ASSOCIATIVE CLASSIFICATION APPROACH CAN PREDICT PROSTATE CANCER BASED ON THE EXTRACTED ASSOCIATION RULES
İpek Balıkçı Çiçek, Zeynep Küçükakçalı, Cemil Çolak · The Journal of Cognitive Systems · 2020
Aim: This study aims to classify the diagnosis status of prostate cancer and determine the related factors by applying the associative classification method, one of the data mining methods, to the open-access prostate cancer data set. Materials and Methods: In the current study , an open-access data set named Prostate Cancer is used for classification. The performance of the associative classification model is evaluated using the classification performance metrics such as sensitivity, selectivity, accuracy, balanced accuracy, negative predictive value, positive predictive value, and F1-score. Results: According to the prostate cancer classification results obtained from the associative classification model, the accuracy, balanced accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F1 score values were obtained as 0.968, 0.789, 0.9, 0.879, 0.938, 0.882 and 0.923, respectively. Conclusion: In the analysis of the open-access data set, the proposed associative classification model has distinctively successful results in classifying prostate cancer on the performance metrics.