Recently, decision tree algorithms have been widely used in dealing with data mining problems to find out valuable rules and patterns. However, scalability, accuracy and efficiency are significant concerns regarding how to effectively deal with large and complex data sets in the implementation. In this paper, we propose an innovative machine learning approach (we call our approach GAIT), combining genetic algorithm, statistical sampling, and decision tree, to develop intelligent decision trees that can alleviate some of these problems. We design our computational experiments and run GAIT on three different data sets (namely Socio-Olympic data, Westinghouse data, and FAA data) to test its performance against standard decision tree algorithm, neural network classifier, and statistical discriminant technique, respectively. The computational results show that our approach outperforms standard decision tree algorithm profoundly at lower sampling levels, and achieves significantly better results with less effort than both neural network and discriminant classifiers.

Using genetic algorithms-based approach for better decision trees: A computational study

A briefing on data mining technology used in criminal investigation work and the importance of using ID3 decision tree to structure the Decision Tree algorithm method is given by this paper. It makes a combination of criminal cases criminal suspects training data sets, using its decision tree analysis of the classification and uses Microsoft SQL Server 2005 Office 2007 Add to the data mining of Visio 2007 graphics. It shows and shares the form of mining model.

Data Mining in the Application of Criminal Cases Based on Decision Tree

In order to compare the differences of machine classification in the application to medical data, the experimental study is carried on using the diabetes data collected from HIS system. In this paper, analysis and assessment with various classification algorithms such as Decision Tree Algorithm, Naive Bayesian and Neural Network Algorithm have been applied to explore the advantages and disadvantages of each machine classification algorithm, and optimize them. The results showed that the algorithm of Naive Nayes and Neural Network received high accuracy and the diagnosis performance was good.

The Construction and Comparative Optimization of Classification Prediction Model in Diabetic Cases

Data Mining in the Application of Criminal Cases Based on Decision Tree - TI Journals

In the modern age of computing, sparse and irregularity in a sample of data is needed for reconstruction of the large and different types of dimensions of data. However, there is a challenge to analyze this type of data. One issue is the robust selection of data. There are many analytical tools are available but crucial part is the decomposition, and make the clusters as well as gradient estimated of data. When one extracts the attributes of sparsely sample data then most likely common attributes may lead to inaccurate results. Due to these, the present paper consists of the solutions through entropy calculation and GINI Index calculation of an insurance company. After calculation of entropy and GINI Index, a decision tree of sample data of an insurance company is presented.

/pdf/construction-of-decision-tree-for-insurance-policy-system-3ahdyjlc9u.pdf

Construction of Decision Tree for Insurance Policy System through Entropy and Gini Index

Decision tree algorithm is a very active research area of data mining. This paper describes the basic decision tree idea in data mining, then discusses the computational complexity of the classical decision tree algorithm (ID3 algorithm). And the improved algorithm to construct a decision tree by using statistical theory and ideas of conditional probability is proposed in this paper. Experiments show that the computational complexity of this decision tree algorithm is superior to the traditional algorithm, and its efficiency is greatly improved.

Research and application of conditional probability decision tree algorithm in data mining

Contrary to TSP with clustering features, clustering processing ant colony algorithm (CPACA) is studied in this paper. CPACA first clusters cities in TSP, the TSP problem is decomposed into many small-scale sub-problems (the number of sub-problems equals to the clustering number of cities), then each sub-problem is solved using the ant colony algorithm in parallel, and solutions for all sub-problems are to be merged into the solution to solve the problem according to certain rules. As the problem is decomposed using the clustering characteristics of the problem itself, to solve each sub-problem in parallel, then to speed up the solving speed greatly.

XianMin Wei

Papers

Research and application of conditional probability decision tree algorithm in data mining

Clustering processing ant colony algorithm