基于RFID的物流大数据资产管理及数据挖掘研究

发布时间：2018-04-13 05:41

本文选题：移动设备 + RFID　；参考：《上海师范大学》2015年硕士论文

【摘要】：移动互联网的飞速发展,给资产管理技术提供了更多的机遇,如物联网、云计算和大数据技术的兴起。物流领域对RFID技术的广泛应用也使得该技术趋于成熟,大数据时代的到来也将物流业引入新的发展空间。虽然RFID技术已经在资产管理方面有很多应用,但其可移动性,便捷性仍有待改进。在物流配送领域,随着移动设备端所产生的越来越多的交易量,不断产生的巨量数据也需要越来越高效实用的数据处理方式,而在物流领域,物流大数据资产管理也成为优化物流管理和提高配送效率的新技术手段。本文提供的一种资产信息采集通道和数据挖掘分类算法解决方案。资产信息采集通道充分利用现有移动设备,通过OTG连线实现移动设备通过USB接口与RFID模块的连接,以网站访问和移动设备应用程序两种方式进行数据管理,满足不同客户的需求。当前以手机为主的移动设备是互联网大潮的趋势之一,综合手机端应用是移动业务的重要载体,并以RFID读写模块作为数据采集的关键部分。本论文提出一种新的解决方案,并提供了多种工作模式,移动性好,附加投资成本低,其充分利用了高性能手机和移动网络优势,使基于RFID的资产管理更加便捷。在大数据来临时,数据作为新时代的资产和价值体现,已经越来越成为潜在价值的直接来源,如何有效、高效和有意义地挖掘和处理庞大的数据成为当今数据处理的主要问题,而分类算法作为数据挖掘的关键技术,已经广泛应用TB,PB级别的数据分析和处理中。数据在完成采集和存储,且在未来物联网中多种格式或制式的数据汇成非结构数据后,直接面临的问题就是在体量庞杂的数据库中寻找有价值的数据并为人们所用,继而为人们的进一步行动提供建议或预测。资产信息在完成数据化过程后,需要进行数据预处理、数据集成与挖掘和结果输出三个主要部分。本文研究了大数据技术在资产管理上的应用,以及数据挖掘中,应用基于一种超球覆盖仿生模式识别的分类算法,以及该算法在数据分类中的应用,分析该算法相较于其它算法的优缺点比较,以及在较大数据量时的算法性能。研究物流资产管理所应用的大数据处理算法及其性能以改进数据挖掘算法效率。对数据挖掘中分类算法的研究有助于数据处理时算法运算时间和运行效率的改善,在分类中也要平衡分类准确度和运算速度。本文所提出的基于RFID的资产信息采集通道是应用移动终端和移动互联网的一种终端数据采集方式,基于超球串的仿生模式识别分类算法应用于数据挖掘中,则在一定程度上对数据分类算法性能有一定改善和提高。
[Abstract]:The rapid development of mobile Internet provides more opportunities for asset management technology, such as the Internet of things, cloud computing and big data technology.The wide application of RFID technology in the field of logistics makes the technology more mature, and the arrival of big data also introduces the logistics industry into a new development space.Although RFID technology has many applications in asset management, its mobility and convenience still need to be improved.In the field of logistics and distribution, with more and more trading volume generated by mobile devices, the huge amount of data that is constantly generated also needs more and more efficient and practical data processing, while in the field of logistics,Logistics big data asset management has also become a new technical means to optimize logistics management and improve distribution efficiency.This paper provides a solution of asset information collection channel and data mining classification algorithm.The asset information collection channel makes full use of the existing mobile devices, realizes the connection between the mobile device and the RFID module through the USB interface through the OTG connection, and manages the data in two ways: website access and mobile device application.Meet the needs of different customers.At present, mobile devices mainly based on mobile phone are one of the trends of the Internet. The integrated mobile phone application is an important carrier of mobile services, and the RFID reading and writing module is the key part of data acquisition.This paper proposes a new solution and provides a variety of working modes with good mobility and low additional investment cost. It makes full use of the advantages of high performance mobile phone and mobile network and makes asset management based on RFID more convenient.With the arrival of big data, data, as the embodiment of assets and value in the new era, has become a direct source of potential value. How to mine and deal with huge data effectively, efficiently and meaningfully has become the main problem of data processing nowadays.As the key technology of data mining, classification algorithm has been widely used in data analysis and processing.When data is collected and stored, and many formats or formats of data in the Internet of things are assembled into unstructured data in the future, the immediate problem is to find valuable data in a large and complex database and to use it for people.In turn, provide suggestions or forecasts for further action.After completing the process of data processing, asset information needs three main parts: data preprocessing, data integration and mining, and result output.This paper studies the application of big data technology in asset management, the classification algorithm based on hypersphere covering bionic pattern recognition in data mining, and the application of this algorithm in data classification.The advantages and disadvantages of the algorithm compared with other algorithms are analyzed, and the performance of the algorithm in large amount of data is also analyzed.In order to improve the efficiency of data mining algorithm, big data algorithm and its performance in logistics asset management are studied.The research of classification algorithm in data mining is helpful to improve the operation time and efficiency of the algorithm in data processing, and to balance the accuracy and speed of classification in classification.The asset information acquisition channel based on RFID proposed in this paper is a kind of terminal data acquisition method based on mobile terminal and mobile Internet. The bionic pattern recognition and classification algorithm based on hypersphere string is applied to data mining.To some extent, the performance of data classification algorithm is improved.
【学位授予单位】：上海师范大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP391.44;TP311.13

【参考文献】