From-Below Approximations in Boolean Matrix Factorization: Geometry and New Algorithm

Abstract

We present new results on Boolean matrix factorization and a new algorithm based on these results. The results emphasize the significance of factorizations that provide from-below approximations of the input matrix. While the previously proposed algorithms do not consider the possibly different significance of different matrix entries, our results help measure such significance and suggest where to focus when computing factors. An experimental evaluation of the new algorithm on both synthetic and real data demonstrates its good performance in terms of good coverage by the first k factors as well as a small number of factors needed for exact decomposition and indicates that the algorithm outperforms the available ones in these terms. We also propose future research topics.

References in corpus (1)

• What is the dimension of your binary data?

Cited by in corpus (11)

• Factorizing formal contexts from closures of necessity operators
• FCA2VEC: Embedding Techniques for Formal Concept Analysis
• Generalized Inverses of Boolean Tensors via Einstein Product
• Recent Developments in Boolean Matrix Factorization
• Boolean Matrix Factorization via Nonnegative Auxiliary Optimization
• Noisy and Incomplete Boolean Matrix Factorizationvia Expectation Maximization
• What is the $\text{intrinsic}$ dimension of your binary data? -- and how to compute it quickly
• topFiberM: Scalable and Efficient Boolean Matrix Factorization
• Improved Algorithms for Exact and Approximate Boolean Matrix Decomposition
• Independent subcontexts and blocks of concept lattices. Definitions and relationships to decompose fuzzy contexts
• Multimodal Clustering for Community Detection