Application of the Support Vector Machine and Neural Network Model Based on Particle Swarm Optimization for Breast Cancer Prediction
Autor: | Esty Purwaningsih |
---|---|
Rok vydání: | 2019 |
Předmět: |
Artificial neural network
business.industry Computer science lcsh:Electronics lcsh:TK7800-8360 Particle swarm optimization Value (computer science) Confusion matrix Pattern recognition breast cancer support vector machine neural network particle swarm optimization rapid miner 02 engineering and technology lcsh:QA75.5-76.95 Field (computer science) Support vector machine 03 medical and health sciences ComputingMethodologies_PATTERNRECOGNITION 0302 clinical medicine Software 030220 oncology & carcinogenesis 0202 electrical engineering electronic engineering information engineering Gasal 1920-Esty Purwaningsih 020201 artificial intelligence & image processing lcsh:Electronic computers. Computer science Artificial intelligence business |
Zdroj: | Sinkron, Vol 4, Iss 1, Pp 66-73 (2019) Sinkron : Jurnal dan Penelitian Teknik Informatika; Vol. 4 No. 1 (2019): SinkrOn Volume 4 Number 1, October 2019; 66-73 |
ISSN: | 2541-2019 2541-044X |
DOI: | 10.33395/sinkron.v4i1.10195 |
Popis: | There are several studies in the medical field that classify data to diagnose and analyze decisions. To predict breast cancer, this study compares two methods, the Support Vector Machine method and the Neural Network method based on Particle Swarm Optimization (PSO) which is intended to determine the highest accuracy value in the Coimbra dataset data. To implement the Support Vector Machine and Neural Network method based on PSO, RapidMiner software is used. Then the application results are compared using Confusion Matrix and ROC Curve. Based on the accuracy of the two models, it is known that the PSO-based Neural Network model has a higher accuracy value of 84.55% than the results of the PSO-based Vector Support Machine with an accuracy value of 80.08%. The calculation results, the accuracy of the AUC performance obtained by the results of the study are, the two methods are PSO-based Neural Network with AUC value of 0.885 and PSO-based Support Vector Machine with a value of 0.819 included in the category of Good Classification. |
Databáze: | OpenAIRE |
Externí odkaz: |