|
|
A Study on Knowledge Graph-based Subject Knowledge Discovery in Life Sciences and Medicine
|
Hu Zhengyin,Liu Leilei,Dai Bing,Qin Xiaochu
|
(Chengdu Library and Information Center, Chinese Academy of Sciences, Chengdu 610041, China)
(Department of Library, Information and Archives Management, School of Economics and Management, University of Chinese Academy of Sciences, Beijing 100190, China)
(Guangzhou Regenerative Medicine and Health Guangdong Laboratory, Guangzhou 510700, China)
(Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences, Guangzhou 510530, China)
|
|
|
Abstract
[Objective] This paper aims to study the method of deep subject knowledge discovery by fusing multi-source heterogeneous data. [Methods] This paper constructs a SPO semantic network of literatures to form the core of a domain knowledge graph, implements multi-source heterogeneous data fusion through “entity alignment, concept level fusion and relationship fusion” to achieve the whole domain knowledge graph, and discovers deep subject knowledge based on the knowledge graph. Hematopoietic Stem Cell for Cancer Treatment (HSCCT) was chosen as a case study. [Results] This paper proposes a framework of knowledge graph-based subject knowledge discovery (KGSKD), which can fuse multi-source heterogeneous data fine-grandly from multi-dimension, define complex semantic relationships between the data, and support knowledge discovery techniques such as knowledge inference, pathfinder, and link prediction and so on natively. [Limitations] KGSKD has some limitations including data supersaturation, poor interpretability of knowledge discovery results and difficulty in communicating with domain experts. [Conclusions] KGSKD has the advantages of “richer data types”, “more comprehensive knowledge linkage”, “more advanced mining methods” and “deeper discovery results”, which can more effectively support the research and service of deep knowledge discovery in life sciences and medicine.
|
Published: 02 September 2020
|
|
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
|
Shared |
|
|
|
|
|
Discussed |
|
|
|
|