Multi-Attribute Density Estimation Based Location Selection Approach in Multi-Agent Disease Prediction Model for Decision Support System Using Diagnosis Pattern and Data Mining
M. Inbavalli, G. Tholkappia Arasu, Perumal Manimekalai · 2016
2 Abstract: The presence of decision support systems plays a vital role in many situations like business intelligence and medical solutions. There are many designs has been proposed earlier to support decision making and suffers with the problem of accuracy and time complexity. We propose a novel approach which uses Multi-Attribute Density Estimation technique (MADE) to choose the set of locations from where the data can be retrieved. The method uses various meta data which represents the availability of data in different locations of the network and based on the meta data, the method computes the MADE measure to choose most optimal locations. From identified locations, the method generates diagnosis patterns which contain various information about the medical history available in the location. The number of agent generation is performed according to the MADE factor and based on the patterns generated, the DSF (Decisive Support Factor) is computed which shows the possibility of the disease. The proposed method reduces the time complexity and improves the accuracy of decision making system.