Research And Implementation Of Birth Defect System Based On Decision Tree

Birth Defect is an important problem faced by countries all over the world. Theoccurrence of Birth Defect makes our society and families under more pressure and had areverse influence on population development. China, one of the countries the occurrence rateof Birth Defect is the most highest in the world, is facing a growing statement of theoccurrence of Birth Defect。The traditional research to control the non-infectious disease, suchas Birth Defect , was used to be the Linear Reduction method. However, from the point ofview in nowadays, this method is with some considerable restrictions. For the restrictions ofthe traditional research, we try to use Data Mining method to find the pathogenesis of BirthDefect。Data Mining is a advance international research subject of academe circle, it hassyncretized knowledge of database, artificial intelligence machinery study and statisticstheoretical knowledge. Data Mining is a course to draw the implicit, unknown information orpattern from large scale database or data warehouse which has latent value for application. Itis a new field in database research with highly applicable value, has great guidance functionto draw the decision for people in the aspects such as economic science and technology。Classification is one important branch of Data Mining. Classification has been successfullyapplied to wide range of application areas, such as medical diagnosis, weather prediction,credit approval, customer segmentation and fraud detection.Many different techniques have been proposed for classification, such as decision tree,bayes net, genetic Algorithms, Decision tree classifiers have found the widest applicability inlarge-scale data mining environments. The decision tree model, which can directly manifestthe characteristic of the data besides to be easily understood, is the most frequently adopted inthe data mining. Moreover, the decision tree model, owning the ability of classification andprediction, can draw the decision rule conveniently.ID3 algorithm and C4.5 algorithm is twomost famous Decision Tree classification algorithm. They are constructed by Quilan forinducing classification model from data. Carrying on all the advantages of ID3 algorithm,C4.5 algorithm is a kind of inducing study algorithm. Firstly, it choose a part of case toconstruct decision tree. Then, it test and restructure decision tree by the last case.This paper first introduced the concept of date mining , data mining technology and datamining algorithm that used frequently. Secondly, the paper discussed classification and thetechniques of classification. Then, base on these, give a systemic and deeply analysis ofdecision tree classifiers. By using C4.5 algorithm, we learn the knowledge of illness regularityand rules from Diabetes Mellitus data, construct decision system, and generate a set of rulesof Birth Defect data, and generate a set of rules of Birth Defect diagnostics and predictiondepending on the preprocessed Birth Defect data.