wsea: Word Set Enrichment Analysis (WSEA) across corpora

In kkdey/WEAVER: Word Enrichment Analysis using Vector Representations

Description

Performs a word set enrichment across multiple corpora in relation to a topic of interest, specified by the header_set.

Usage

wsea(model_list, word_sets, header_set, model_names = NULL,
  add_groups = NULL, numwords = 1000, fgsea.control = list())

Arguments

model_list	A list of VectorSpace models for a number of corpora. The length of the list is equal to the number of corpora.
word_sets	A list of vector of words (word sets) whose enrichment is to be studied.
header_set	a word or a vector of words defining a topic of interest.
model_names	The names of the corpora that are compared. Defaults to numerical groups.
add_groups	A list of vectors indicating group labels that the user wants to compare. Using add_groups, the user can compare a set of corpora with another set of corpora beyod the 1v1 and 1vall comparisons the default version does.
numwords	The number of neighboring words to each word in the word sets pooled into the word set enrichment analysis.
fgsea.control	The control parameters for the GSEA model fitting using the fgsea() package.

Value

Produces 1v1, 1vall WSEA output for all copora (each output similar to the fgsea output) together with group based WSEA output as specified by the add_groups.

kkdey/WEAVER documentation built on May 8, 2019, 9:24 a.m.