Multinomial Naive Bayes Categorization for Semantic Web Services

Abstract

The significant existence of web services is challenging to the researchers regarding their diversity of types and their diffusion. It may lead to difficulty in identifying the relevant service during the discovery or composition process. To tackle this problem, we propose a new method to categorize semantic web services based on the Naive Bayes algorithm using a weighting method (TF-IDF), which binds a service according to its description importance offered by the service provider to be categorized in a relevant class. It enhances the performance by proposing a compatible combination of the preprocessing techniques (Natural language processing) to achieve a better classification result. This method has been tested on the OWLS-TC dataset, categorized into seven classes, and its accuracy is 93%.