Effective Online Knowledge Graph Fusion
Recently, Web search engines have empowered their search with knowledge graphs to satisfy increasing demands of complex information needs about entities. Each engine offers an online knowledge graph service to display highly relevant information about the
- PDF / 1,812,091 Bytes
- 17 Pages / 439.37 x 666.142 pts Page_size
- 36 Downloads / 227 Views
st China University of Science and Technology, Shanghai 200237, China [email protected] 2 University of Aberdeen, Aberdeen, UK
Abstract. Recently, Web search engines have empowered their search with knowledge graphs to satisfy increasing demands of complex information needs about entities. Each engine offers an online knowledge graph service to display highly relevant information about the query entity in form of a structured summary called knowledge card. The cards from different engines might be complementary. Therefore, it is necessary to fuse knowledge cards from these engines to get a comprehensive view. Such a problem can be considered as a new branch of ontology alignment, which is actually an on-the-fly online data fusion based on the users’ needs. In this paper, we present the first effort to work on knowledge cards fusion. We propose a novel probabilistic scoring algorithm for card disambiguation to select the most likely entity a card should refer to. We then design a learning-based method to align properties from cards representing the same entity. Finally, we perform value deduplication to group equivalent values of the aligned properties as value clusters. The experimental results show that our approach outperforms the state of the art ontology alignment algorithms in terms of precision and recall.
1
Introduction
With the prevalence of entity search [1], a large portion of Web queries are to search entity related information. To support the ever growing information needs, search engines leverage public available knowledge bases such as Wikipedia and Freebase to build their own knowledge graphs. When submitting a query to Google (Bing or Yahoo!), the engine will provide a structured summary called knowledge card describing attributes of the given entity and relations with other entities. Such a card can be regarded as a query-based online form of the knowledge graph. Since a query might be ambiguous, it could return several cards corresponding to different real-world entities. Google returns three cards for the query “Fox” while Bing returns two more different cards. Even though the two cards represent the same entity, some property may just appear in one card. For example, only Google gives an attribute named “Daily sleep” in the card describing “Fox (animal)”. So it is necessary to fuse knowledge cards from various search engines automatically to provide a more comprehensive summary with This work was partially supported by the National Science Foundation of China (project No: 61402173), the Fundamental Research Funds for the Central Universities (Grant No: 22A201514045) and the EC MSC K-Drive project (286348). c Springer International Publishing Switzerland 2015 M. Arenas et al. (Eds.): ISWC 2015, Part I, LNCS 9366, pp. 286–302, 2015. DOI: 10.1007/978-3-319-25007-6 17
Effective Online Knowledge Graph Fusion
287
all important facts for a given entity. Also, search engines usually update their contents quickly so that the fused cards always contain up-to-date information. Knowledge cards fusion
Data Loading...
