refine_posterior: After generating probabilities for a list of pairs this will...

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

After generating probabilities for a list of pairs this will refine the probabilities

Description

After generating probabilities for a list of pairs this will refine the probabilities

Usage

refine_posterior(
  p,
  x_dat,
  y_dat,
  weights_nm = "x_weight",
  priori_delta = 0.01,
  args_x = list(col_nms = "company_name"),
  args_y = list(col_nms = "company_name"),
  token_types = "company_name",
  token_index = "",
  suffix = TOKEN_SUFFIX_DEFAULT,
  token_join_by = TOKEN_TOKEN_TYPE_VEC
)

Arguments

p	should be a dataframe like object with atleast 3 columns 'x', 'y' indicating the row names of the x_dat and y_dat that are being compared
x_dat	dataframe to check
y_dat	dataframe to check
weights_nm	Name of the column in pairs object that contains some kind of score or probability, which will be re scaled to a probability
priori_delta	passed to scale_to_prob. Default = 0.01
args_x	passed to token_links. Default list(col_nms = 'company_name')
args_y	passed to token_links. Default list(col_nms = 'company_name')
token_types	passed to token_links. Default 'company_name'
token_index	passed to token_links. Default is a blank string.
suffix	passed to token_links. Default is TOKEN_SUFFIX_DEFAULT

Value

returns a dataframe with x and y columns as well as priori and posterior from the original dataframe 'p' containing pairs with priori

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