节点文献

基于决策树的数据挖掘算法在空气质量评估中的应用

Application of Decision Tree Data Mining Algorithm in Air Quality Assessment

【作者】 汪滢

【导师】 陈炼;

【作者基本信息】 南昌大学 , 计算机软件与理论, 2009, 硕士

【摘要】 空气质量评估是人类保护自然资源、兼顾环境与效益、合理控制污染排放及预防严重污染事件发生的重要评估手段。其中空气质量评估算法则是空气质量评估中的关键。目前常用的空气质量评估算法有空气污染指数法、模糊综合评价法、欧几里得贴近度法等。但这些算法中存在单因子评价、指数范围的人为因素、指数值界限过于清晰等缺点,在日常使用中还是存在评估的误差。本课题综合分析国内外研究现状,在探讨空气质量评估常用方法的基础上提出结合决策树算法的评估方法。它基于ID3算法多值偏向的特点和属性相似度的决策树具有少值偏向的特点,并将两者结合起来,在选择测试属性时采用属性相似度作为校正系数的信息增益计算方法对空气质量进行评估。本文结合某市空气质量的具体数据进行仿真实验,分析实验结果表明改进的算法对于决策树的建立和规则的生成,在分类的效率和正确性上都有显著的提高。基于决策树的空气质量评估算法,不仅能指导研究人员正确地分析和掌握空气中污染物的迁移变化规律,而且具有良好的社会经济效益与应用前景。

【Abstract】 Air quality assessment is an important assessment tools of human protection of natural resources, taking into account the environment and benefits, reasonable control of pollution emissions and prevent serious pollution incident. Air quality assessment algorithm is the key to air quality assessment.Now commonly used in air quality assessment algorithm for air-pollution index method, fuzzy comprehensive evaluation method, Euclid close to the degree method. But there are some shortcoming in single-factor evaluation, the index range of human factors, the index value limit is too clear .The Project comprehensive analysis of the present study of the status at home and abroad, to explore the air quality assessment methods commonly used decision tree algorithm based on the combination of assessment methods. It is based on ID3 algorithm for multi-valued bias of the characteristics and properties of similarity of the decision tree has a few characteristics of the value of bias, and combination of the two, the choice of test properties attribute similarity used as a correction factor of information gain calculation.In this project, a city of specific air quality simulation data analysis results show that the algorithm improved the establishment and rules for decision tree generation, the efficiency and correctness of the classification on both a significant increase. The air quality assessment based on decision tree algorithm, not only to guide us to correctly analyze and control the migration of pollutants in the air changes in the law, but also has a good social and economic benefits and application prospects.

【关键词】 空气质量评估决策树ID3算法
【Key words】 Air quality assessmentdecision treeID3 algorithm
  • 【网络出版投稿人】 南昌大学
  • 【网络出版年期】2011年 S1期
节点文献中: 

本文链接的文献网络图示:

本文的引文网络