A comparison of Boosting techniques for Classification of Microarray data

Abstract

The advancements in technology, particularly microarrays, have played a pivotal role in enhancing crucial aspects within the domains of genomics and bioinformatics. These advancements have significantly contributed to the enhancement of illness diagnosis, evaluation of therapy response in patients, and advancements in cancer research. A single tissue sample obtained from a microarray has the potential to encompass a substantial number of distinct gene expressions, reaching into the tens of thousands. Consequently, the process of scrutinizing the data to identify significant patterns can be quite daunting when employing conventional statistical approaches. Microarray data often exhibits a substantial likelihood of encompassing extraneous and duplicative factors, hence introducing noise into the dataset. Numerous studies are currently being conducted to enhance the analysis of microarray data, with the aim of enhancing performance and prediction accuracy at an accelerated pace.  Boosting algorithms were utilized in this study to classify microarray data. The microarray data underwent pre-processing, during which significant features carrying essential information were retrieved. These features were then inputted into the AdaBoost, Gradient Boost, and XGBoost algorithms. The experimental findings indicate that XGBoost demonstrates superior performance compared to other boosting approaches, with an accuracy rate of 98.18%.