Puspita, Ayu (2025) Empirical tests comparison of the performance of the Self-Organizing Maps (SOM) post-dimensional reduction classification algorithm / AYU PUSPITA</p>. Diploma thesis, Universitas Negeri Malang.
Full text not available from this repository.Abstract
p This study examines the effectiveness of the Self-Organizing Maps (SOM) algorithm as a dimension reduction method on the performance of two classification algorithms namely Naive Bayes and Random Forest. The datasets used totaled 248 datasets and came from the UCI Machine Learning Repository with preprocessing carried out through normalization and selection of attributes. The results showed that in the Na iuml ve Bayes algorithm as many as 65.32% of the datasets experienced a decrease in accuracy after the implementation of SOM while 31.45% of the datasets experienced an increase and 3.23% of the datasets did not change. Meanwhile in the Random Forest algorithm the impact of dimension reduction has a percentage with 63.31% of the dataset experiencing a decrease in accuracy 32.67% experiencing an increase and 4.03% experiencing no change. Overall the implementation of SOM tends to reduce classification performance so the effectiveness of dimension reduction using SOM is highly dependent on the characteristics of each dataset. /p
| Item Type: | Thesis (Diploma) |
|---|---|
| Divisions: | Fakultas Teknik (FT) > Departemen Teknik Elektro (TE) > S1 Teknik Informatika |
| Depositing User: | library UM |
| Date Deposited: | 02 May 2025 04:29 |
| Last Modified: | 09 Sep 2025 03:00 |
| URI: | http://repository.um.ac.id/id/eprint/400175 |
Actions (login required)
 |
View Item |