贝叶斯网络预测模型的应用研究毕业论文
2022-02-02 22:26:16
论文总字数:15048字
摘 要
贝叶斯网络起源于国外,最近几年在中国兴起,用于解决各领域的问题,它是通过研究概率的一套算法,通常用来解决分类,预测和概率估计问题,在经济学中更有着很大的成果。
近年来,随着人类社会物质文化水平的提高,人们不再仅仅追求物质上的满足,更加注重精神方面的追求,旅游已成为当代人主要的休闲娱乐活动之一。我国政府颁发了许多促进旅游业发展的政策,国内外旅游人数显著增加,交通餐饮旅馆饭店行业发展迅速,这都是围绕旅游业所兴起的产业,如何有效的加快旅游业结构升级显得尤为主要,换句话说,如何准确预测旅游业收入,分析影响旅游业收入的因素,从而达到有效促进旅游业发展已然成为政府部门眼下着力应当解决的问题之一。然而影响旅游业的因素很多,具有很大的不确定性,贝叶斯网络适合研究不确定性问题,因此,本文采用贝叶斯网络来研究旅游业的问题。
本文主要工作有:
(1)介绍了时间序列预测法模型,灰色预测模型以及贝叶斯网络预测模型,通过比较,发现贝叶斯模型是不是更适合预测旅游业收入。
(2)贝叶斯网络模型是当代比较热门的一种建模方式,介绍了贝叶斯相关理论基础,贝叶斯结构学习,模型优化评分以及贝叶斯参数学习。
(3)通过建立时间序列模型以及贝叶斯网络预测模型,通过比较发现贝叶斯模型更适合预测适合分析事故险情等特殊情况发生,而不适合研究均值附近的大多数情况
(4)通过分析研究的结果,提出一些建议,希望可以对有关部门提供一些参考。
关键词:国内旅游业 时间序列模型 贝叶斯网络预测模型 极大似然函数
Application of Bayesian Network Prediction Model
Abstract
The Bayesian network originated in foreign countries and has risen in China in recent years. It is used to solve problems in various fields. It is a set of algorithms for studying probability. It is usually used to solve problems of classification, prediction and probability estimation. Has great results.
In recent years, with the improvement of the material and cultural level of human society, people no longer only pursue material satisfaction, but also pay more attention to the pursuit of spirituality. Tourism has become one of the major leisure and entertainment activities of the contemporary people. The Chinese government has issued many policies to promote the development of the tourism industry. The number of domestic and foreign tourists has increased significantly. The transportation and catering hotel industry has developed rapidly. This is all about industries that have risen in the tourism industry. How to effectively accelerate the upgrading of the tourism industry is particularly important. In other words, how to accurately forecast tourism income and analyze the factors that affect tourism income, so as to effectively promote the development of tourism has become one of the problems that government departments should focus on now. However, there are many factors affecting the tourism industry and there is a great deal of uncertainty. The Bayesian network is suitable for researching uncertainty issues. Therefore, this paper uses the Bayesian network to study the tourism issues.The main tasks of this article are:
(1) Introduce the time series forecasting model, grey forecasting model and Bayesian network forecasting model. Through comparison, we find that the Bayesian model is more suitable for forecasting tourism income.
(2) The Bayesian network model is a relatively popular contemporary modeling method. It introduces Bayesian related theoretical basis, Bayesian structure learning, model optimization scoring and Bayesian parameter learning.
(3) By establishing a time series model and a Bayesian network prediction model, it is found that the Bayesian model is more suitable for predicting accidents and other special situations and is not suitable for studying most situations near the mean value.
(4) Through the analysis and research results, some suggestions are made, and it is hoped that some relevant departments can provide some references.
Keywords: domestic tourismtime;series model; Bayesian network prediction model; maximum likelihood function
目录
摘要 I
Abstract II
第一章绪论 1
1.1研究的背景意义 1
1.2国内外研究现状综述 2
1.3 研究的目的方法 4
1.4常见的模型综述介绍. 4
第2章贝叶斯网络相关的理论基础 6
2.1概率论与数理统计基础 6
2.2贝叶斯结构学习 7
2.2.1常用的模型结构评分函数介绍 8
2.2.2 模型优化的方法介绍 10
2.3贝叶斯参数学习 11
2.3.1极大后验概率(MAP) 11
2.3.2贝叶斯参数估计 11
2.3.3 贝叶斯预测公式 12
第3章旅游发展的贝叶斯预测模型 13
3.1 旅游发展的现状 13
3.2旅游发展因素影响 13
3.3旅游发展的计量经济模型(时间序列模型) 14
3.4贝叶斯网络预测模型 18
第4章结论及展望 22
致谢 23
参考文献 24
附录 25
第一章 绪论
1.1研究的背景意义
图表 1国内近年旅游业收入
我们从上面的图像可以看出,近几年,国内旅游业发展迅速,从2010年到2016年,这七年时间,国内旅游业收入,从一开始的12579.80亿元,提高到了39390亿元,2017年比2010年的收入整整翻了一倍还要多,由此可以看出,世界已经迈入了旅游时代。人们随着时代发展,休闲娱乐的时间逐渐增多,休闲度假旅游成为了新的市场方向,各种旅游公司,特色餐馆,以及旅馆,在旅游业的带动下,都发展迅速,旅游业发展前景广阔。根据2017年的旅游业市场调查显示,我国国内旅游业目前正处于高速发展的趋势下,预计到2020年底,我国旅游业收入将增加至5000亿元,以年均5%的速度增长。越来越多的数据指明了当前旅游业如今的重要地位,占市场经济的比重更是越来越大,有专家预测,到2020年,旅游业收入将占世界经济的10%之多。有旅游业带动的相关产业经济发展更是不用多说,称它为新的经济增长点毫不夸张。我国旅游业虽然起步较晚,但有着先天优势,因为我国名胜古迹资源丰富、交通发达,对国内外旅游者的吸引力力强,截至2016年6月,我国已有50处文化遗址和自然景观列入联合国科教文组织《世界遗产名录》,位列世界第二位,目前我国旅游业主要以观光旅游为主,这50多处景点,每年都吸引着各种各样的旅客;另一方面,我国的旅游资源尚未开发完全,尤其是东部,我相信随着东部资源的不断开发,旅游业将反推动着东部经济的不断增长。
请支付后下载全文,论文总字数:15048字