Abstract
Knowledge bases often utilize graphs as logical model. RDF-based knowledge bases (KB) are prime examples, as RDF (Resource Description Framework) does use graph as logical model. Graph databases are an emerging breed of NoSQL-type databases, offering graph as the logical model. Although there are specialized databases, the so-called triple stores, for storing RDF data, graph databases can also be promising candidates for storing knowledge. In this paper, we benchmark different graph database implementations loaded with Wikidata, a real-life, large-scale knowledge base. Graph databases come in all shapes and sizes, offer different APIs and graph models. Hence we used a measurement system, that can abstract away the API differences. For the modeling aspect, we made measurements with different graph encodings previously suggested in the literature, in order to observe the impact of the encoding aspect on the overall performance.
Subject
Computer Vision and Pattern Recognition,Software,Computer Science (miscellaneous),Electrical and Electronic Engineering,Information Systems and Management,Management Science and Operations Research,Theoretical Computer Science
Reference42 articles.
1. Blazegraph products. https://web.archive.org/web/20171125161035/https://www.blazegraph.com/product/
2. Blazegraph TinkerPop implementation. https://web.archive.org/web/20180611150556/https://github.com/blazegraph/tinkerpop3
3. DB-Engines ranking of graph DBMS. https://web.archive.org/web/20180911002043/https://db-engines.com/en/ranking/graph+dbms
4. Grakn.AI - the knowledge graph. https://web.archive.org/web20180918085112/http://www.grakn.ai/grakn-core
5. HypergraphDB - a graph database. https://web.archive.org/web/20180809121925/http://hypergraphdb.org/
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献