Private Mining of Association Rules (2005) examines how parties can share association rules for data mining, without sharing data.
The authors develop a secure collaborative association rule mining protocol based upon a homomorphic encryption scheme.
Developing a similar approach for topic maps would be a nice doctoral project. Association rule data mining and the associated privacy concerns are well known. Combining those in a topic map context would be an interesting piece of work.
Author Bibliographies:
[…] Another Word For It Patrick Durusau on Topic Maps and Semantic Diversity « Private Mining of Association Rules […]
Pingback by Association Rules « Another Word For It — May 29, 2010 @ 6:28 am