Inductive Matrix Completion (IMC) is an algorithm for recommender systems with side-information of users and items. The IMC formulation incorporates features associated with rows (users) and columns (items) in matrix completion, so that it enables predictions for users or items that were not seen during training, and for which only features are known but no dyadic information (such as ratings or linkages).

Download

Please download the IMC software from the following links:

• C/C++ version with Python and Matlab (mex file) interfaces [Download]
• Matlab only version (.m file) [Download]

Citation

This software is released under the BSD License but please acknowledge its use with a citation to at least one of the following publications:

• Large-scale Multi-label Learning with Missing Labels (pdf, software)
  H. Yu, P. Jain, P. Kar, I. Dhillon.
  In International Conference on Machine Learning (ICML), pp. 593–601, June 2014.
• Inductive matrix completion for predicting gene-disease associations (pdf, software)
  N. Natarajan, I. Dhillon.
  Bioinformatics 30(12), pp. i60-i68, June 2014.

Related Publications

• Matrix Completion with Noisy Side Information (pdf, software)
  K. Chiang, C. Hsieh, I. Dhillon.
  In Neural Information Processing Systems (NIPS), pp. 3447–3455, December 2015. (Spotlight)
• Tumblr Blog Recommendation with Boosted Inductive Matrix Completion (pdf, software)
  D. Shin, S. Cetintas, K. Lee, I. Dhillon.
  In ACM Conference on Information and Knowledge Management (CIKM), pp. 203–212, October 2015.
• Efficient Matrix Sensing Using Rank-1 Gaussian Measurements (pdf, software)
  K. Zhong, P. Jain, I. Dhillon.
  In International Conference on Algorithmic Learning Theory (ALT), pp. 3-18, October 2015.
• PU Learning for Matrix Completion (pdf, software)
  C. Hsieh, N. Natarajan, I. Dhillon.
  In International Conference on Machine Learning (ICML), pp. 2445-2453, July 2015.
• Large-scale Multi-label Learning with Missing Labels (pdf, software)
  H. Yu, P. Jain, P. Kar, I. Dhillon.
  In International Conference on Machine Learning (ICML), pp. 593–601, June 2014.
• Inductive matrix completion for predicting gene-disease associations (pdf, software)
  N. Natarajan, I. Dhillon.
  Bioinformatics 30(12), pp. i60-i68, June 2014.

Developers: Hsiang-Fu Yu   Nagarajan Natarajan   Donghyuk Shin   

Associated Projects: Gene-Disease Prediction: A Link Prediction Approach   , Large-scale Multi-label Learning