Business, scientific and engineering experiments, medical studies, and governments generate huge amount of data. The problem is how to manage these large volumes of data, efficiently process them and extract hidden information and knowledge from all these data. Data mining (and other data analytics methods) provides means for at least a partial solution to this problem. However, it would be much too expensive to all these areas of human activity and companies to develop their own data mining solutions, develop software and deploy it on their private infrastructure.

With the CloudMiner project we offer a cloud of data mining services (Software as a Service) that run on a cloud service provider infrastructure. We design the architecture of the CloudMiner and its main components: MiningCloud that contains all published data mining services, BrokerCloud which mining service providers publish services to, DataCloud that contains the collected information, and Access Point. The Access Point allows users to access the Service Broker to discover mining services and supports mining service selection and their invocation.