Font Size: a A A

The Practice Of Data-mining Technique In Criminals Database Of Prisons

Posted on:2007-03-18Degree:MasterType:Thesis
Country:ChinaCandidate:Y ZhaoFull Text:PDF
GTID:2178360185459879Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Data-mining becomes popular in computer research in recent years. It widely finds its way in every field. Data-mining is kind of technique applied to practice at the very beginning. It not only applies to consulting specific database, but also deals with the statistic data, analysis, integration and reasoning the given data so as to find the connection among the compositions, guide to seek for the solution to problems and even to predict the future on the basis of the data.The paper introduces some common methods used in data-mining's calculating, process and data pretreatment. It digs out the implied connection among criminals'data information after analyzing and dealing with all kinds of criminals'data information in a certain prison's database. And though the study on decision tree C4.5 algorithm, we utilize the algorithm set up a model of criminal shortening the term of imprisonment, and give some classify of factor of criminal shortening the term of imprisonment. These help to analyze group criminals and manage them as well. The main achievements are as follows:1. Analyzes the relationship between criminals'education and its term of imprisonment; points out the degree the education level affecting on the term of imprisonment and the term that is really carried out.2. Analyzes the relationship between criminals'age and the term of imprisonment; points out that the age can be the factor that influences the sentence and execution.3. Through the study on decision tree C4.5 algorithm , taking criminal's information database , we use C4.5 decision tree algorithm to generate a decision tree, and use post-pruning method to pruning the decision tree .And then according to the decision tree, we obtain the classification rules finally.
Keywords/Search Tags:Data Mining, Decision Tree, Classifier, Classification
PDF Full Text Request
Related items