"Must-link" and "Cannot-link" in semi-supervised clustering papers and good book, "Web Data Mining"

[Kazunari SUGIYAMA]

Dear all

The followings are some information related to today's paper:

(1) "Must-link" and "Cannot-link" in semi-supervised clustering
The concept of "must-link" and "cannot-link" are
originally used in semi-supervised clustering.

Semi-supervised clustering is classified into
"constraint-based approach" and "distance-based approach."

[Constraint-based approach]
K. Wagstaff and C. Cardie:
"Clustering with Instance-level Constraints."
In Proc. of the 17th International Conference on Machine Learning
(ICML2000), pages 1103 - 1110, 2000.

K. Wagstaff, S. Rogers, and S. Schroedl:
"Constrained K-means Clustering with Background Knowledge."
In Proc. of the 18th International Conference on Machine Learning
(ICML2001), pages 577 - 584, 2001.


[Distance-based approach]
D. Klein, S. D. Kamvar, and C. D. Manning:
"From Instance-level Constraints to Space-level Constraints:
Making the Most of Prior Knowledge in Data Clustering."
In Proc. of the 19th International Conference on Machine Learning
(ICML2002), pages 307-314, 2002.

E.P. Xing, A.Y. Ng, M.I. Jordan, S.J. Russell:
"Distance Metric Learning with Application to Clustering with
Side-Information."
Advances in Neural Information Processing Systems 15,
pages 521 - 528, 2003.

A. Bar-Hillel, T. Hertz, N. Shental:
"Learning Distance Functions Using Equivalence Relations."
In: Proc. of the 20th International Conference on Machine Learning
(ICML2003), pages 577-584, 2003.


The comparative approaches with today's papers
(DF-LDA, MC-LDA, and GK-LDA) may review
the semi-supervised clustering.


(2) Book "Web Data Mining" authored by B. Liu
Today's paper cites it in "References" section.
This is a good book to learn recent techniques:



I hope that the information above would be helpful!

Best regards,
Kazunari Sugiyama