A Sequential Process of Clustering and Classification is Employed to Predict the Occurrence of Lung Cancer

In order to effectively identify severe illnesses such as lung cancer, a robust big data tool is needed to assess the enormous amount of data in the medical field and extract meaningful insights. This work offers a thorough evaluation of several different supervised and unsupervised learning techniques found in literature. The study demonstrates that the foggy k-means technique can effectively cluster the real-time lung cancer dataset despite several obvious challenges, including noise, outliers, inliers, missing values, and multidimensionality. The research proposes a cascade of C4.5 classification on clustered data using the foggy k-means approach to improve the results of the cluster validity indices given in the literature. It also advises using a supervised learning method to estimate the lung cancer data set more accurately.