find_posterior_all_evidence: t_dat should have been run through...

In csps-efpc/TokenLink: Joins two dataframes using tokens or like words

t_dat should have been run through find_posterior_positive_evidence_only() first to create the list of candidates

Description

t_dat should have been run through find_posterior_positive_evidence_only() first to create the list of candidates

Usage

find_posterior_all_evidence(
  t_dat,
  min_posterior = MIN_POSTERIOR_ALL_EVIDENCE,
  token_join_by = TOKEN_TOKEN_TYPE_VEC,
  positive_evidence_only = NULL,
  return_all = TRUE,
  priori_func = calculate_priori,
  ...
)

Arguments

t_dat	t_dat object
min_posterior	Filter results below this threshold Default = 0.01
token_join_by	What to join the all_tokens by. Default = TOKEN_TOKEN_TYPE_VEC,
positive_evidence_only,	NULL or a dataframe with at least two columns indicating the row_names to be checked in each of the datasets
return_all	If TRUE it returns the whole object if FALSE it just returns the dataframe. Default TRUE
priori_func	A function that will calculate the priori. Default calculate_priori
...	Not used

Examples

t_dat <-
  token_links(
     dat_x = readr::read_csv('https://tinyurl.com/2p8etjr6'),
     dat_y = readr::read_csv('https://tinyurl.com/2p8ap4ad'),
     args_x = list(col_nms = 'coname'),
     args_y = list(col_nms = 'companyName'),
     token_types = 'company_name',
     token_index = '',
     suffix = c('ceo', 'alb')
  )

results <-
  t_dat |>
    find_posterior_positive_evidence_only() |>
    find_posterior_all_evidence(return_all = FALSE)

csps-efpc/TokenLink documentation built on Feb. 10, 2023, 3:30 a.m.