search

software-shao / Collective-Kernel-Learning

Matlab Code for CoKL, which was proposed in Clustering on Multiple Incomplete Datasets via Collective Kernel Learning. ICDM 2013
7 stars 4 forks source link

Generalization of CoKL for more than two incomplete data sets #1

Closed amirms closed 7 years ago

amirms commented 7 years ago

The CoKL algorithm as implemented here is restricted to two incomplete views. I would like to generalize the implementation for more than two views. The paper suggests as to how this should be done. Would you describe how the current implementation can be extended to generalize this?

software-shao commented 7 years ago

Hi amirms,

One way to extend this to multiple views is to apply this method pairwise or in a loop fashion. For example, if we have view 1,2,3,...,k. We can apply this method to view 1,2, then view 2,3, then view 3,4. ... then view k-1, k. until it converges.

However, this paper is my very early attempt for multi-view clustering problem. You can also read/use the code from my more recent papers [1] and [2]. [1] https://github.com/software-shao/Multi-Incomplete-view-Clustering Shao, W., He, L. and Philip, S.Y., 2015, September. Multiple Incomplete Views Clustering via Weighted Nonnegative Matrix Factorization with L_ {2, 1} Regularization. In Joint European Conference on Machine Learning and Knowledge Discovery in Databases (pp. 318-334). Springer International Publishing. [2] https://github.com/software-shao/online-multiview-clustering-with-incomplete-view Shao, W., He, L., Lu, C.T. and Philip, S.Y., 2016, December. Online multi-view clustering with incomplete views. In Big Data (Big Data), 2016 IEEE International Conference on (pp. 1012-1017). IEEE.