Font Size: a A A

Research Andachievement Of License Plate Recognition System Based On Neural Network

Posted on:2011-01-06Degree:MasterType:Thesis
Country:ChinaCandidate:L WangFull Text:PDF
GTID:2198330338991839Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Healthy housing concentrates on how residential area impacts the people's health. It devotes to not only satisfy the architecture provisions, but also meet the requirements of physiology, psychology, morality and social adaptation. Consequently, the housing environments will be more comfortable and healthier.Data mining is used to explore information and knowledge from mass data, which we depend on to analysis the survey data from experimental residential area. Factors which are important to people's health and how they react on it will be discovered.On the basis of the framework for healthy housing, we propose the method that association rules mining applies to the domain. The specific researches are as follows.First of all, through comparative analysis on the difference between two data mining process models, CRISP-DM is adopted. We analyze the advantages and shortcomings of two association rules algorithms, and then chose the more comfortable to our conditions.Secondly, we use statistical analysis on the survey data, so as to acquire the property of the data, by which experts can acquire appropriate mining model. According to the method of multi-dimension association rules mining, we accomplish data preparing, creating mining model and evaluate the results by analyze the accuracy. Besides, we introduce domain driven data mining, improve the quality of the results.Finally, in the light of the requirement analysis, we complete the system architecture, and design the main functional module and data storing. Association rules module is included into the system, and is implemented.In this paper, we apply CRISP-DM and multi-dimension association rules mining into healthy housing, solve the mining problem of multi-dimension data, and acquire the rules that the factors affect people's health. Hence, we propose a new method for research of healthy housing, and the software system,Healthy Housing Evaluation and Rules Discovery Decision Support System,can contribute to the further research on it. Further research will focus on how to join the domain intelligence in healthy housing into data mining, so as to improve the actionability of results.
Keywords/Search Tags:healthy housing, data mining process model, multi-dimension data mining, healthy housing evaluation and rules discovery decision support system
PDF Full Text Request
Related items