HYPERPARAMETER TUNING IN ARTIFICAL NEURAL NETWORKS FOR BREAST CANCER CLASSIFICATION USING PARTICLE SWARM OPTIMIZATION

ALFAROQ, AKBAR UMAR (2026) HYPERPARAMETER TUNING IN ARTIFICAL NEURAL NETWORKS FOR BREAST CANCER CLASSIFICATION USING PARTICLE SWARM OPTIMIZATION. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (COVER) 19081010043-cover.pdf Download (1MB)
	Text (BAB I) 19081010043-bab1.pdf Download (336kB)
	Text (BAB II) 19081010043-bab2.pdf Restricted to Repository staff only until 23 July 2029. Download (601kB) \| Request a copy
	Text (BAB III) 19081010043-bab3.pdf Restricted to Repository staff only until 23 July 2029. Download (583kB) \| Request a copy
	Text (BAB IV) 19081010043-bab4.pdf Restricted to Repository staff only until 23 July 2029. Download (810kB) \| Request a copy
	Text (BAB V) 19081010043-bab5.pdf Download (248kB)
	Text (DAFTAR PUSTAKA) 19081010043-daftarpustaka.pdf Download (235kB)

Abstract

Artificial Neural Networks (ANNs) are widely used in medical data classification due to their ability to recognize complex patterns. However, the performance of ANNs is greatly influenced by the selection of appropriate hyperparameter values. This study aims to evaluate the effectiveness of Particle Swarm Optimization (PSO) in optimizing hyperparameters for an Artificial Neural Network model used in breast cancer classification. The study was conducted using the Wisconsin Breast Cancer Dataset from the UCI Machine Learning Repository, with hyperparameter optimization applied to the MLPClassifier model from Scikit-learn. Model evaluation was performed using the 3-fold cross-validation method to obtain more stable performance. The experimental results showed an increase in classification accuracy from 93.1% to 95.6%, or by 2.5%, compared to the baseline model without optimization. These results indicate that the use of PSO has the potential to improve the performance of Neural Network models in breast cancer classification tasks. Keywords: Particle Swarm Optimization, Hyperparameter Tuning, Artificial Neural Network, Breast Cancer Classification, MLPClassifier

Item Type:

Thesis (Undergraduate)

Contributors: