Indian Journal of Science and Technology
Year: 2016, Volume: 9, Issue: 38, Pages: 1-9
Amandeep Kaur1 and Rachna Soni2
1 Faculty of Computer Applications, Chitkara University, Rajpura - 140401, Punjab, India; [email protected]
2 Faculty of Computer Applications, DAV College, Yamuna Nagar - 135001, Haryana, India; [email protected]
Background/Objectives: The web client gets easily lost in the web’s rich hyper structure as the utilization of web is expanding more step by step. The primary point of proprietor of the website is to give the important data in terms of satisfactory QoS (Quality of Service) factors like throughput, response time, accuracy and content availability. From the client point of view, Web Service based QoS Discovery is a multi-criteria decision mechanism that requires knowledge about the service and its QoS description. These clients are not experienced enough to acquire the best selection of web service and trust the QoS information published by the provider. Methods/Analysis: The existing t Model was used with XML based SOAP protocol in order to solve the problem of UDDI registry which holds QoS description. Findings: The new discovery approach is expected to be the solution for contemporary web service discovery problems. A comparative performance analysis of prominent page rank algorithms was made on the basis of metrics like throughput, response time, recall rate and precision rate etc. Simulation Interface has been designed for classification algorithms. The program is developed for the Fuzzy Logic, Naïve Bayes, Neural Network, Linear Discriminant Analysis and Support Vector Machine using MATLAB application Improvements: The experiment revealed the fact that recall and precision rate are the best to predict the Quality of Service (QoS) supported by various E-Commerce web sites like Amazon, Jabong and Shop Clues etc. Detailed performance analysis further concluded that Neural Network could be the best algorithm to rate the service quality of E-Commerce websites.
Keywords: Fuzzy Classification, Linear Discriminant Analysis, Neural Network, Support Vector, Universal Discovery Description and Integration (UDDI)
Subscribe now for latest articles and news.