tuning_algorithm: Active EM Parameter Tuning Algorithm

In activetext/activeR: a semi-supervised active learning algorithm for text classification.

Active EM Parameter Tuning Algorithm

Description

Tries to find optimal combination of pre-processing and active algorithm parameters.

Usage

tuning_algorithm(
  docs,
  doc_name,
  index_name,
  labelsName,
  measure = "accuracy",
  max_runs = 30,
  seed = 123,
  stem = c(F, T),
  ngrams = c(1:3),
  trimPct = c(0, 1e-04, 0.001, 0.01),
  min_doc_freq = c(1, 2, 3),
  idfWeight = c(F, T),
  removeStopWords = c(F, T),
  minChar = 1:4,
  weight = seq(0, 1, 0.1),
  bound = seq(0, 1, 0.1),
  initSize = 5,
  maxActive = 20,
  maxQuery = 1,
  queryType = c("basic_entropy", "random"),
  forceSweep = NA,
  dfmOut = F
)

activetext/activeR documentation built on May 31, 2024, 10:21 a.m.