Revision as of 15:28, 24 August 2023 edit Citation bot (talk \| contribs) Bots 5,864,624 edits Alter: title. Add: chapter, doi-access. \| Use this bot. Report bugs. \| #UCB_CommandLine ← Previous edit		Revision as of 20:51, 2 October 2023 edit undo Blandamson (talk \| contribs) 5 edits m Reworded the sentence to help clarify that inferring latent structure in the data is a requirement for representing 'V' using less data rather than simply a side affect. Next edit →
Line 107: Usually the number of columns of {{math\|'''W'''}} and the number of rows of {{math\|'''H'''}} in NMF are selected so the product {{math\|'''WH'''}} will become an approximation to {{math\|'''V'''}}. The full decomposition of {{math\|'''V'''}} then amounts to the two non-negative matrices {{math\|'''W'''}} and {{math\|'''H'''}} as well as a residual {{math\|'''U'''}}, such that: {{math\|1='''V''' = '''WH''' + '''U'''}}. The elements of the residual matrix can either be negative or positive. When {{math\|'''W'''}} and {{math\|'''H'''}} are smaller than {{math\|'''V'''}} they become easier to store and manipulate. Another reason for factorizing {{math\|'''V'''}} into smaller matrices {{math\|'''W'''}} and {{math\|'''H'''}}, is that if one's isgoal ~~able~~is to approximately represent the elements of {{math\|'''V'''}} by significantly less data, then one has to infer some latent structure in the data. === Convex non-negative matrix factorization ===

Non-negative matrix factorization: Difference between revisions