Data Mining Technology Exploration Applications, The Relationship Of Weight And Eating Habits

With computer technology, especially the rapid development and wide application of database technology, all walks of life accumulated huge amount of data, the traditional data-processing method has been very difficult to be fully utilized in order to find useful information in these mass data. Therefore, data mining technology came into being. Since the birth date, data mining technology has been given more and more attention by various industries, but in domestic and international epidemiological studies, statistical methods has always been the mainstream, but the statistical methods still needs a certain degree of requirements, such as researchers need to have some general understanding of the data relationship, assuming that there is a certain relationship between attributes and then to collect data according to the relationship, establish statistical models to test hypotheses, statistical models can not to study large amounts of data, if there are more variables, statistical model also has more difficulties. Data mining technology is widely used in financial, retail, telecommunications and so on, but in the medical field is still in the exploratory phase, this article attempts to take data mining technology as a complementary research method, according to analysis the weight and eating habits to explore the relationship, with using association rules algorithm to find out which food products and frequency that different body type of people often eat, helping people understand the relationship between daily diet intake and body weight, using classification algorithms to establish classification rules, establish forecasting model, predict obesity by entering personal eating habits, serve as an early warning so that timely adjust eating habits, in order to obesity prevention and reduction of the obesity production, indirect reduction of cardio-cerebral blood vessel diseases. In addition, as the same time to predict obesity, comparing the two algorithm decision trees and Bayesian classification, sum up the advantages and disadvantages of the two algorithms.