Researcher Compass

What are Fingerprints?

Fingerprints are automatically generated key concepts derived from text data in Pure, the backend system of CityUHK Scholars. They create an index of weighted terms that provides a snapshot of the main topics covered in research outputs, researcher profiles, projects, and other records.

Using Natural Language Processing (NLP), the system scans text such as publication titles, abstracts, and descriptions, matching key concepts against scientific thesauri. It identifies relevant scientific areas and ranks concepts based on their frequency and location in the text, with higher weight given to concepts in titles. Multiple fingerprints can be aggregated by summing individual concept ranks and normalizing the results. The final fingerprints are displayed in CityUHK Scholars backend and portal as appropriate.

Which Content Type Generates Fingerprints?

Fingerprints are generated from various content types in Pure, and they update dynamically as new content is added:

• Research Outputs: Title, subtitle, and abstract
• Researcher Profiles: Aggregated from research interests, linked research outputs and prizes
• Projects: Aggregated from project descriptions, linked research outputs and awards
• Prizes: Title and description
• Research Units: Aggregated from all linked research outputs within academic units or organizations

Can I Manage Fingerprints?

• Add or Modify Fingerprints

Fingerprints are generated automatically using NLP, so manual addition or modification of fingerprints is not possible. However, you can influence which terms appear by including key concepts in your research interests within your researcher profile, as well as in the titles and abstracts/descriptions of your research works.

• Remove or Disable Fingerprints

If you find the fingerprints irrelevant or not representative of your research, you can remove specific concepts or disable fingerprints entirely for your researcher profile or records. Below are the steps on how to do this for researcher profiles.

1. Log in to the CityUHK Scholars backend Pure and click Edit profile.
2. Click Fingerprints in the left panel.
3. Hover over the unwanted fingerprint and click the red cross to remove it.
4. The removed fingerprint will move to the OMITTED CONCEPTS section. Click to restore it to the INCLUDED CONCEPTS section.
5. To disable all fingerprints or those under a subject, click the blue On button to switch it to Off.
6. Click Save to apply the changes.

For more information, please refer to Elsevier's guide What does the Fingerprint Engine do?.