A MMD-Based Non-Parametric Online Anomaly Detection Algorithm over Big Data Streams in Cloud Collaborative Environments

Big Data and Cloud Computing are complementary technological paradigms with a core focus on scalability, agility, and on-demand availability. The rise of cloud computing and cloud data stores have been a precursor and facilitator to the emergence of big data. As a result, a number of enterprises are building efficient and agile cloud environments, and cloud providers continue to expand service offerings. One of the major security challenge in cloud collaboration systems are detection of anomalous data patterns that reflect malicious intrusions into cloud data storage systems. This problem typically involves design of statistical tests to identify data variations. The main scope of this paper is to exploit information theoretic and statistical techniques to address the above security issue in order to provide information theoretically provable security (i.e., anomaly detection with vanishing probability of error) based on Maximum Mean Discrepancy (MMD) that measures the distance between mean embedding of distributions into a Reproducing Kernel Hilbert Space (RKHS). To the best of our knowledge, the detection of anomalous access requests in cloud-based collaborations through non-parametric statistical technique has not been studied in earlier works. This paper proposes an online anomaly detection algorithm based on MMD technique to detect anomalous access requests in cloud environment at runtime.