Intelligence Platform Design Based on Big Data of Trade and Logistics

Abstract. The aim of the article is to construct an electronic commerce intelligence platform based on the data of trade and logistics ， according to the platform， three different layers have been prososed ：application layer， data layer and core layer. elaborate design for layers of the platform has been displayed. a clear description about the key technologies to implement the intelligent platform is also made.The platform design is so unique and innovative that has strong reference and application value.

Key words： component； formatting； style； big data； trade and logistics； intelligence platform design

1 Introduction

The concept of big data is， in fact， the integrated concept such as cloud computing， Internet of things from the perspective of huge amounts of data and effective use. At present， no generally accepted definition of big data， but there is a common cognition： the big data’s database system processing capacity is far more than the traditional one， higher requirements are also needed to transfer the speed and the scale of it. big data not only contains the "big data"， tell from the connotation， also contains complex types of data. Big data is all of the data set， including trade and interactive data set. Its size or complexity beyond the commonly used technology in accordance with the reasonable cost and time limit to capture， manage， and the ability to deal with these data sets. three periods can be traced to the development of big data：The first period is the data integration and "mining" within the industry，which involves some industries of a large amount of data derived. the enterprise internal data fusion and data mining is the second period.The third period is the organizational structure change based on data driven centering the consumers. Specifically， big data era mainly has the following features： information is highly personalized， barriers of time and space are broke down，Social innovation and public innovation are emerged. Big data era fully learn and absorb the latest achievements of information technology such as semantic web， business intelligence， pervasive computing， cloud computing， etc.

With the rapid development of electronic commerce， a large amount of data and accumulation of big data applications are produced in trade and logistics field. The classic example of "beer and diapers" in the data mining textbook is the best explanation. Under the background of big data in early 2008， a sharp fall in the number of buyers inquiries on alibaba platform. Enquiries index is a precursor， early active inquiry index marks a certain volume will be reached on later， shrinking trading volume can be speculated in the future if the inquiry index declines abnormally. From above we can predict that there is huge application space for big data in the trade and logistic field .It is of great significance for government， enterprises and individuals to explore regularity and law from big data in the trade and logistic field and then apply them to forecast analysis in the future， while an intelligence platform based on big data is the better solution and the gripper to solve the problem of trade and logistics. For the government， intelligence platform will provide guidance for industry development as well as provide intelligence support for the government decision-making through promoting further development of government informatization improving work efficiency and ensuring national economic security. intelligence platform will provide the enterprise with lower cost， higher work efficiency ，more profits and achieve the industry collaboration and fine management through coordinating the supply chain and carrying out precision marketing， for instance， analysising customer demand more accuratelly， guiding the customer consumption tendency and strategic decision-making .For individuals， the platform will ensure us with more abundant and valuable information to acquire， which benefits to our study， social intercourse， online shopping， and scientific research activities， etc.

2 Trade and logistics intelligence platform design

based on big data

Electronic commerce have higher request of its intelligence platform than before in the big data age. According to the needs of real life in today's world， a design framework is created based on the point in the figure below.

Application layer： provide application interface for foreign trade， commerce and trade， trade logistics， trade finance， trade education and other industries . Provide the enterprise with help in the following aspects： strategic decision-making， business opportunities found， precise marketing and supply chain coordination， fine management， improve work performance， across the channel analysis ability and industry collaboration. Providing enterprises with corresponding efficient solution through the real-time data exchange so as to improve the competitiveness of the enterprise.Providing application interface for the government to find some more effective ，practicial and feasibility schemes from the micro level of big data analysis. The design offsets deficiency of conventional government regulates business only through the macro data， which improves decision-making， the precision and efficiency of guidance and control in the government decision-making departments.

Data layer： includes customs clearance data， trade data， logistics data， enterprise data， government regulation， government information resources data services and corporate libraries， etc.）， Internet data （sina weibo， baidu， etc.）， port customs clearance data standards， exchange of heterogeneous data acquisition transmission， buying structured data， etc. A large data repository of trade and logistics is built with the fastest speed through the accumulation of above data and existing resources as well as through the purchases of government and enterprise of structured and unstructured data with segmentation， continuously expanded， optimization and improvement.

Core layer： big data intelligent analysis engine： industry experts and data mining as well as machine learning cooperation， according to different data and applications areas ，establishing all kinds of data processing mechanism and the business analysis model by using all kinds of algorithms （such as： classification and regression bayesian， support vector machine SVM， logistic regression， genetic algorithm （ga）， principal component analysis （PCA； such as clustering， k-means， spectral clustering， etc.）. According to different data and applications， a big data repository is built aims to looking for the data value of the trade and logistics industry to provide targeted and efficient service for the government and enterprises. The big data repository is built on the basis of complex large big data in structured form and unstructured form through intelligent analysis of big data. the methods are uesd in intelligent analysis of big data including association rule mining， rough set， neural network learning algorithm， extracting rules， multi-agent genetic algorithm optimization algorithm， particle swarm optimization algorithm， the decision tree classification algorithm， a one-dimensional search learning algorithm，quantum evolutionary algorithm， simulated annealing hybrid algorithm demo， the space image data mining， statistics and regression analysis method.

3 The key technology

Based on the 3 v characteristics of big data ，including data cleaning， unstructured data calibration and research of efficient algorithms. efficient and reliable analysis model is set up in response to the structured data by absorbing knowledge of industry expert . Practical，efficient and feasible algorithm model is set up in response to the large amounts of unstructured internet data by using a semi-supervised learning algorithm. schemes has been developed according to characteristics and requirements of the industry users， Which benefits to the development of the industry application.

4 Platform features and innovative points

4.1 The Platform Features

The acticle is unique interms of the depth and breadth of data acquisition own to there is so little in-depth research in domestic at present in the data collection of related information system that is based on the foreign trade enterprise of a small amount of data . In addition， this study is a big breakthrough in the field of data intelligent application in the trade logistics own to rarely research in the information system level while more exiting research on the theoretical research in this field. This system involves the forecast analysis in the field of trade and logistics， there are a variety of intelligent analysis of innovation method and simulation aim at big data. This research design platform has strong data visualization， especially its presentation layer has strong attraction and appeal. easily used， strong expressivity， actual operability and practicalness are main distinctive features. corresponding intelligent analysis algorithm are adopted in view of different application field， this platform through the real-time data exchange of industry and building industry structured and unstructured data repository， which provides accurate and efficient services for enterprises and government.

4.2 Platform innovation points：

a） This design platform in this article based on the analysis of the supply chain business process and state change of traceability documents and retrieval way as well as on the basis of related content， covering the whole process of logistics tracking model is put forward that based on internet of things technology. Visualization of logistics supply chain and tracking system can be constructed by adopting traceable unit， unit identification information， unit record information.

b） This design platform in this article build a spatial geographic information comprehensive repository with rich content that support multidimensional comprehensive information query，Which has realized the logistics management in real-time scheduling， increases the effective working time of the staff as well as save the traffic resources.

c） This design platform in this article is put forward based on the RFID information of multidimensional view model and a variety of end products logistics tracking mode， which provides scientific decision basis for the enterprise decision-making in different levels of the enterprise as well as increases the value of RFID data resources.

d） The platform in this article analysises and defines the port logistics business model. It can quickly definite and complete the data cleaning work according to the business model and object program. at the same time it ensures the accuracy of the data and qualified rate of sampling. Multidimensional data warehouse is established that based on business law that is explored in the shanghai port logistics as well as by using on-line analytical processing （OLAP） analysis，which ensure quickly and flexibly of data results of electronic commerce exhibition.Business intelligence service system is established that oriented to the port logistics of government departments and enterprises，which can realize data exchange between enterprise and service system as well as present the analysis report of enterprise operation in a short period of time on the basis of the standard template and the amount of data is determined.

e） This design platform in this article based on the analysis of transaction types， trade characteristics and trade process， scrap electronic trading and integrity model is put forward to ensure open and fair in the process of releaseing transaction information，bidding and similar occasions.

f） This platform realizes the cooperative interaction in work flow， documents， and information data among enterprises in logistics supply chain.This platform also realizes the transmission of documents and the collaborative business information sharing among the owner ，customs broker and 3PL， which improves the efficiency of foreign trade and reduce the cost of documents transmission.

References

[1] Cao lei. in 2012 （top） China B2B e-commerce market data monitoring report [R]. China e-business research center， 2012 （8）.

[2] Yang Wenlan. The structure of China's trade imbalance problem research [J]. Economic BBS， 2011 （3）.

[3] Wang Yuting. Empirical study on relationship between foreign trade and economic growth in liaoning province [J]. Journal of logistics economy， 2010 （36）.

[4] Wu Yanfeng. Trade logistics urgently need standardized in Asean free trade area [J]. Journal of commerce in China， 2012 （6）.