Research On The Descriptive Geographic Information Retrieval Based On Relevance Evaluation

In recent years, as the Internet and 3s are developing rapidly, the data people contact shows explosive growth. According to research, more than 80% of the data related to the spatial information. Geographic information which delegate is 3s has become a hot topic of today’s society, especially the use of geographical location information which implied in geographic information and the information extraction research is increasingly attracting the attention of scholars both at home and abroad, and become the research hot spot. In the long run, the research on the information retrieval based on the geographic location and the related services have broad application prospects. In this paper, the research is based on the Heilongjiang organizations code information vector data and uses the information retrieval technology to complete descriptive geographic information retrieval based on relevance evaluation. The research has certain science significance and application value. The main researches of the thesis are as follows:1、Geographic Descriptive Information Retrieval:This paper introduced the progress of the descriptive retrieval statement parsing in detail. Statement segmentation work was done by the processing of word segmentation and part-of-speech tagging for retrieving statements. Then combined with the characteristics of the geographic information retrieval, it summarized four geographic information retrieval sentences. With these sentences, the contents of retrieving statement were matching to the corresponding sentence parts to complete information extraction.2、Geography-Semantic March Model:From the geography-semantic marching, according to the corresponding relationship between the map and the natural language, this paper put forward an improved geography-semantic marching model, and discussed the work process and implementation steps of the models to determine the geographic scope of the query.3、Correlation Calculation:It respectively introduces the text relevance and geographical correlation calculation method. In the process of text correlation, query expansion used made the retrieval results more ideal.4、Prototype System Design and Experiment Verification:The above contents and algorithms has carried on the practice by examples. It implements the entire process from statement parsing、the determine of the geographic scope of the query, query expansion, correlation calculation to the express of the retrieval results. This is the full display for the descriptive geographic information retrieval based on relevance evaluation.