Article Per Year (5 Year)
p-Index From 2019 - 2024
0.61
P-Index
This Author published in this journals
All Journal JURNAL MEDIA INFORMATIKA BUDIDARMA
Pahala Sirait
STMIK Mikroskil, Medan
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
Optimalisasi Kinerja Klasifikasi Melalui Seleksi Fitur dan AdaBoost dalam Penanganan Ketidakseimbangan Kelas Tanti Tanti; Pahala Sirait; Andri Andri
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 5, No 4 (2021): Oktober 2021
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v5i4.3280

Abstract

One of the problems in data mining classification is class imbalance, where the number of instances in the majority class is more than the minority class. In the classification process, minority classes are often misclassified, because machine learning prioritizes the majority class and ignores the minority class so that this can cause the classification performance to be not optimal. The purpose of this study is to provide a solution to overcome class imbalances so as to optimize classification performance using chi-square and adaboost on one of the classification algorithms, namely C5.0. In this study, the majority class in the dataset used is dominated by the negative class, so the performance appraisal should focus more on the positive class. Therefore, a more suitable assessment is recall/sensitivity/TPR because the resulting value only depends on the positive class. The results showed that both methods were able to increase the recall/sensitivity/TPR value, meaning that the application of chi-square and adaboost was able to improve the classification performance of the minority class
Pembobotan Kriteria Dalam Prediksi Meningitis Tuberkulosis Menggunakan Metode SWARA dan Nearest Neighbor Dwika Assrani; Pahala Sirait; Andri Andri
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 5, No 4 (2021): Oktober 2021
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v5i4.3276

Abstract

Weights greatly affect the value and results of decisions or predictions of a test data, a problem that often occurs in the results of the prediction process is the weighting of symptom attributes which is less certain of the value of the weight, thus affecting the prediction results and the level of accuracy of a prediction itself. This study predicts a data using the Nearest Neighbor method where in the process of predicting the attribute weight value does not yet have a definite value for testing. Then we need an attribute weighting for each test attribute to get a definite weight value result. One method that can be applied to attribute weighting is the SWARA method. Based on research conducted to compare the prediction of Meningitis Tuberculosis without SWARA weighting and with SWARA weighting, testing with a ratio of 90:10, 80:20, 70:30 results in disease prediction using the Nearest Neighbor method, there are differences in results and levels of prediction accuracy and the process in prediction helps shorten the time to find prediction results, the highest prediction result using the swara method is 100% accurate and without weighting method is 91%.
Optimasi Klasifikasi Bayesian Network Melalui Reduksi Attribute Menggunakan Metode Principal Component Analysis Surizar Rahmi; Pahala Sirait; Erwin Setiawan Panjaitan
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 4, No 4 (2020): Oktober 2020
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v4i4.2370

Abstract

Dimensionality reduction is a hot topic being discussed in its development has been carried out in various fields of research one of which is machine learning by reducing can reduce the capacity of dimensions without reducing (eliminating) information contained in the data. Principal Component Analysis is one of the proven reduction techniques capable of reducing data capacity without significantly eliminating the information contained in the dataset. In this research attribute reduction using principal component analysis using a dataset of factors affecting employee absence was taken from the University of California repository at Irvine (UCI). Combination with Bayesian Network to classify data as a comparison between before and after attribute reduction. This can be seen in the initial results before the reduction with an accuracy of 100% and after the fifth attribute reduction there is a decrease in accuracy by 89,7%
Co-Authors Andri Andri Dwika Assrani Erwin Setiawan Panjaitan Surizar Rahmi Tanti