Brain Knowledge Graph Analysis Based on Complex Network Theory

	Brain Knowledge Graph Analysis Based on Complex Network Theory
	Zhu, Hongyin1 ; Zeng, Yi1,2 ; Wang, Dongsheng1 ; Xu, Bo 1,2
	2016-10
会议日期	October 13-16, 2016
会议地点	Omaha, Nebraska, USA
英文摘要	Domain knowledge about the brain is embedded in the literature over the whole scientific history. Researchers find there are intricate relationships among different cognitive functions, brain regions, brain diseases, neurons, protein, gene, neurotransmitters, etc. In order to integrate, synthesize, and analyze what we have known about the brain, the brain knowledge graph is constructed and released as part of the Linked Brain Data (LBD) project, to reveal the existing and potential relationships of brain related entities. However, there are some incorrect and missing relationships in the extracted relations, and researchers also cannot find the key topics overwhelmed in the massive relations. Some researchers analyze the properties of vertices based on the network topology, but they cannot verify and infer the potential relations. In order to address the above problems, we propose a framework which consists of 3 parts. Firstly, based on complex network theory, we adopt the embeddedness to verify the relations and infer the potential links. Secondly, we use the network topology of existing knowledge to build the self-relations graph. Finally, the structural holes theory from sociology is adopted to discover the key and core vertices in the whole brain knowledge graph and we recommend those topics to users. Compared with logic inference methods, our methods are lightweight and capable of processing large-scale knowledge efficiently. We test the results about relation verification and inference, and the result demonstrates the feasibility of our method.
内容类型	会议论文
源URL	[http://ir.ia.ac.cn/handle/173211/14354]
专题	自动化研究所_类脑智能研究中心
通讯作者	Zeng, Yi
作者单位	1.Institute of Automation, Chinese Academy of Sciences, Beijing, China 2.Center for Excellence in Brain Science and Intelligence Technology, Chinese Academy of Sciences, Shanghai, China
推荐引用方式 GB/T 7714	Zhu, Hongyin,Zeng, Yi,Wang, Dongsheng,et al. Brain Knowledge Graph Analysis Based on Complex Network Theory[C]. 见:. Omaha, Nebraska, USA. October 13-16, 2016.