Abstract | ||
---|---|---|
This paper presents an efficient hybrid feature selection model based on Support Vector Machine (SVM) and Genetic Algorithm (GA) for large healthcare databases. Even though SVM and GA are robust computational paradigms, the combined iterative nature of a SVM-GA hybrid system makes runtime costs infeasible when using large databases. This paper utilizes hierarchical clustering to reduce dataset size and SVM training time, multi-resolution parameter search for efficient SVM model selection, and chromosome caching to avoid redundant fitness evaluations. This approach significantly reduces runtime and improves classification performance. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1145/1389095.1389362 | GECCO |
Keywords | Field | DocType |
large healthcare databases,svm training time,support vector machine,large databases,efficient svm-ga feature selection,efficient hybrid feature selection,genetic algorithm,efficient svm model selection,runtime cost,svm-ga hybrid system,classification performance,machine learning,data mining,genetic algorithms,model selection,feature selection,optimization,support vector machines,hybrid system,hierarchical clustering | Structured support vector machine,Hierarchical clustering,Data mining,Feature selection,Computer science,Support vector machine,Model selection,Artificial intelligence,Hybrid system,Genetic algorithm,Database,Machine learning | Conference |
Citations | PageRank | References |
5 | 0.46 | 13 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Rick Chow | 1 | 12 | 2.05 |
Wei Zhong | 2 | 5 | 0.46 |
Michael Blackmon | 3 | 5 | 0.46 |
Richard Stolz | 4 | 6 | 1.18 |
Marsha Dowell | 5 | 6 | 1.18 |