Bayesian Kernel Methods for Non-Gaussian Distributions: Binary and Multi-class Classification Problems.

摘要

Recent advances in data mining have integrated kernel functions with Bayesian probabilistic analysis of Gaussian distributions. These machine learning approaches can incorporate prior information with new data to calculate probabilistic rather than deterministic values for unknown parameters. This paper analyzes extensively a specific Bayesian kernel model that uses a kernel function to calculate a posterior beta distribution that is conjugate to the prior beta distribution. Numerical testing of the beta kernel model on several benchmark data sets reveal that this model's accuracy is comparable with those of the support vector machine and relevance vector machine, and the model runs more quickly than the other algorithms. When one class occurs much more frequently than the other class, the beta kernel model often outperforms other strategies to handle imbalanced data sets. If data arrive sequentially over time, the beta kernel model easily and quickly updates the probability distribution, and this model is more accurate than an incremental support vector machine algorithm for online learning when fewer than 50 data points are available.