Skripsi
OPTIMASI ALGORITMA NAÏVE BAYES MENGGUNAKAN ARTIFICIAL BEE COLONY UNTUK KLASIFIKASI DATA PENDERITA PENYAKIT STROKE
This research aims to optimize the Naïve Bayes algorithm for feature selection of stroke disease patient data using the Artificial Bee Colony (ABC) method. Naïve Bayes is one of the simple yet effective classification algorithms. However, Naïve Bayes performance can decrease when there are irrelevant features in the data. To overcome this problem, a feature selection method with Artificial Bee Colony is applied. ABC is a population-based optimization algorithm that mimics the food search behavior of honey bees with employed bee, onlooker bee, and scout bee types, which are effective in finding optimal solutions. In this study, ABC is used to select the most relevant feature subset, thereby improving the classification accuracy of Naïve Bayes. The results show that Naïve Bayes has an accuracy of 86.10% without oversampling, while with feature selection using Artificial Bee Colony, the accuracy increases to 95.20% without oversampling. With the application of oversampling, the accuracy of Naïve Bayes reaches 82.23%, while the accuracy with Artificial Bee Colony optimization is 84.5%. The experimental results show that the combination of Naïve Bayes with ABC significantly improves the prediction accuracy compared to Naïve Bayes without feature selection. This approach is effective for stroke disease diagnosis.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2407003784 | T146611 | T1466112024 | Central Library (REFERENCES) | Available |
No other version available