US Library of congress

Machine learning model for automated cataloguing (concept prototype)

This is a work in progress
I'm currently working on completing this case study and should be available soon.

The short read

The US Library of Congress runs experiments to develop future service and system improvements. This experiment, run by the cataloguing office, was about exploring automated cataloguing of new e-books entering the LOC system.

By training a sample via machine learning, we found an effective way to automate the main aspects of cataloging by title and by author

We worked to:

  1. Optimise the main combinatons of algorithm inputs needed to establish an acceptable level of accuracy that would actually help the manual work of human Cataloguers at the LOC.
  2. Develop an interface skeleton that would work in terms of hierarchy and layout, to extract this information optimally.

↓ Read the full story below.

Case study

Case study coming soon.

My role

- Prototyping of the interface; information architecture.

Thanks to

Finlay McCourt (Machine Learning Enginneer).