Data mining supported hospital information systems solutions

Data mining is the process of obtaining meaningful information from large-scale datasets. It attempts to obtain meaningful results from these data by analyzing the data with the help of a variety of methods and techniques. Data mining is used in many areas such as machine learning, pattern recognition, statistics and medicine in order to analyze and interpret the information.The purpose of this study is to examine DBSCAN clustering algorithm which is a data mining technique and to allocate appropriate number of clusters by analysing mammography data with a software developed using this algorithm.In the study, mammography data were divided into optimal number of clusters with DBSCAN algorithm. The noisy data that not included in any cluster were added in the appropriate clusters with the K-NN classification algorithm. Thus, it is aimed to be analysing mammography data?s easier.In a conclusion, parameter values (Eps, MinPts) which give optimal result on MIAS database were determined by examining the values of these parameters. The specifications of clusters which created by this parameter value were defined. For noisy data, k parameter value that produces the best result for the K-NN, was detected by examining all k nearest-neighbors parameter values.