The Development And Text Information Mining Based On Rail Transportation Document Management System

Today, China is entering into a booming era of rail transportation construction. Take Beijing for example: there are 6 subway programs to be under construction in 2009; combined with those 5 programs right in construction, there will be 11 in all. Rail transportation project data management is an important part in rail transportation construction as well as important files for the construction organization in charge. These data fully reflects information about the quality and checking process of rail project programs, as a result they are attached with vital importance. However, since there are no standard and unified computerized codes and the idea of computerizing has not been employed in solving the problem of how to code Documents and Volume them, data management can only be done by person handedly. This brings many problems, such as low efficiency, weak accuracy, which prevent us from calculating, searching, storing and managing these data well. As a result, proper software is in need to solve the problem of coding and volume. We analyze this problem in this thesis in detail and provide solutions to it.Firstly, we introduced the present situation in China's rail transportation project data management. We did some on-field research, evaluated the process of data management, categorized and numbered different rail transportation project data according to computerizing codes, and finally met the need of volume and coding. In addition to this, to improve the efficiency of data using, we introduced tree-style catalog, reduced the work of data repeating and the need of EMS memory, and the maneuverability and read-write speed of the whole system is greatly improved.Secondly, this thesis combined the principle of on-field data management process and software projects, designed and explored a rail transportation data management system, by using the technique of target-oriented software development methods such as demand analysis, system analysis, overall design and system evaluation. In this new system, we solved some shorts that existed in former data management system by designing some functions such as project data management, volume management and system storage. Also we did experiments in several pilot locations, and the results showed that the system could meet the basic needs of front staff in rail transportation projects construction.In addition to the above things, this thesis also provided theoretical analysis to rail transportation projects construction data. Since quality problems of the project in construction can bring tremendous negative influences for the society, environment, transportation and economy for cities, so in-time analysis of under-construction data is closely related with the operation of a project. In the end, this thesis gave data evaluation to Beijing Subway, Line 4, Xidan station. The idea of data mining was employed. Relevant data involved in construction was abstracted, and then these information was pre-treated and attribute is curtailed, and the information about the quality of construction and classification information is finished by using the Clustering algorithm system finally.