HAPE: A programmable big knowledge graph platform

Ruqian LU,Chaoqun FEI,Chuanqing WANG,Shunfeng GAO,Han QIU,Songmao ZHANG,Cungen CAO
DOI: https://doi.org/10.1016/j.ins.2019.08.051
IF: 8.1
2020-01-01
Information Sciences
Abstract:Heaven Ape (HAPE) is an integrated big knowledge graph platform supporting the construction, management, and operation of large to massive scale knowledge graphs. Its current version described in this paper is a prototype, which consists of three parts: a big knowledge graph knowledge base, a knowledge graph browser on the client side, and a knowledge graph operating system on the server side. The platform is programmed in two high level scripting languages: JavaScript for programming the client side functions and Python for the server side functions. For making the programming more suitable for big knowledge processing and more friendly to knowledge programmers, we have developed two versions of knowledge scripting languages, namely K-script-c and K-script-s, for performing very high level knowledge programming of client resp. server side functions. HAPE borrows ideas from some well-known knowledge graph processing techniques and also invents some new ones as our creation. As an experiment, we transformed a major part of the DBpedia knowledge base and reconstructed it as a big knowledge graph. It works well in some application tests and provides acceptable efficiency.
What problem does this paper attempt to address?