城市物流效率分析自适应DBN算法研究

发布时间：2018-02-27 06:00

本文关键词： 深度学习深度信念网络物流效率自适应聚类　出处：《计算机工程与应用》2017年12期 　论文类型：期刊论文

【摘要】：深入研究了城市物流效率分析的研究现状,结合深度学习相关理论,针对具体问题构建三隐层连续型深度信念网络(DBN),对网络知识集进行了定义,提出了自适应DBN算法,分析了算法的收敛性。利用Iris数据集和Wine数据集验证了网络及算法的模式分类能力,分类精度高于双隐层深度信念网络与深度误差反向传播网络。根据新丝绸之路经济带沿线城市物流特点,以物流效率为评估目标,选取4个维度的13项指标建立评价指标体系,以20个核心节点城市为研究对象,利用自适应DBN算法和社会网络分析法(SNA)进行聚类分析,结果表明自适应DBN算法相对更为合理有效。研究结果为确定新丝绸之路经济带沿线城市物流发展策略、促进国内物流业未来的协作与发展奠定了研究基础。
[Abstract]:In this paper, the current research situation of urban logistics efficiency analysis is deeply studied. Based on the theory of depth learning, a three-hidden layer continuous depth belief network is constructed, and the knowledge set of the network is defined, and an adaptive DBN algorithm is proposed. The convergence of the algorithm is analyzed. The ability of pattern classification of the network and the algorithm is verified by using the Iris data set and the Wine data set. The classification accuracy is higher than that of double-hidden layer depth belief network and depth error back-propagation network. According to the characteristics of urban logistics along the New Silk Road economic belt, the evaluation index system is established by selecting 13 indexes of 4 dimensions, taking logistics efficiency as the evaluation objective. Taking 20 core node cities as the research object, the cluster analysis is carried out by using adaptive DBN algorithm and social network analysis method. The results show that the adaptive DBN algorithm is more reasonable and effective. The results of the study lay a foundation for the determination of the logistics development strategy along the new Silk Road economic belt and the promotion of the future cooperation and development of the domestic logistics industry.
【作者单位】：中国(西安)丝绸之路研究院;西安财经学院管理学院;
【基金】：中国(西安)丝绸之路研究院基金项目(No.2016SY14) 陕西省社科联项目(No.2017Z055) 陕西省自然科学基金(No.2014JM9360) 西安财经学院校级科研基金项目(No.14XCK08)
【分类号】：F259.2;TP18

【相似文献】