The integration of data from numerous, disparate sources makes the transformation of those data into actionable information extremely difficult. Applying analysis, synthesis, and predictive modeling techniques using human interaction is impractical, especially for converting data to information in situations which are rapidly changing and in which human life might be at risk, e.g. battlefield. Combining data which are sparse, varying in reliability, and with a time dependent factor to other data such as inventory, capability, and availability to make a decision falls into the category of complex decision support systems. Addressing this problem requires adaptive Big streaming data analytics that can: (a) model cyber physical infrastructures encompassing realistically complex warfighter scenarios, (b) ingest the massive data sets to capture large-scale dynamic systems complexity, and (c) process and update the analytics results in a timely manner to test contrasting mechanistic models and drive the next set of analyses. Currently, there is no adaptive big data analytics framework processing data streams in real-time and adopt to changes in cyber physical environments. The goal of this project is to develop an Adaptive Big Data Analytics Environment (ABDAE) that can adjust computations to respond promptly to rapid changes in data and cyber physical environments.
Benefit: The hallmark of the proposed ABDAE is its ability to dynamically adjust computations to respond efficiently to the data received from physical systems, environmental sensors and effectors. This capability will enable a unique tactical decision support tool that can be deployed to cyber battle management to increase the effectiveness of weapon selection, provide greater battlefield awareness, speed up the order of battle, extend actionable information to the tactical edge, and reduce exposure of friendly forces to unneeded risk. The outcomes of this phase will be: (1) a proof-of-concept prototype system that implements active change detection and passive learning models for non-stationary environments; (2) a framework for hybrid/passive learning methods for learning in high-volume data streams; (3) provide a recommendation system for cybersecurity applications that face non-stationary streaming environments; and (4) provide a benchmark using synthetic and real-world data sets along with statistics that measure the overall efficacy of the proposed approaches.
Keywords: bug data stream, bug data stream, Life-Long Learning, Big Data, ADAPTIVE LEARNING, cyber-security, Classification
