Novel HGDBO: A Hybrid Genetic and Dung Beetle Optimization Algorithm for Microarray Gene Selection and Efficient Cancer Classification; [Nuevo HGDBO: Un Algoritmo Hrido de Optimizaci Genica y de Escarabajos Peloteros para la Selecci de Genes en Microrrays y la Clasificaci Eficiente del Ccer]
- Title
- Novel HGDBO: A Hybrid Genetic and Dung Beetle Optimization Algorithm for Microarray Gene Selection and Efficient Cancer Classification; [Nuevo HGDBO: Un Algoritmo Hrido de Optimizaci Genica y de Escarabajos Peloteros para la Selecci de Genes en Microrrays y la Clasificaci Eficiente del Ccer]
- Creator
- Alluri V.L.; Kanadam K.P.; Helen Josephine V.L.
- Description
- Introduction: ovarian cancer ranked as the seventh most common cancer and the eighth leading cause of cancer-related mortality among women globally. Early detection was crucial for improving survival rates, emphasizing the need for better screening techniques and increased awareness. Microarray gene data, containing numerous genes across multiple samples, presented both opportunities and challenges in understanding gene functions and disease pathways. This research focused on reducing feature selection time in large gene expression datasets by applying a hybrid bio-inspired method, HGDBO. The goal was to enhance classification accuracy by optimizing gene subsets for improved gene expression analysis. Method: the study introduced a novel hybrid feature selection method called HGDBO, which combined the Dung Beetle Optimization (DBO) algorithm with the Genetic Algorithm (GA) to improve microarray data analysis. The HGDBO method leveraged the exploratory strengths of DBO and the exploitative capabilities of GA to identify relevant genes for disease classification. Experiments conducted on multiple microarray datasets showed that the hybrid approach offered superior classification performance, stability, and computational efficiency compared to traditional methods. Ovarian cancer classification was performed using Nae Bayes (NB) and Random Forest (RF) algorithms. Results and Discussion: the Random Forest model outperformed the Nae Bayes model across all metrics, achieving higher accuracy (0,96 vs. 0,91), precision (0,95 vs. 0,91), recall (0,97 vs. 0,90), F1 score (0,95 vs. 0,91), and specificity (0,97 vs. 0,86). Conclusions: these results demonstrated the effectiveness of the HGDBO method and the Random Forest classifier in improving the analysis and classification of ovarian cancer using microarray gene data. 2024; Los autores.
- Source
- Data and Metadata, Vol-3
- Date
- 2024-01-01
- Publisher
- Editorial Salud, Ciencia y Tecnologia
- Subject
- Dung Beetle Optimization; Evolutionary Computation; Gene Feature Selection; Genetic Algorithm; Hybrid Algorithm; Microarray Data
- Coverage
- Alluri V.L., Acharya Nagarjuna University, Department of Computer Science and Engineering, AP, Guntur, India; Kanadam K.P., R.V.R & J.C. College of Engineering, Department of Computer Applications, AP, Guntur, India; Helen Josephine V.L., Christ University, Business Analytics, School of Business and Management, Karnataka, Bangalore, India
- Rights
- Restricted Access
- Relation
- ISSN: 29534917
- Format
- Online
- Language
- English
- Type
- Article
Collection
Citation
Alluri V.L.; Kanadam K.P.; Helen Josephine V.L., “Novel HGDBO: A Hybrid Genetic and Dung Beetle Optimization Algorithm for Microarray Gene Selection and Efficient Cancer Classification; [Nuevo HGDBO: Un Algoritmo Hrido de Optimizaci Genica y de Escarabajos Peloteros para la Selecci de Genes en Microrrays y la Clasificaci Eficiente del Ccer],” CHRIST (Deemed To Be University) Institutional Repository, accessed February 23, 2025, https://archives.christuniversity.in/items/show/13496.