Applied Mechanics and Materials Vols. 513-517

Abstract: BBS (Bulletin Board System) forum of campus provides a main platform that students from the same university can communicate with each other by posting a post or comment others posts. The posts and comments in campus BBS contain useful information about the students. In this paper, we design and implement a program to collect the emotional data that were generated in the last two months in BBS of Shanghai Jiao Tong University, and analyze the fluctuation of the emotion along with time. Firstly, we implement a simple crawler to fetch the content of posts and comments; secondly, using Chinese lexical analysis system to get the words; then, give each word an emotional tendency (positive or negative) by using lexicon and analyze the result.

Abstract: Regarding to the technical difficulties of reachingvaluable knowledge and information from CRM mass of data and information, this article brings up the idea that we could apply the data mining technology in CRM. The data mining classification methods helps us to find out the potential customers with high value. The article describe how the methods and processes how could we apply the data mining technology to CRM and on the basis of decision tree algorithm to construct the CRM data mining model, and help companies better understand the customers behaviors and improve the core competitiveness of enterprises.

Abstract: Cloud computing is a novel network-based computing model, in which the cloud infrastructure is constructed in bottom level and provided as the support environment for the applications in upper cloud level. The combination of clouding computing and GIS can improve the performance of GIS, and it can also provide a new prospect of GIS information storage, processing and utilization. By integrating cloud computing and GIS, this paper presented a cloud computing based GIS model based on two features of cloud computing: data storage and transparent custom service. The model contains two layers: service layer and application layer. With this two-layer model, GIS can provide stable and efficient services to end users by optimized network resource allocation of underlying data and services in cloud computing.

Abstract: The co-occurrence word emphasize the word and word internal relations, so its use can improve shortage from the hypothetical of Bayesian algorithm. To build Token Dictionary, Information Gain algorithm is used to choose Tokens, and Synonyms Dictionary is used to acquire more Tokens. By large amounts of training, the matching scores of Token are counted, according to the matching rate the Tokens that is valuable are selected, and the Token Dictionary is established. The proposed method is used to E-mail classification experiment, the results show that the accuracy of spam filter has a well improvement.

Abstract: Many legacy applications rely on a single instance relational database. In the age of big data, applications with growing popularity may continue to generate large amount of data. The data would be beyond the limit of a traditional relational database. These applications can be migrated to a distributed database system for acquiring high scalability and performance. But the migration process is complicated especially for user defined stored procedure. In order to obtain the correct behavior, a stored procedure usually has to be rewritten in the application code. This rewriting is usually a time consuming and unpleasant task. This study gives a mechanism for migrating stored procedure from single instance relational database to distributed sharded database without the need of application code rewriting. In the migration solution offered by this study, Oracle stored procedure code is parsed, analyzed, and translated so as to be stored and executed globally in a sharded database.

Abstract: The design of books analysis system is beneficial to sufficiently using the accumulated data resources to guide work for librarians and help library leaders make decisions. The paper presents a system architecture of library books based on data warehouse. It focuses on several key technologies in building data warehouse, analyze and design the client application, It uses the accumulated data of college library as data resources , simulate simple data warehouse and use OLAP technology. It makes manager analysis data on different levels and aspects and show the result in the form of forms and charts.

Abstract: With the widespread use of mobile devices, universities are speeding up the construction of digital campus. Mobile digital campus system described in this paper contains campus view, voice campus navigation, voice dialogue, information query, campus news, books query, study BBS and other functions. Its key technology combines the query with voice, maps and other features forms an integrated Android mobile application platform with powerful function and program scalability.

Abstract: EDI is an important part of logistics information technology and is also the basis and prerequisite of the implementation for supply chain logistics information system. This paper researches on electronic data interexchange of embedded Linux to promote EDI technology to be widely used in the field of logistics control. First, the paper studies kernel structure of embedded Linux that includes process scheduling, memory management, virtual file system, network interface, inter-process communication and methods and procedures of kernel cutting; then, we can research a number of issues in the logistics field of electronic data interchange supported by embedded Linux. The issues include workflow, hierarchical structure and main applications and so on. Contents of this paper extend the application area and support environment of EDI. In addition, it is an important research direction in logistics informatization used EDI technology.

Abstract: Discrete event systems are widely used in the production and life, it is difficult to use conventional differential equations, differential equations, and other models to describe, the theoretical analysis method is difficult to obtain analytical solutions, computer simulation techniques to solve these problems provides an effective means. Arrival event is a typical discrete system event; on arrival event handling is always one of the difficulties of computer simulation, in this paper, banking customer arrival system as an example to study. For banks queuing system, customers arrive to obey the parameter of Poisson distribution is, the probability mass function through the distribution curves and cumulative distribution function curves to study the distribution of customer arrival; construction of single-queue multi-server system of customer arrival event subroutine flow chart, and processing steps will be described. Content of this study, it is suitable for the developed area bank to adopt "number ticket machine" approach to service.

Abstract: University scientific research management platform for the scientific management research personnel to carry out the work to provide convenience, for the leadership decision-making to provide auxiliary support, for teachers to carry out research activities to provide services. This paper studies the basic work of building university scientific research management platform. For the problem that the function is difficult to determine, it researches functional framework composed of scientific research information, scientific research results, scientific research projects, scientific research statistics, portals, system management and others subsystem; For the problem that the software development is difficult to achieve, it based on B / S structure to research three layers framework structure consisted of the user presentation layer, business logic, data access layer, and to describe each layer. Using the contents of this paper to build scientific research management platform, for the standard software development methods and improving the efficiency of system development has great significance.