严正声明:近期,《海洋技术学报》编辑部发现某些中介组织利用假冒网站、电子信箱、电话等方式,以可以快速刊发稿件为由,向作者提供假冒录用通知并收取费用。对此,我刊发布严正声明,详见动态信息公告。
基于海洋地理知识图谱的智能搜索引擎系统
Intelligent Search Engine System Based on Marine Geographic Knowledge Map
  
DOI:
中文关键词:  知识图谱  搜索引擎  海洋地理  多模态
英文关键词:knowledge graph  search engine  marine geography  multimodality
基金项目:
作者单位
贺鹏艺, 毛邓添, 赵瑞可, 宁方辉  
摘要点击次数: 275
全文下载次数: 495
中文摘要:
      随着大数据时代的到来, 网络上各行各业的信息浩如烟海。在海洋地理信息领域, 由于信息来源间相互封闭, 垂直信息的搜索查询过程带来极大不便。对于海洋地理信息领域的静态数据及动态网站数据, 为实现数据的融合及智能搜索, 本文针对通用搜索引擎结果不够专业化、数据单一的问题进行改进, 搭建了一种面向海洋地理信息并基于知识图谱的智能搜索引擎系统, 首先采集与海洋地理相关的文献、档案、地图、数据、气象等类型的多模态信息, 而后使用neo4j 图数据库和Mysql 关系数据库融合结构化和非结构化数据, 构建及存储海洋地理专业知识图谱。最后, 当计算相似得分时, 将结果集中每条搜索结果的全部文本信息与检索词进行频率计算, 根据得分对搜索实例排序。选择部分搜索实例对搜索引擎进行实验和验证, 与通用搜索引擎进行比较, 实验结果证明该系统能够较好地适用于海洋地理专业信息的精准查询和智能推荐。
英文摘要:
      In this paper, an intelligent search engine system for marine geographic information based on knowledge map is established. vue+element-ui framework is adopted to design the front display layer, and nine subject query methods including disciplines, institutions, personnel, literature, information, data, models, terms and place names are designed. Use the springboot framework to build a back-end service layer for users, project design, modification, and management. On the basis of building the system, data collection and map construction are carried out. Firstly, multi-modal information related to marine geography such as literature, documents, maps and data is retrieved by selenium, and then the knowledge map is constructed and stored by using neo4j map database and Mysql relational database. Finally, some search examples are selected to experiment and verify the system. The experimental results show that the system can be well applied to the accurate query of professional information of marine geography.
查看全文  查看/发表评论  下载PDF阅读器