Decision tree model

About: Decision tree model is a research topic. Over the lifetime, 2256 publications have been published within this topic receiving 38142 citations.

GUI testing techniques based on Event Interactive Graph tree model

Abstract: Aimed at the GUI (graphical user interface) software testing the existence of the complexity and inefficiency issues, propose an Event Interactive Graph tree model. The structural events and system interactive events separate to establish component interactive graph tree and event interactive model, and gives a detailed GUI software modeling methods, and according to the model designed to test coverage criteria and test cases generation method. Finally applied to achieve good results in the actual system has greatly enhanced the quality and efficiency of GUI testing.

GQM-based Tree Model for Automatic Recommendation of Design Pattern Category

Abstract: Software Design Patterns (DP) are formal approaches that propose generic reusable solutions to different design problems. Building DP automatic recommendation system is one of the most challenging topics in the field of software industry to improve the final software quality. Proposing a DP for a design problem requires good base knowledge about each DP and its functionality. In this paper, we propose an approach that automatically recommends the appropriate design pattern category. The proposed approach is a Goal Question Metric (GQM) based tree model of questions. The software engineer answers these questions based on the user requirements, and finally the approach recommends the category of the suitable DP category based on our designed tree model. The GQM is responsible for weight calculation process at each node based on the questions' answers. The software engineer is responsible for delivering the user requirements to our system, via answering the proposed model. The precision and accuracy obtained by our system is 80\% while the recall is 100\%.

Design method of shoe tree model and related device

Abstract: The invention discloses a design method of a shoe tree model. The design method comprises the steps that three-dimensional surface data of the foot of a user is obtained; cutting-out processing is conducted on the three-dimensional surface data according to shoe tree design standards to obtain multiple cross section lines, and fitting processing is conducted on all the cross section lines to obtain three-dimensional curves; the three-dimensional curves are adjusted according to the shoe tree design standards to obtain a three-dimensional shoe tree sketch; curved surface modeling processing isconducted on the three-dimensional shoe tree sketch to obtain the shoe tree model. By processing the three-dimensional surface data of the foot to obtain the three-dimensional curves describing foot characteristics and then obtaining the three-dimensional sketch through processing to model the corresponding shoe tree model, the model can very well conform to the foot characteristics of a customer,the integrality requirement of a shoe tree can be also met by adjusting the curves, and the comfort of a follow-up made shoe sample is improved. The invention also discloses a design device of the shoe tree model, a computer and a computer readable storage medium. The design device has the above advantages.

Method and system for analyzing test cases for automatically generating optimized business models

Abstract: A method and a system for analyzing one or more test case trees for automatically generating an optimized test tree model are disclosed. The one or more test case trees are received and analyzed to identify a source tree and a target tree. Further, levels of each node of the source tree and the target tree are analyzed to identify a source node and a target node. The source node and the target node are compared to obtain a match. The source node is merged with the target node if the match is obtained. Else, a next source node and a next target node are identified based on one or more parameters. The identified nodes are subsequently compared and merged. The merged nodes are further optimized to generate the optimized tree model.

New decision tree based on genetic algorithm

Abstract: The decision tree based on the k-means algorithm has recently been proposed. However, the drawback of the k-means algorithm is that the users must determine the number of branches for each node before the decision tree is designed. The users are usually hard to determine the number of branches for each node. In this study, the new decision tree with variable-branches is proposed. The genetic algorithm is proposed to determine the number of branches for each node in the new decision tree. Thus, the proposed new decision tree approaches to near-optimization.