Web service reliability and scalability determination using optimized depth wise separable convolutional neural network
Gokulakrishnan Dhakshnamoorthy,Jeyabal Sridhar,Ramakrishnan Ramanathan,Muruga Radha Devi Dharmalingam
DOI: https://doi.org/10.1002/qre.3530
2024-03-17
Quality and Reliability Engineering International
Abstract:Web service composition (WSC), a distributed architecture, creates new services atop existing ones. Ensuring trust and assessing performance and dependability in online services coordination is essential. In this paper, "Web Service Reliability and Scalability Determination Using Depth Wise Separable Convolutional Neural Network" (WSRS‐DWSCNN) is proposed to assess the trustworthiness of online service compositions, particularly focusing on performance and dependability. This work addresses the need to predict the reliability and scalability of Business Process Execution Language (BPEL) composite web services. The proposed approach transforms the BPEL specification into a Depth Wise Separable Convolutional Neural Network (DWSCNN) and annotates it with probabilistic properties for prediction. The DWSCNN model classifies the outcomes as correct or incorrect, and to enhances the prediction of web service composition scalability and reliability, we optimize the DWSCNN's weight parameters using the Adolescent Identity Search Algorithm (AISA). The proposed technique is activated in Python and its efficacy is analyzed under some metrics, such as reliability, scalability, accuracy, sensitivity, specificity, precision, F‐measure. The proposed method provides 12.36%, 45.39%, and 25.97% better reliability, 41.39%, 11.39%, 34.16% better accuracy compared with existing methods like, Web service reliability prediction depending on machine learning (WSRS‐K‐means), reliability prediction method for multiple state cloud/edge‐basis network utilizing deep neural network (WSRS‐DNN‐BO), and improving reliability of mobile social cloud computing utilizing machine learning in content addressable network (WSRS‐CAN), respectively.
engineering, industrial, multidisciplinary,operations research & management science