Robust Approach for Estimating Probabilities in Naive-Bayes Classifier

摘要: Naive-Bayes classifier is a popular technique of classification in machine learning. Improving the accuracy naive-Bayes will be significant as it has great importance using numerical attributes. For numeric attributes, conditional probabilities are either modeled by some continuous probability distribution over range that attribute's values or conversion attribute to discrete one discretization. The limitation discretization does not classify those instances for which any value every class zero. proposed method resolves this estimating and improve noisy data. efficient robust classifier. been tested number databases UCI learning repository comparative results existing also illustrated.