Context-aware Machine Learning Middleware for Real-time Distributed Streaming Big Data Analysis (COMAND)
Department of Energy awards InfoBeyond million dollars on the R&D of COMAND for big data streaming. Emerging applications and instrumental systems produce massive streaming data every day. It is essential that the big data streaming to be quickly and accurately analyzed to extract useful information for timely decision making. Despite substantial works have been proposed for streaming data processing, e.g., SAMOA, Jubatus, etc., many shortcomings have not well addressed. InfoBeyond advocates COMAND for massive streaming data analysis. COMAND is a real-time and context-aware ML (Machine Learning) system that has the ability to adapt ML algorithms to the time-varying data streams, which is the key difference than the current approaches. Furthermore, COMAND is developed as a middleware architecture that enables efficient and scalable capabilities in distributed environments.
The COMAND system addresses the challenges of massive streaming data analysis. It enables efficient and accurate massive streaming data analysis through collaborative machine learning technology for energy, security, scientific, and business big data in a real-time fashion.
Key words: Distributed big data processing, massive streaming data analysis, machine learning based data analysis, large-scale data processing and dissemination.