Abstract: Unsupervised learning is often derived from motivations that appear to be independent of supervised learning. While supervised models aim at prediction, unsupervised models are mainly used for grouping records or fields and for the detection of events or attributes that occur together. Predictive analytics that predict the present are based on existing data, preferably as much as possible.Predictive techniques for customer segmentation can be used for Intelligent decision towards customer preferences.In this paper unsupervised clustering based analysis towards very large datasets for analysis towards health insurance dataset for their preferences towards health insurance products for intelligent decision making.