A Flexible and Hybrid Feature Mapping Technique Based on multivariate Beta Distribution Kernel Applied to Medical Applications
Narges Manouchehri, Nizar Bouguila · 2022 IEEE International Conference on Industrial Technology (ICIT) · 2022
In this work, we propose two novel feature mapping techniques based on multivariate Beta distribution and its mixture models. Support vector machine (SVM) is one of the most famous and powerful discriminative classifiers which has been applied in various domains. We improved its discrimination power by considering the nature of data and integrating discriminative approach with generative method. Such a novel hybrid method could improve the accuracy of the model compared to SVM with traditional kernels. To evaluate our model performance, we applied it to medical applications including Barrett’s oesophagus detection and colonoscopy image analysis. The outputs indicate that our proposed model could be considered as a promising alternative.