Knowledge Graph-based Research Synthesis
This is a demonstration of using knowledge graph to represent and synthesize research findings from the literature into a cumulative knowledgebase. The source of the data is Chen, Duran, Sauerwald, Hitt, and van Essen (in-press).
Step 1. Construct and upload a taxonomy of concepts/constructs
This is typically done by senior domain experts who are curators of a knowledge domain. Systematic reviewers and senior journal editors are well positioned to be the knowledge source for this step.
Step 2. Construct and upload a map of causal influences among concepts/constructs
This is typically done by domain experts and theorists. Primary research studies, cause studies, and seasoned practitioners are well positioned to be the knowledge source for this step.
Step 3. Construct and upload a list of measures for the concepts/constructs
This is typically done by domain experts and empirical researchers. Primary empirical studies are well positioned to be the knowledge source for this step.
Step 4. Upload statistical findings and sample meta data
This is typically done by data scientists. Statistical and machine learning analyses are well positioned to be the knowledge source for this step.
Step 5. Now the full knowledge graph is completed, and it continues to cumulate and evolve as new uploads are made.
See XKEP for how users can teach (upload) and learn (query) knowledge from this knowledge graph database.