Abstract: The study focuses on Hadoop-based approach for web service management. Due to small amount of web service repository, performance of traditional web service discovery approaches does not offers relevant services based on user given query. To carry all those complications and also focuses on discovering potentially value added web services define architecture as WSDH (Web Service Discovery using Hadoop). Hadoop is a framework which is designed to support processing of large data sets in distributed computing environment. We propose a service discovery approach which retrieves relevant web services based on the user query. In which, we first perform scheduling. In scheduling, schedule the user given query based on time and then logical query execution tree is constructed. After that, preprocessing the relevant data which are maintained in hadoop frame work. To offer a deeper understanding of time series data we use deep learning technique. In deep learning the relevant data are maintained in Hbase-OpenTSDB. So efficiently merge time series data into open TSDB. We assess TF-IDF to calculate matching using Map/Reducer frame and similarity is calculated using vector space model. Comparing with the existing system proves that the proposed system provide an optimal solution for user query with minimal search time and maximum accuracy.
M. Akila Rani and D. Shanthi, 2016. Web Mining for Potentially Value Added Services. Asian Journal of Information Technology, 15: 2908-2926.