Skip to main content Skip to main navigation


Convex Non-negative Matrix Factorization in the Wild

Christian Thurau; Kristian Kersting; Christian Bauckhage
In: Wei Wang; Hillol Kargupta; Sanjay Ranka; Philip S. Yu; Xindong Wu (Hrsg.). ICDM 2009, The Ninth IEEE International Conference on Data Mining. IEEE International Conference on Data Mining (ICDM-2009), December 6-9, Miami, Flordia, USA, Pages 523-532, IEEE Computer Society, 2009.


Non-negative matrix factorization (NMF) has recently received a lot of attention in data mining, information retrieval, and computer vision. It factorizes a non-negative input matrix V into two non-negative matrix factors V = WH such that W describes "clusters" of the datasets. Analyzing genotypes, social networks, or images, it can be beneficial to ensure V to contain meaningful "cluster centroids", i.e., to restrict W to be convex combinations of data points. But how can we run this convex NMF in the wild, i.e., given millions of data points? Triggered by the simple observation that each data point is a convex combination of vertices of the data convex hull, we propose to restrict W further to be vertices of the convex hull. The benefits of this convex-hull NMF approach are twofold. First, the expected size of the convex hull of, for example, n random Gaussian points in the plane is ¿(¿log n), i.e., the candidate set typically grows much slower than the data set. Second, distance preserving low-dimensional embeddings allow one to compute candidate vertices efficiently. Our extensive experimental evaluation shows that convex-hull NMF compares favorably to convex NMF for large data sets both in terms of speed and reconstruction quality. Moreover, we show that our method can easily be applied to large-scale, real-world data sets, in our case consisting of 1.6 million images respectively 150 million votes on World of Warcraft ® guilds.

Weitere Links