author(s) -
Gadis Retno Apsari,
Mohammad Syaiful Pradana,
Novita Eka Chandra
Publication year - 2020
Publication title -
ujmc (unisda journal of mathematics and computer science)
Language(s) - English
Resource type - Journals
eISSN - 2579-907X
pISSN - 2460-3333
DOI - 10.52166/ujmc.v6i2.2392
Subject(s) - matlab , fuzzy logic , cluster analysis , computer science , data mining , attendance , point (geometry) , value (mathematics) , function (biology) , mathematics education , artificial intelligence , mathematics , machine learning , biology , economic growth , geometry , evolutionary biology , economics , operating system
Students are the most important component in a university, especially private universities especially Universitas Islam Darul ‘ulum (Unisda) Lamongan. One of the most important roles of students for higher education is achievement. This study aims to determine the role of Fuzzy Clustering in classifying student performance data. The data includes GPA (Grade Point Average), ECCU (Extra-Curricular Credit Unit), attendance, and students' willingness to learn. So that groups of students who have the potential to have achievements can be identified. In this case, the grouping of student performance data uses Fuzzy Clustering by applying the Fuzzy C-Means (FCM) and Possibilistic C-Means (PCM) algorithms with the help of Matlab. In the FCM algorithm, the membership degree is updated so as to produce a minimum objective function value. Meanwhile, the PCM algorithm uses a T matrix, which shows the peculiarities of the data which are also based on minimizing the objective function.