Breast cancer disease classification using fuzzy-ID3 algorithm based on association function

Nur Farahaina Idris, Mohd Arfian Ismail, Mohd Saberi Mohamad, Shahreen Kasim, Zalmiyah Zakaria, Tole Sutikno
2022 IAES International Journal of Artificial Intelligence (IJ-AI)  
Breast cancer is the second leading cause of mortality among female cancer patients worldwide. Early detection of breast cancer is considerd as one of the most effective ways to prevent the disease from spreading and enable human can make correct decision on the next process. Automatic diagnostic methods were frequently used to conduct breast cancer diagnoses in order to increase the accuracy and speed of detection. The fuzzy-ID3 algorithm with association function implementation (FID3-AF) is
more » ... oposed as a classification technique for breast cancer detection. The FID3-AF algorithm is a hybridisation of the fuzzy system, the iterative dichotomizer 3 (ID3) algorithm, and the association function. The fuzzy-neural dynamicbottleneck-detection (FUZZYDBD) is considered as an automatic fuzzy database definition method, would aid in the development of the fuzzy database for the data fuzzification process in FID3-AF. The FID3-AF overcame ID3's issue of being unable to handle continuous data. The association function is implemented to minimise overfitting and enhance generalisation ability. The results indicated that FID3-AF is robust in breast cancer classification. A thorough comparison of FID3-AF to numerous existing methods was conducted to validate the proposed method's competency. This study established that the FID3-AF performed well and outperform other methods in breast cancer classification.
doi:10.11591/ijai.v11.i2.pp448-461 fatcat:v2x5comvkren5nsoplubs2pofu