|
|||||
|
|
| 多源异构土地基础数据一体化管理检索方法研究 | |
| 引用本文： | 张书瑜,张定祥,王荣彬,季宏伟.多源异构土地基础数据一体化管理检索方法研究[J].浙江大学学报(理学版),2018,45(5):589-594. |
| 作者姓名： | 张书瑜 张定祥 王荣彬 季宏伟 |
| 作者单位： | 1. 浙江大学 地球科学学院, 浙江 杭州 310027; 2. 中国土地勘测规划院, 北京 100035 |
| 基金项目： | “十二五”国土资源调查评价--土地基础数据库整合集成与共享平台建设项目（DCPJ131707-01）. |
| 摘 要： | 为了从多源异构的复杂土地基础数据中快速准确地提取用户所需信息，提出了基于元数据的一体化管理检索方法.在元数据信息提取、元数据加权索引、实体同义词扩展检索3个环节中，结合土地领域专业知识和用户实际需求，设计和开发了共享元数据表结构、加权元数据中字段相对重要性和信息熵因子，构建地名实体和专题数据层实体同义词库，并集成到包括中文分词、实体识别、同义词扩展、索引检索和相似度计算的一体化管理检索框架中，解决了多源异构土地基础数据统一管理和精确检索的问题.实践表明，该方法较传统的通用信息检索方法具有更好的适用性和更高的准确率. |
| 关 键 词： | 多源异构土地基础数据 管理检索一体化 元数据信息提取 元数据加权索引 实体同义词扩展检索 |
| 收稿时间： | 2017-03-06 |
Research on integrated management and retrieval method of multi-source heterogeneous land basic data |
|
| ZHANG Shuyu,ZHANG Dingxiang,WANG Rongbin,JI Hongwei.Research on integrated management and retrieval method of multi-source heterogeneous land basic data[J].Journal of Zhejiang University(Sciences Edition),2018,45(5):589-594. | |
| Authors: | ZHANG Shuyu ZHANG Dingxiang WANG Rongbin JI Hongwei |
| Affiliation: | 1. School of Earth Sciences, Zhejiang University, Hangzhou 310027, China; 2. China Land Surveying and Planning Institute, Beijing 100035, China |
| Abstract: | In order to obtain the required information quickly and accurately from the complex multi-source heterogeneous land basic data, an integrated management and retrieval method based on metadata is proposed. More concretely, during the process of metadata information extraction, metadata weighted indexing and entity synonyms extended retrieval, three optimized methods are performed combined with the field expertise of land and the actual needs of users, which are design and development of sharing metadata structure, construction of weighted index based on relative importance of metadata columns and information entropy factor, and building synonym database of geographic name entities and thematic data layer entities. An integrated management and retrieval method is proposed, including features of word segmentation, entity recognition, synonym extension, index retrieval and similarity computation. And, the optimized methods mentioned above are integrated into the framework for unified management and precise retrieval for multi-source and heterogeneous land basic data. Experimentation and practical application show that the proposed method presents higher accuracy and better applicability than the traditional general information retrieval method. |
| Keywords: | multi-source heterogeneous land basic data integrated management and retrieval method metadata information extraction metadata weighted index entity synonyms extended retrieval |
| 本文献已被 CNKI 等数据库收录！ | |
| 点击此处可从《浙江大学学报(理学版)》浏览原始摘要信息 | |
| 点击此处可从《浙江大学学报(理学版)》下载全文 | |
