To search, Click below search items.

 

All Published Papers Search Service

Title

Data Mining Algorithms for Classification of Diagnostic Cancer Using Genetic Optimization Algorithms

Author

Rafaqat Alam Khan, Taseer Suleman, Muhammad Sajid Farooq, Muhammad Hassan Rafiq, Muhammad Arslan Tariq

Citation

Vol. 17  No. 12  pp. 207-212

Abstract

The breast tumor is the primary driver of female casualty everywhere throughout the world and the real area of study from a long time but with slighter development than an-ticipated. Numerous establishments and as-sociations are working in this field to prompt to a conceivable arrangement of the issue or to prompt to additionally comprehension of the issue. Numerous past inquiries about the said were contemplated for improved com-prehension of the issue and the research per-formed previously was to reduce dimension-ality and to contribute to the betterment in the field of cancer, Wisconsin-Madison Di-agnostic Breast cancer (WDBC) dataset was taken from learning repository of UCI data-base with 569 distinct instances for training by choosing finest features out of 32 differ-ent attributes. Different feature selection algorithms were used with data mining algo-rithms for better classification. Numerous enhancements in classification accuracy of WDBC were discovered by utilizing distinc-tive methodologies than the prior reviews directed in a similar field. The Logistic Re-gression, Linear Regression, and SVM algo-rithms showed better classification accuracy i.e. 98.24 %, 98.24 % and 98.07 % than the previous outcome results known for the said classification algorithms. The results were generated using 10 fold cross validation, by using different classification algorithms with feature selection and generation algorithms.

Keywords

SVM, Logistic Regression, Linear Regres-sion, Accuracy, Benign, Malignant.

URL

http://paper.ijcsns.org/07_book/201712/20171230.pdf