Metode Algoritma Support Vector Machine (SVM) Linier Dalam Memprediksi Kelulusan Mahasiswa

Abstract

The accumulation of student databases can occur if students are unable to complete their studies, namely graduating at a predetermined time. Data mining techniques are often used to process student data so that they can produce predictions of student graduation in order to graduate at a predetermined time. One of the data mining techniques that is often used is the Support Vector Machine (SVM) algorithm. This study aims to analyze the performance of the SVM algorithm to produce a predictive model of student graduation in order to graduate at a predetermined time in the Public Health Study Program, Faculty of Public Health, Deli Husada Health Institute. The method used in this study is a linear SVM algorithm starting from data retrieval by selecting the attributes that will be used for the next stage, data processing consists of cleaning data whose contents do not exist and data transformation which is the determination of the category of each data, modeling is done with the SVM algorithm. from training data and testing and evaluation data to validate and measure the accuracy of the model. The test results with the amount of training data as much as 70% and testing data as much as 30% shows that the linear SVM algorithm provides an accuracy value of 90%

Implementation of Exploratory Data Analysis and Artificial Neural Networks to Predict Student Graduation on-Time

“…By applying the Naïve Bayes algorithm, which uses a dataset of alumni from several universities in Palembang as test data, the results obtained show that the accuracy level of the algorithm is 81% (Qisthiano et al, 2021). Other algorithms, such as support vector machines, were also applied, and the results obtained were 90% accuracy (Bangun, Mawengkang, & Efendi, 2022). Application of the K-Nearest Neighbors algorithm with the highest accuracy rate of 98.5% (Muliono, Lubis, & Khairina, 2020).…”