Tips to Build and Work with Knowledge Graphs -


Knowledge graphs are becoming an indispensable tool for information and knowledge managers, researchers, and executives. It is therefore essential to deploy industry best practices to get the best out of them.

Traditional information management approaches often rely on forcing information into controlled structures that impose assumptions. Therefore, mapping the relationship between two synonymous words can help modify the relationship at a later stage, if needed. This will prevent the potential difference in meaning from getting lost. Another best practice would be to make informed decisions about where data needs to be cleansed and where it needs to be mapped. These informed decisions will facilitate understanding which data to put in a knowledge graph and what to discard.

Furthermore, the value of knowledge graphs lies in the connections between objects. Therefore, it is essential to design the connections to answer questions you need to ask. Another good practice would be to maintain and show the provenance of the information in the knowledge graph to reassure users enough about its sources' reliability. Moreover, one of the significant advantages of knowledge graphs is that they are designed keeping users of the information in mind. It follows that the results should be presented in a way that the users would find useful.

Finally, it is vital to keep in mind that knowledge graphs should be living objects. Therefore, it is important to think carefully about the design of the graph and how things are mapped. Equally important is to allow users to provide feedback as it would help to evolve the knowledge graph and make it even more powerful.

Click here to read the original article published by Copyright Clearance Center (CCC).

STORY TOOLS

  • |
  • |

Brought to you by Scope e-Knowledge Center, an SPi Global Company, a trusted global partner for Digital Content Transformation Solutions, Knowledge Modeling (Taxonomies, Thesauri and Ontologies), Abstracting & Indexing (A&I), Metadata Enrichment and Entity Extraction.

Please give your feedback on this article or share a similar story for publishing by clicking here.


sponsor links

For banner adsĀ click here