mpmiExcraction: Extract words based on MPMI+BE alogrithm

In Flaretie/mpb: New Word Miner Based on modified point mutual information and branch entropy

Description

Extract words based on MPMI+BE alogrithm

Usage

mpmiExcraction(
  df,
  ChineseCharFilter = T,
  lambda = 0.25,
  optics,
  steps = 100,
  threshold,
  bayesianCutoff
)

Arguments

df	One column dataframe.
ChineseCharFilter	Drop non-Chinese characters. Default is True.
lambda	Weight distribution between PMI and BE, which will directly affect the results.
optics	Automatically optimize the result, select TURE if you don't know how to choose lambda.
steps	Number of iterations in the optimization process.
threshold	threshold for the final score.
bayesianCutoff	threshold for the bayesian probabilty value in word extension process

Value

List of extracted strings.

Examples

mpmiExcraction(drugnames_test,
               ChineseCharFilter = T,
               lambda = 0.25,
               optics = F,
               threshold = 30,
               bayesianCutoff = 0.6)

Flaretie/mpb documentation built on Jan. 24, 2020, 3:18 a.m.