Full-text Retrieval Based On Vessels’ Technical Data Integrated Management System

With the rapid development of marine vessel technology, various technical equipmentshave been being loaded on vessels, which enlarges the volume of technical data. What’smore, there are various kinds of technical data and the formats of them are also diverse.Therefore, how to manage these technical data and to find the required information is anurgent problem. According to the practical needs of marine vessel technicians, this paper didan intensive study and design on the technical data integrated management system, whichcan provide more accurate information for users.First of all, according to the development process of software engineering andcombining with the user’s requirement, this paper analyzed and described thoroughly thetechnical data integrated management system, made the general design of logistic andphysical structure of this system, then analyzed and designed the function modules ofpersistence layer. The author laid an emphasis on the full-text retrieval module of businesslogic layer on account of the direct function of this system——search.Moreover, there are two kinds of important technologies of full-text retrieval——Chinese words segmentation technology and query optimization technology. As a result of it,this paper did a research in the two technologies. The one is analyzing the currentalgorithms of Chinese words segmentation, introducing a difficult problem——theambiguity problem. After comparing the algorithms of ambiguity recognition andprocessing, the author proposed an algorithm to recognize ambiguities and adopted a usefulambiguity processing algorithm. The other one is analyzing the query expansion technologywhich is a branch of query optimization. According to the practical needs of the system, thispaper adopted a query expansion method, and did an experiment to verify it. The resultshows this algorithm can improve the search efficiently.This paper has done the following research. First, the author proposed an algorithmwhich was an optimized algorithm of ambiguity recognition of Chinese words segmentation,combining Maximum Matching and Literal Scanning Algorithm. The second one is solvingthe recognized ambiguity by using the ambiguity solving algorithm based on statistical rules.The third one is using the query expansion algorithm based on context of document andsearch results in order to improve the query.