A Practice of Tourism Knowledge Graph Construction based on Heterogeneous Information

A Practice of Tourism Knowledge Graph Construction based on Heterogeneous Information Dinghe Xiao Hainan Sino-intelligent-Info Technology Ltd. xiaodinghe@e-zzx.com Nannan Wang Beijing University of Posts and Telecommunications wangnannan@bupt.edu.cn Jiangang Yu Hainan Sino-intelligent-Info Technology Ltd. yujiangang@e-zzx.com Chunhong Zhang� Beijing University of Posts and Telecommunications zhangch@bupt.edu.cn Jiaqi Wu Hainan Sino-intelligent-Info Technology Ltd. wujiaqi@e-zzx.com Abstract The increasing amount of semi-structured and unstructured data on tourism websites brings a need for information extraction (IE) so as to construct a Tourism-domain Knowledge Graph (TKG), which is helpful to manage tourism information and develop downstream applications such as tourism search engine, recommendation and Q & A. However, the existing TKG is de- ficient, and there are few open methods to promote the construction and widespread application of TKG. In this paper, we present a systematic framework to build a TKG for Hainan, collecting data from popular tourism websites and structuring it into triples. The data is multi-source and heterogeneous, which raises a great challenge for processing it. So we develop two pipelines of processing methods for semi-structured data and unstructured data respectively. We refer to tourism InfoBox for semi-structured knowledge extraction and leverage deep learning algo- rithms to extract entities and relations from unstructured travel notes, which are colloquial and high-noise, and then we fuse the extracted knowledge from two sources. Finally, a TKG with 13 entity types and 46 relation types is established, which totally contains 34,079 entities and 441,371 triples. The systematic procedure proposed by this paper can construct a TKG from tourism websites, which can further applied to many scenarios and provide detailed reference for the construction of other domain-specific knowledge graphs. 1 Introduction Tourism has become increasingly popular in people’s daily life. Before people set out to travel, they often need to make clear the travel guides and matters needing attention for their destinations. Nowadays, with the development of the Internet, many tourism websites have appeared and provide a variety of travel information, such as attractions, tickets, bus routes, travel guides, etc. However, there may be some errors in the miscellaneous information on the tourism websites, and information on different tourism websites may be inconsistent. As shown in screenshots of Sina Micro-Blog users’ blogs in Figure 1, there are still tourists who are worried about making travel strategies despite rich information on all kinds of tourism- related search engines. How to collect and integrate valuable tourism knowledge on websites is a very important issue. Recently, Knowledge Graph (KG) has received much attention and research interest in industry and academia. The KG utilizes a set of subject-predicate-object triplets to represent the diverse entities and their relations in real-world scenes, which are respectively represented as nodes and edges in the graph. The KG is a graph-based large-scale knowledge representation and integration method, which has been applied in various scenarios such as enterprise (Miao et al., 2015), medical (Rotmensch et al., 2017) and industry (Zhao et al., 2019). Naturally, we consider applying KG in the field of Tourism to integrate and organize relevant knowledge, so as …

Content truncated to 500 words