The Design And Implementation Of A College Score Analysis System

In-depth analysis of student test scores is the effective basis for objectively evaluating and positioning on students’learning and teachers’teaching effectiveness. Currently, most of the high schools use simple statistical methods to retrieve some characteristics of test scores. Based on the low level analysis results, it is hard for us to get profound information with directive significance for further work. In addition, although colleges have already built academic information management systems, different colleges, even different faculties within a same college may use different score management systems. As different score management systems have different organizations and storing methods, it is difficult to integrate score information among these score management systems, which makes it even harder to perform score analysis on a higher level.A good result analysis system should possess the function on macroscopic mathematical statistics, as well as on detailed analyses. In addition, although some of the more innovative educational management softwares have a relatively comprehensive statistical functions, the expansion of the software itself is too weak due to the great differences of the performance management system among all colleges and universities. Besides, the existing performance analysis software is difficult to adapt to the actual situation of the majority of colleges and universities. The expended software still has many problems in data management and delivery. Therefore, it is necessary to design a performance analysis system to adapt to the current situation, which is easily extended for colleges and universities.In this paper, the relevant data collection is analyzed, combined with the real situations that colleges are facing. We present an information extraction technique based on the Hidden Markov Model and tree automata theory, and an in-depth analysis and conclusions as well. Then we introduced the Design and Implementation of the college score analysis system based on C/S structure. The system successfully implemented the information extraction technique based on Hidden Markov Model and tree automata theory. The system first uses Hidden Markov Model to find the relationship between the data fields and the elements in the score page of the original score management system. And then builds a tree automata used for dumping field data, so that the automatic score information extraction is implemented for the original score management system. By analyzing the extracted score information, the gap between different data storing structures formed by different score management systems is filled, and then the score analysis among different score management systems will be successfully realized.