A novel framework for classification of syncope disease using K-means clustering algorithm

For valuable decision making, the extraction of information enriched data from a collection of large and unstructured data is of high significance. Data mining can be used to extract hidden knowledge from a data set. Mining unstructured attributes is renowned technique for predicting the potential causes of diseases. However, it is complex process to develop prediction mechanism for diseases those comprise characteristics like dataset-unavailability and lengthy diagnoses procedures. Syncope is classified as one of such disease. This paper presents novel framework for predicting possible causes of syncope disease. The validation of framework is performed through real case study. Data set used in this research is obtained from Armed forces institute of cardiology and National institute of heart disease (AFIC & NIHD), Rawalpindi, Pakistan. K-means clustering algorithm is used for classification of dataset. Results are compared by applying K-means fast, K medoids and X-means algorithms. Empirical results prove that proposed framework improve the predication accuracy through novel clustering approach for possible syncope causes.