Abstract :
EHR Electronic Health Records collects data on yearly basis and it is used in many countries for healthcare.HER Health Examination Records collects the data on regular basis and identifies the participants at risk that is important for early warning and prevention.the fundamental challenge is for learning classification model for risk prediction with unlabelled data and live data string that established the majority of the collected dataset.the unlabelled data string describes the participants in health examintions whose health conditions can be vary from healthy to highly risky or very ill.in this paper, we propose a graph based,semisupervised learning algorithm called SHG health semi supervised heterogenous graph on Health for risk prediction and assessment to classify a progressively developing condition with the majority of the data unlabelled. An efficient iterative algorithm is designed and developed to proof the convergence is given.extensive experiments based on both real health examination dataset and live datasets to show effectiveness of our method. Jayashri A. Sonawane | Dr. Swati A. Bhavsar "Mining Health Examination Records - A Graph Based Approach" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-3 | Issue-3 , April 2019, URL: https://www.ijtsrd.com/papers/ijtsrd22810.pdf Paper URL: https://www.ijtsrd.com/engineering/computer-engineering/22810/mining-health-examination-records---a-graph-based-approach/jayashri-a-sonawane
Keyword :
livedatastring, heteroHER, classifier