High performance social data computing with development of intelligent topic models for healthcare

Data mining and big data computing are the emerging domains in the current era of predictions for societal applications. Millions of people are interested in sharing their views through tweets. Healthcare predictions are one of the attractive researches in big data social mining. Healthcare predictions are derived by implementing topic models by the ailments data. An ailment refers to either illness or sign of a particular health problem. Millions of tweets are collected based on conditions and assessed with ailment topic aspect models. The existing topic model, Latent Dirichlet Allocation (LDA), Latent Semantic Indexing, Probabilistic LSI (PLSI), limits the healthcare results assessment concerning any one of the ailments aspects. Recent ailments topic aspect model (ATAM) overcome the problems of these topic models and delivers the healthcare assessment results concerning the fundamental aspects of ailments data except side-effects analysis of treatments. The scalability performance of ATAM is degraded in showing healthcare results over the massive amounts of health data. A high-performance computing model of ATAM has been developed in the distributed environment to address scalability. Its intel-ligent model is designed in the cloud and multi-node Hadoop environment to deliver high-performance social computing results for healthcare. Experiments are conducted on many comparative studies is demonstrated between the existing and proposed high-performance models using the massive amount of health-related tweets concerning the ailments aspects.