A System for Parallel Data Mining Service on Cloud

摘要

We present a cloud-based data mining platform which demonstrates the solution of data mining as a service (DMaaS). In the backend, the data processing engine is based on hadoop, an open-source implementation of Google MapReduce. Implementation of the data mining algorithms in Apache Mahout is deployed in the platform. The user can access DMaaS from his browser for analyzing general purpose data mining problems. In this paper, we give an overview of DMaaS, present the system architecture and implementation techniques, and elaborate on a demonstration scenario.