Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Bulletin of Applied Mathematics and Mathematics Education

Optimization of feature selection on semi-supervised data Wijayanti, Dian Eka; Afriyani, Sintia; Surono, Sugiyarto; Dewi, Deshinta Arrova
Bulletin of Applied Mathematics and Mathematics Education Vol. 4 No. 2 (2024)
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/bamme.v4i1.11104

Abstract

This research explores feature selection optimization in semi-supervised text data by utilizing the technique of dividing data into training and testing sets and implementing pseudo-labeling. Proportions of data division, namely 70:30, 80:20, and 90:10, were used as experiments, employing TF-IDF weighting and PSO feature selection. Pseudo-labeling was applied by assigning positive, negative, and neutral labels to the training data to enrich information in the classification model during the testing phase. The research results indicate that the linear SVM model achieved the highest accuracy with a 90:10 data division proportion with a value of 0.9051, followed by Random Forest, which had an accuracy of 0.9254. Although RBF SVM and Poly SVM yielded good results, KNN showed lower performance. These findings emphasize the importance of feature selection strategies and the use of pseudo-labeling to enhance the performance of classification models in semi-supervised text data, offering potential applications across various domains that rely on semi-supervised text analysis.
Identification of fingerprint image with Minkowski distance algorithm approach Tulloh, Wisnu Catur Rachmad; Wijayanti, Dian Eka
Bulletin of Applied Mathematics and Mathematics Education Vol. 3 No. 2 (2023)
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/bamme.v3i2.9949

Abstract

In the digital era, fingerprint identification plays a critical role in information technology administration. Various studies have been conducted to improve the fingerprint identification process, but there are still cases of identification failures that are fatal. This research discusses fingerprint identification with the Minkowski distance method. The data of fingerprints are taken from Mathematics students and the Kaggle site. Data analysis includes the steps of image retrieval, dimensioning, conversion to grayscale, pattern matching, and accuracy measurement. Results show an improvement in data accuracy with a structured approach to data capture and preprocessing. Results from primary data obtained an accuracy of 56.67% while from secondary data obtained an accuracy of 93%.