type_probability: Topic model key term summaries
In MansMeg/topicmodeltoolbox: A Topic Modeling Toolbox for R

Description Usage Arguments Details Value References

View source: R/topic_summaries.R

The classical way of representing topics by order by the probability for a type within a topic.

type_probability(state, j, beta = 0)

topic_probability(state, j, beta = 0)

KR1(state, j, beta = 0)

KR2(state, j, beta = 0)

relevance(state, j, beta = 0, lambda = 0.6)

`state`	A tidy topic model state file
`j`	The number of top words to return
`beta`	Beta hyper parameter. Default is 0 (no smoothing).
`lambda`	Relevance weight. Default is 0.6.

To save space the calculations are done using a sparse format, only returning values for type-topic combination that exist in the model. This means that unless beta is set to 0, the returning probabilities will not sum to 1.

Not all reweighting schemes return a probability (such as KR1, KR2 and relevance)

Returns a data_frame with topic and top terms

Topic and keyword re-ranking for LDA-based topic modeling (2009) LDAvis: A method for visualizing and interpreting topics (2014)

MansMeg/topicmodeltoolbox documentation built on May 7, 2019, 2:45 p.m.