IMPLEMENTASI METODE BACKPROPAGATION NEURAL NETWORK UNTUK MENDIAGNOSIS PENYAKIT KANKER PAYUDARA

  • Muhamad Bahrul Ulum
  • Ravie Kurnia Laday
Keywords: Neural network, breast cancer, classification

Abstract

Breast cancer is one of lethal diseases feared by women. Early symptoms that are hard to recognize are the main factors which caused early treatment to patients receive few attention. One of the ways to solve this is by diagnosing someone to know whether or not the individual harbor malignant cancer. The research in t his paper will explain and implement backpropagation neural network method to diagnose breast cancer. The implementation process is by conducting tests using breast cancer data set. In initial stage, backpropagation neural network method conducted training to training data set and then the model obtained from the training’s network would be tested on test data. The experiment used 569 observation data divided into two parts; which were 70% training data set and 30% test data set for each class. Based on the result of the experiment, backpropagation neural network method can be used to diagnose breast cancer. The accuracy value obtained from this method was 100%. Choosing training data sets that will be used for learning will affect the accuracy level of the system.

References

[1] Mangasarian OL, Street WN, Wolberg WH. Breast cancer diagnosis and prognosis via linear programming. Operations Research. 43(4):570-577. 1995.
[2] Ruby CW, Chih JL. A Bayesian Approximation Method for Online Ranking. Journal of Machine Learning Research. 12:267-300. 2011.
[3] Stéphane R, Joelle P, Brahim Cd,Pierre K. A Bayesian Approach for Learning and Planning in Partially Observable Markov Decision Processes. Journal of Machine Learning Research. 12:1729-1770. 2011.
[4] UCI Data Set. 2015. Breast Cancer Wisconsin (Diagnotic) Data Set.http://archive.ics.uci.edu/ml/datasets/ [diakses maret 2015]
[5] Wisconsin. 2015. Breast Cancer Wisconsin (Diagnotic) Data Set. http://archive.ics.uci.edu/ml/datasets/Breast+Cancer+Wisconsin+ %28 Diagnostic%29. (Diakses maret 2015).
Published
2016-10-29
Section
Articles