A PROCESS BASED ON DATA MINING TOOLS FOR HEALTHCARE PATTERN ANALYSIS

Public healthcare (PHC) is usually provided by the government through National Health Care (NHC) systems. In India, the Indian components and territories administer a universal healthcare system. A conceptual framework has been established for the systematic use of data mining technologies to public healthcare management concerns because of the research. The study only includes public healthcare institutions in India, specifically in the states of Jammu and Kashmir, Punjab, and Haryana. The research goal of this mining task is to discover patient’s medical behavior patterns for event detection. Patient’s medical behavior is best captured by the patient’s medical visits ordered by Doctor’s. The time-orientated medical visiting sequence can be used to draw a multi-dimensional profile, which preserves the history of the medical treatments and indicates every medical behavior change over the time. By analyzing the medical profiles rather than a snapshot of the records, patient’s medical behaviors are understood. Feature selection has also been discussed with respects to what is available in the source data and what is expected as data mining results. Setting up profile has prepared the experimental dataset for advanced mining steps and more importantly assisted us in the mining method development. Iterative training process results in robust models (HMMs) and at the same time reduces the expense of training HMM in a data mining application.