With the rapid development of Internet and big data, place retrieval has become an indispensable part of daily life. However, traditional retrieval technology cannot meet the semantic needs of users. Knowledge graph has been introduced into the new-generation retrieval systems to improve retrieval performance. Knowledge graph abstracts things into entities and establishes relationships among entities, which are expressed in the form of triples. However, with the expansion of knowledge graph and the rapid increase of data volume, traditional place retrieval methods on knowledge graph have low performance. This paper designs a place retrieval method in order to improve the efficiency of place retrieval. Firstly, perform data preprocessing and problem model building in the offline stage. Meanwhile, build semantic distance index, spatial quadtree index, and spatial semantic hybrid index according to semantic and spatial information. At the same time, in the online retrieval stage, this paper designs an efficient query algorithm and ranking model based on the index information constructed in the offline stage, aiming at improving the overall performance of the retrieval system. Finally, we use experiment to verify the effectiveness and feasibility of the place retrieval method based on knowledge graph in terms of retrieval accuracy and retrieval efficiency under the real data.

中文翻译：

---

知识图中的位置检索

随着互联网和大数据的飞速发展，场所检索已成为日常生活中不可或缺的一部分。但是，传统的检索技术无法满足用户的语义需求。知识图已引入到新一代检索系统中，以提高检索性能。知识图将事物抽象为实体，并在实体之间建立关系，这些关系以三元组的形式表示。但是，随着知识图的扩展和数据量的迅速增加，传统的知识图场所检索方法的性能较低。本文设计了一种场所检索方法，以提高场所检索的效率。首先，在离线阶段执行数据预处理和问题模型构建。同时，建立语义距离索引，空间四叉树索引，以及根据语义和空间信息的空间语义混合索引。同时，在在线检索阶段，基于离线阶段构造的索引信息，设计了一种高效的查询算法和排序模型，旨在提高检索系统的整体性能。最后，通过实验验证了基于知识图的场所检索方法在真实数据下的检索精度和检索效率的有效性和可行性。旨在提高检索系统的整体性能。最后，通过实验验证了基于知识图的场所检索方法在真实数据下的检索精度和检索效率的有效性和可行性。旨在提高检索系统的整体性能。最后，通过实验验证了基于知识图的场所检索方法在真实数据下的检索精度和检索效率的有效性和可行性。