Skip to contents

Clean a GBIF download of suspicious records using complex rules

Source: R/clean_download.R
clean_download.Rd

Removes records that have been marked as suspicious by annotation users using complex rules. Only filters based on SUSPICIOUS annotations, ignoring other annotation types. If project_id is provided, only rules belonging to those project(s) will be used. When use_higher_taxonomy is TRUE, rules can match records at any taxonomic level present in the download.

Usage

clean_download(
  d,
  rm_suspicious = TRUE,
  handle_conflicts = "favor_suspicious",
  project_id = NULL,
  use_higher_taxonomy = TRUE,
  min_support = NULL,
  exclude_contested = FALSE
)

Arguments

d

a GBIF download in DWCA format

rm_suspicious

removes records with suspicious annotations

handle_conflicts

how to handle conflicting annotations

project_id

optional project ID(s) to filter rules by specific project(s). Can be a single ID or vector of IDs.

use_higher_taxonomy

logical; if TRUE (default), rules will match records based on higher taxonomic ranks. The function checks for matches against kingdomKey, phylumKey, classKey, orderKey, familyKey, genusKey, speciesKey, and taxonKey columns in the download data. For example, a rule created for taxonKey 212 (Aves/Birds) will match all bird species records that have classKey = 212.

min_support

integer; minimum number of supports required for a rule to be applied. NULL (default) means no minimum. Only rules with at least this many supports will be used.

exclude_contested

logical; if TRUE, exclude any rules that have been contested (downvoted). Default is FALSE.

Value

A cleaned download `data.frame()`

Examples