cshaowang / pic

Spectral Perturbation Meets Incomplete Multi-view Data

Spectral Perturbation Meets Incomplete Multi-view Data

This repo hosts the code for paper "Spectral Perturbation Meets Incomplete Multi-view Data, in IJCAI-2019". Beyond multi-view clustering, the paper tackles incomplete multi-view clustering (a.k.a. partial multi-view clustering), where a number of data instances are missing in certain views.

The proposed method shows a strong link between spectral perturbation and incomplete multi-view clustering. The key idea is to transfer the missing problem from data matrix to similarity matrix and reduce the spectral perturbation risk among different views while balancing all views to learn a consensus representation for the final clustering results.

The repo also hosts some baseline systems as we compared in the paper. We would like to thank the authors of the baseline systems for their codes. If any baseline systems cannot be licensed freely here, please drop me an email, so we can remove it from the collection.

If you find this repo useful, please kindly cite the paper below.

@inproceedings{wang2019spectral,
  title={Spectral Perturbation Meets Incomplete Multi-view Data},
  author={Wang, Hao and Zong, Linlin and Liu, Bing and Yang, Yan and Zhou, Wei},
  booktitle={Proceedings of the 28th International Joint Conference on Artificial Intelligence (IJCAI)},
  pages={3677--3683},
  year={2019}
}

pic's Issues

关于flower数据

博主您好！请问一下为什么我在网上下载的Flower数据里面是距离度量而不是特征？请问您有这个数据集吗？

Error when running PIC (quadprog)

Hi, first of all, thanks so much for your implementation on PIC and all the partial multi-view clustering methods.

However, when I am trying to run PIC, I run into the following error:

Error using quadprog (line 197)
The 'active-set' algorithm has been removed from quadprog. To avoid this error, choose a different algorithm: 'interior-point-convex' or 'trust-region-reflective'

The error is from PIC.p and I don't think I am able to edit the script on my own. Thanks for your help in advance :)