# Factorization of Binary Matrices: Rank Relations, Uniqueness and Model Selection of Boolean Decomposition

@inproceedings{DeSantis2020FactorizationOB, title={Factorization of Binary Matrices: Rank Relations, Uniqueness and Model Selection of Boolean Decomposition}, author={Derek DeSantis and Erik West Skau and Duc P. Truong and Boian S. Alexandrov}, year={2020} }

The application of binary matrices are numerous. Representing a matrix as a mixture of a small collection of latent vectors via low-rank decomposition is often seen as an advantageous method to interpret and analyze data. In this work, we examine the factorizations of binary matrices using standard arithmetic (real and nonnegative) and logical operations (Boolean and Z2). We examine the relationships between the different ranks, and discuss when factorization is unique. In particular, we… Expand

#### References

SHOWING 1-10 OF 44 REFERENCES

Boolean decomposition of binary matrices using a post-nonlinear mixture approach

- Computer Science
- Signal Process.
- 2021

This work proposes a simple yet efficient algorithm for solving the proposed BMF problem based on multiplicative update rules and provides for the first time in the binary data literature, a necessary and sufficient condition for the uniqueness of the Boolean matrix factorization, as well as several other uniqueness results. Expand

Modeling by shortest data

- description. Automatica,
- 1978

Boolean Matrix Factorization via Nonnegative Auxiliary Optimization

- Computer Science, Mathematics
- IEEE Access
- 2021

The proofs for the equivalencies of the two solution spaces under the existence of an exact solution of the BMF problem are provided, and the nonincreasing property of the algorithm is also proven. Expand

Finding the Number of Latent Topics With Semantic Non-Negative Matrix Factorization

- Computer Science
- IEEE Access
- 2021

This paper considers a semantic-assisted non-negative matrix factorization (NMF) topics model, which is based on Kullback-Leibler(KL) divergence and integrated with a method for determining the number of latent topics, and demonstrates that it not only determines the correct number of topics, but also extracts topics with a high coherence and accurately classifies the documents of the corpus. Expand

From-below Boolean matrix factorization algorithm based on MDL

- Computer Science, Mathematics
- Adv. Data Anal. Classif.
- 2021

This paper proposes a novel from-below Boolean matrix factorization algorithm based on formal concept analysis that outperforms—from different standpoints—existing state-of-the-art BMF algorithms. Expand

Model order selection for approximate Boolean matrix factorization problem

- Computer Science
- Knowl. Based Syst.
- 2021

It is shown that the number of factors for DBP cannot be used in AFP, and a novel way is presented, reflecting the nature of AFP, how to establish the proper number of Factors. Expand

A neural network for determination of latent dimensionality in non-negative matrix factorization

- Physics, Computer Science
- ArXiv
- 2020

A supervised machine learning approach in combination with a recent method for model determination, called NMFk, to determine the number of hidden features automatically, and compared the accuracy of the method to ARD, AIC and stability-based methods. Expand

Recent Developments in Boolean Matrix Factorization

- Computer Science
- ArXiv
- 2020

A concise summary of the efforts of all of the communities studying Boolean Matrix Factorization is given and some open questions which in this opinion require further investigation are raised. Expand

A Practical Randomized CP Tensor Decomposition

- Mathematics, Computer Science
- SIAM J. Matrix Anal. Appl.
- 2018

The CANDECOMP/PARAFAC (CP) decomposition is a leading method for the analysis of multiway data and involves a series of alternating least squares algorithms. Expand

MDL4BMF: Minimum Description Length for Boolean Matrix Factorization

- Mathematics, Computer Science
- TKDD
- 2014

An existing algorithm for BMF is extended to use MDL to identify the best Boolean matrix factorization, analyze the complexity of the problem, and perform an extensive experimental evaluation to study its behavior. Expand