Optimizing Cancer Forecasting with Cutting-Edge Data Mining Techniques

The needs of the machine learning domain are especially pressing in the medical field. The information as in the medical area is quite complicated, and in order to arrive at an accurate diagnosis, it is necessary to analyse many and diverse forms of data, making these needs all the more important. Because of the high stakes involved, areas like cancer detection need the highest levels of accuracy and dependability in the method of detection. The location of a tumour determines the kind of cancer that has developed. As a result, there is a great deal of variation in the parameters of different cancer types. The lung cancer values for prediction are established by the first contribution. The multi-model framework that this work provides may be utilized to effectively forecast lung cancer. The model's clustering techniques are a hybrid of Decision Tree and K-Means. The prediction process is enhanced by the varying degrees of integration between the two models. An efficient pre-treatment method is included into the model to offer a data purification The system with the basic data being acquired from real-time. Predicting how long the patient will survive levels is done using the decision tree framework, which is a form of regression. The accuracy analysis among the three methods indicates that the suggested method Decision Tree and K-Means based survival prediction (DKSP) markedly surpasses both the Decision Tree (DT) and K-Means techniques. The Decision Tree technique exhibits the lowest level of precision, although K-Means shows an improvement; nonetheless, it still fails to equal the effectiveness of the proposed approach. The DKSP model, achieving an accuracy of 99.23, exhibits an important enhancement, establishing it as the most proficient tool for cancer prediction among those evaluated.