Evolutionarily-tuned support vector machines.

Support vector machine (SVM) classifiers can cope with many different classification tasks but improperly selected hyperparameters may deteriorate their performance. Moreover, datasets are getting bigger in terms of their size and the number of features. This is often coupled with low training data quality and presence of redundant features, which can adversely affect classification accuracy and time performance. Furthermore, high memory and computational complexity of SVM training can be a limiting factor for its application over huge datasets. We address these issues with evolutionarily-tuned SVM, where we utilize evolutionary algorithms for optimizing hyperparameters, along with selecting features and training instances. The performance of our method is compared on several benchmark datasets to other methods for optimizing SVMs, as well as to other classifiers. The results show that our algorithm gives high performance in both accuracy and classification time when compared with the state-of-the-art methods for SVM optimization.