Xubin Wang's Site

Accurate screening of cancer types is a critical process in cancer detection, which can help cancer patients by providing them with a precise treatment. Unfortunately, only a few characteristic biomarker genes from gene expression profiles are closely associated with tumors. Therefore, it is relevant and essential to select biomarker genes from these profiles for screening in cancer. Recent nature-inspired computational methods have been proposed for the selection of relevant biomarker genes across multiple types of cancers. However, we note some limitations; such as low efficiency and poor generalization. To address these challenges, this paper proposes a framework called EODE, which incorporates the grey wolf optimizer (GWO) to optimize feature subsets and then creates an optimized ensemble classifier in a collaborative manner. We demonstrate the performance of our proposed model using 35 different cancer gene expression datasets. The experimental results reveal that our model has strong robustness compared with other benchmark methods for the identification of important biomarker genes across different cancer types. Furthermore, we investigate the general applicability of EODE on a pan-cancer dataset containing 33 different cancer types from the The Cancer Genome Atlas (TCGA) PanCanAtlas project. EODE achieved an accuracy of 0.8654 and performed better than the other computational models.

Recently, many evolutionary computation methods have been developed to solve the feature selection problem. However, the studies focused mainly on small-scale issues, resulting in stagnation issues in local optima and numerical instability when dealing with large-scale feature selection dilemmas. To address these challenges, this paper proposes a novel weighted differential evolution algorithm based on self-adaptive mechanism, named SaWDE, to solve large-scale feature selection. First, a multi-population mechanism is adopted to enhance the diversity of the population. Then, we propose a new self-adaptive mechanism that selects several strategies from a strategy pool to capture the diverse characteristics of the datasets from the historical information. Finally, a weighted model is designed to identify the important features, which enables our model to generate the most suitable feature-selection solution. We demonstrate the effectiveness of our algorithm on twelve large-scale datasets. The performance of SaWDE is superior compared to six non-EC algorithms and six other EC algorithms, on both training and test datasets and on subset size, indicating that our algorithm is a favorable tool to solve the large-scale feature selection problem. Moreover, we have experimented SaWDE with six EC algorithms on twelve higher-dimensional data, which demonstrates that SaWDE is more robust and efficient compared to those state-of-the-art methods.