A SURVEY ON RESOURCE OPTIMIZATION AND PREDICTION TECHNIQUES WITH BIG HEALTHCARE DATA

Big data (BD) analytics is utilized to gathers and analyzes large volume of data to find useful information. BD is utilized to mine valuable information for predictive analytics. Healthcare data classification classifies the patient details into many columns based on the user requirement. Resource optimizations are collection of processes to match with the accessible resources to attain the goal. Prediction examines the present and historical events to forecast about the future events. In recent times, many research works are carried out for improving the performance of classification and prediction process with minimal resource utilization. However, the prediction time and resource utilization remained challenging issue in healthcare applications. The key objective of the paper is to present comparative literature survey of existing resource optimization and prediction techniques. The contribution of the survey is two-fold. In first one, a brief description of storage techniques, resource optimization and prediction techniques like NoSQL database (DB), data management methodology (DMM), two-stage scheduling policy, data parallelism (DP), external scheduler (ES), internal scheduler (IS), Fuzzy Linguistic Summarization (LS) approach and thus highlights the significant features, merits, and shortcomings. In second one, one of the existing schemes was selected to test whether it is adequate for real systems. Consequently, our theoretical analysis is compared with experimental results to look forward the results and to afford valuable insight to future researchers