A Privacy Preserving Improvised Approach for QOS Aware Web Service Recommendation

• Published in: 2018 9th International Conference on Computing, Communication and Networking Technologies (ICCCNT) pp. 1 - 7
• Main Authors: Khan, Arif Ahmed, Chakkarwar, V. A.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.07.2018
• Subjects: Collaboration; Collaborative filtering; Data models; Filtering; K-means ICRS clustering; Predictive models; privacy preservation; QoS; QoS prediction; QoS values; Quality of service; recommendation; Reliability; Vectors; Web Service; Web service recommendation; Web services
• DOI: 10.1109/ICCCNT.2018.8493866

Suggestions for web services along with the recommendation have been very popular lately in IT research. When creating composite quality-of-service sequences based on service-oriented systems, it is imperative to evaluate the non-performance characteristics of potential candidates for service selection. In this article, we will present Improvised clustering recommendation System (ICRS), a model for analyzing and predicting the absolute reliability of the predicted atomic web service that can also evaluate the reliability of a continuous service call based on data collected from previous invocations with a purpose. To improve the accuracy of the most advanced forecast models, we include user-specific parameters and user-specific contexts. To reduce the flexibility problems presented in a modern way, we will collect the above usage data using the K-means and ICRS grouping algorithm. When evaluating the different qualities of our models, we experimented with Planet DB Dataset and the services registered in WSDream dataset. The results confirm that our models can be guessed and scaled accurately.