get.factor.matrices

Extract the latent factor matrices for users (rows) and
columns (items) from a Poisson factorization model object, as returned
by function `poismf`.

Creates a non-negative low-rank approximate factorization of a sparse counts matrix by maximizing Poisson
likelihood with L1/L2 regularization (e.g. for implicit-feedback recommender systems or bag-of-words-based topic modeling)
(Cortes, (2018) <arXiv:1811.01908>), which usually leads to very sparse user and item factors (over 90% zero-valued).
Similar to hierarchical Poisson factorization (HPF), but follows an optimization-based approach with regularization
instead of a hierarchical prior, and is fit through gradient-based methods instead of variational inference.

David Cortes

poismf

Factorization of Sparse Counts Matrices Through Poisson
Likelihood

Jean-Sebastien Roy

Stephen Nash

get.factor.matrices function

<dl><dt>model</dt>
<dd>A Poisson factorization model, as produced by `poismf`.</dd>
<dt>add_names</dt>
<dd>Whether to add row names to the matrices if the indices
were internally remapped - they will only be so if the `X` passed to `poismf`
was a `data.frame`. Note that if passing `X` as `data.frame` with integer indices
to `poismf`, once row names are added, subsetting such matrix by an integer will
give the row at that position - that is, if you want to obtain the corresponding
row for ID=2 from `X` in `factors$A`, you need to use `factors$A["2", ]`, not
`factors$A[2, ]`.</dd></dl>

Arguments

Extract Latent Factor Matrices — get.factor.matrices

<dl>

<dt>model</dt>
<dd>A Poisson factorization model, as produced by `poismf`.</dd>


<dt>add_names</dt>
<dd>Whether to add row names to the matrices if the indices
were internally remapped - they will only be so if the `X` passed to `poismf`
was a `data.frame`. Note that if passing `X` as `data.frame` with integer indices
to `poismf`, once row names are added, subsetting such matrix by an integer will
give the row at that position - that is, if you want to obtain the corresponding
row for ID=2 from `X` in `factors$A`, you need to use `factors$A["2", ]`, not
`factors$A[2, ]`.</dd>

</dl>

get.factor.matrices: Extract Latent Factor Matrices

Description

Usage

Value

Arguments

Details

See Also