Automated Classification of Cancer using Heuristic Class Topper Optimization based Naïve Bayes Classifier

Cancer is a complicated illness that is caused by numerous gene mutations or deregulation of gene interactions. This study, on the other hand, proposes a unique method for cancer categorization. Cancer is the major reason of death in this era. Appropriate methods are required to diagnose it as early as possible so that accurate treatment should be started to save human lives. Heuristic Class Topper Optimization describes a vital role in the detection of cancer for classification. A large dataset of tumors has been taken and Naive Bayes classifier exported to categorize them. The Heuristic Class Topper Optimization method (HCTO) is planted to extract the features. The optimization technique is based on the intelligent learning of pupils in a classroom. Weak students are learning from the class topper. There are various class toppers based upon the number of sections. Thus, the characteristics are refined with the aid of HCTO. The HCTO algorithm is a novel artificial intelligence technology that is rapidly converging. The HCTO-NB technique is simple, less complex, accurate, and has a low error rate, all of which are important characteristics in cancer categorization. The recommended method’s achieved parameters are accuracy of 97.6%, precision of 98.4, error rate decreased by 3% on 1000 iterations, and classification efficacy are all demonstrated. The results are also compared to the KNN classifier, which has been used to classify cancer by a number of studies in the past. Experiments on a range of datasets demonstrated that this novel method was more accurate and dependable by ~ 14% compared to KNN. The findings shows that the suggested approach is both rapid and accurate, making it a great alternative for cancer diagnosis in the real world. In this paper 4 types of cancers datasets with relevant features like age, Gender, Tumor size, Tumor area and smoker or non-smoker etc. have been used for real time validation.