Extracting hidden patterns in blood donor database using association rule mining

Data mining, in contrast to traditional data analysis, is discovery driven. Most of the recent data mining tools provide automatic pattern recognition and attempt to uncover patterns in data that are difficult to detect by traditional statistical methods. Most of the result patterns are not meaningful to the domain users at all. As a matter of fact, users have also to cope with the painful process to select patterns that are of actual interest from the result set. In this paper we are using the Apriori algorithm in association rules with the aim of helping experts in the blood organizations to find the best donor subjects within the whole population. We choose patterns according to rules that have some interest for the experts, or rules with values higher than a given threshold. For this purpose, we start from 7 input attributes such as: Age, sex, level of education, marital status, job, level of income and type of donor. On the basis of the results achieved, we are able to find interesting patterns from blood databases showing the healthiest donors within the whole population. Detecting the healthiest donors can guarantee a better quality of blood and blood products can prevent the environmental pollution and save the Blood Transfusion Organization the cost necessary to find the best population.