Ant lion and ant colony optimization integrated ensemble machine learning model for effective cancer diagnosis

Abstract

Statistics from reputable sources, including the World Health Organization (WHO), demonstrate that cancer is a leading cause of death globally, accounting for millions of deaths each year. When it comes to the early identification of cancer, machine learning (ML) is crucial. To analyze complex data and identify minute patterns that may indicate the presence of cancer, it employs robust computational approaches. Improving patient outcomes relies on early cancer detection since it paves the way for faster treatment and intervention, which might lead to better prognoses and higher survival rates. To choose features, this study intends to build an ML-based ensemble model utilizing ant colony optimization (ACO) and ant lion optimization (ALO). Next, ML classifiers are used as the initial predictions' basis learners. The last forecast is the result of combining two ensemble methods: voting and averaging classifiers. Four distinct cancer microarray datasets are used to assess the approach. With an accuracy of 99.08% on the Lung cancer dataset, the voting ensemble classifier outperforms the others, according to the empirical analysis.