基于BP神经网络的湖南省红色旅游资源评价模型研究

发布时间：2018-03-07 06:05

本文选题：红色旅游资源　切入点：BP神经网络　出处：《湘潭大学》2017年硕士论文　论文类型：学位论文

【摘要】：湖南省红色旅游资源具有较强的独特性,具备及其丰富的遗留革命文化资源,是红色旅游资源的重要源泉之一。红色旅游资源评价是基于红色资源的角度,对其传播价值、文化价值、经济价值、教育价值等进行全方位、多角度的综合评价,构建科学合理的评价模型,能够为红色旅游资源的开发利用提供依据。红色旅游资源评价受到多种因素的影响,评价过程是一个多变量、复杂度高的非线性程序。目前,运用较多的评价方法是层次分析法、模糊综合评价法,能够得出较为客观真实的评价结果,但仍可以进一步优化。运用BP神经网络能够实现对红色旅游资源,尤其是未发掘的红色旅游资源(如长征线路、茶马古道、湖南农民运动考察等)进行自学习的智能化评价,有效避免主观随机因素,客观地对红色旅游资源进行评价。在梳理相关文献、掌握BP神经网络基础理论的前提下,从类型、分布、总体特征三个角度介绍模型训练数据来源(湖南省红色旅游资源)的概况,构建出湖南省红色旅游资源评价模型的指标体系,用于收集专家对所选景区的打分数据,并以此作为BP神经网络模型的输入数据。指标体系分为观赏、社会、经济、艺术文化、政治教育以及社会情感价值六个维度,十七个具体指标。将所搜集的数据进行归一化处理,构建出层数依次为17(输入层)、14(隐含层)、1(输出层)的三层BP神经网络模型,输入归一化处理过的数据进行训练,直到误差小于0.001为止,说明训练结束,得到最终的神经网络模型。最后,运用韶山景区、杨开慧故居、胡耀邦故里景区的数据对训练好的神经网络模型进行检验,得出评分分别为0.69914、0.60021、0.57988,最终评价等级为“良好”或接近良好,证明其整体红色资源状况较好,与实际质量评定等级基本相符,同时也说明本文所构建的BP神经网络评价模型能够用于红色旅游资源的发掘、评价与保护。
[Abstract]:The red tourism resources in Hunan Province are unique and rich in revolutionary cultural resources, which is one of the important sources of red tourism resources. The evaluation of red tourism resources is based on the angle of red resources and its spreading value. Comprehensive evaluation of cultural value, economic value, educational value, etc., and the construction of a scientific and reasonable evaluation model. It can provide the basis for the exploitation and utilization of red tourism resources. The evaluation of red tourism resources is influenced by many factors. The evaluation process is a multivariable and highly complex nonlinear program. More evaluation methods are analytic hierarchy process and fuzzy comprehensive evaluation, which can get more objective and real evaluation results, but can be further optimized. The use of BP neural network can realize the red tourism resources, In particular, the unexplored red tourist resources (such as long March routes, Tea Horse Ancient Road, Hunan Peasant Movement investigation and so on) carry out intelligent self-study evaluation, effectively avoiding subjective random factors, Objectively evaluate the red tourism resources. Under the premise of combing the relevant literature and mastering the basic theory of BP neural network, from the type, distribution, This paper introduces the general situation of the model training data source (red tourism resources of Hunan Province) from three aspects of overall characteristics, and constructs an index system of the evaluation model of red tourism resources in Hunan Province, which is used to collect the scoring data of selected scenic spots by experts. The index system is divided into six dimensions: ornamental, social, economic, artistic and cultural, political education and social emotional value. Seventeen specific indicators. The collected data are normalized, and a three-layer BP neural network model with 17 layers (input layer 14 (hidden layer / output layer) is constructed, and the normalized data is input and trained. Until the error is less than 0.001, it shows that the training is over and the final neural network model is obtained. Finally, the trained neural network model is tested using the data from Shaoshan Scenic area, former residence of Yang Kaihui, and Hu Yaobang hometown Scenic area. The final evaluation grade is "good" or "close to good", which proves that the overall red resource condition is better, which is basically consistent with the actual quality evaluation grade. It also shows that the BP neural network evaluation model can be used to explore, evaluate and protect the red tourism resources.
【学位授予单位】：湘潭大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP183;F592.7

【参考文献】