Non-negative Matrix Factorization
Non-negative Matrix Factorization (NMF) is a mathematical technique used to decompose a non-negative matrix into two lower-dimensional non-negative matrices. This process helps in identifying hidden patterns or features within the data, making it useful in various applications like image processing, text mining, and recommendation systems.
In NMF, the original matrix is approximated by multiplying the two smaller matrices, often referred to as the basis and coefficient matrices. The non-negativity constraint ensures that the components are interpretable, as they can represent quantities like pixel intensities in images or word counts in documents, facilitating easier analysis and understanding of the underlying data structure.