Open AccessApplication of Rough Sets to Predict the Breast Cancer Risk Association with Routine Blood Analyses
Author(s) -
Amr H. AbdelHaliem,
Mohammed A. Atiea,
M. El-Sayed Wahed,
Mohammed Metwally
Publication year - 2021
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.b8235.0110321
Subject(s) - rough set , sort , receiver operating characteristic , computer science , sensitivity (control systems) , breast cancer , data mining , machine learning , artificial intelligence , association rule learning , set (abstract data type) , cancer , medicine , engineering , electronic engineering , information retrieval , programming language
For women around the globe, breast cancer has been a significant cause of mortality. Around the same time, early diagnosis and high cancer prediction precision are critical to improving the quality of care and the recovery rate of the patient. Expert systems and machine learning techniques are gaining prominence in this area as a result of efficient classification and high diagnostic ability. This paper introduces a model of hybrid prediction (RS QA) based on a rough set theoryand a quasi-optimal (AQ) rule induction algorithm. To find a minimal set of attributes that completely define the results, a rough set tool is used. The selected characteristics were collected, ensuring the high standard of the classification. Then to produce the decision rules, we use the quasi-optimal (AQ) rule induction algorithm. These hybrid prediction models allow expert systems to be built based on the conceptual rules of the IF THEN sort. The suggested experiment is performed using the Coimbra Breast Cancer Dataset (BCCD) based on sets of measures that can be obtained in routine blood tests. Using classification precision, sensitivity, specificity, and receiver operating characteristics (ROC) curves, the efficiency of our suggested approach was assessed. Experimental results indicate the highest classification accuracy (91.7 percent), sensitivity (83.3 percent), and precision (94.3) obtained by the proposed (RS_QA) model.