Skip to contents

Set master digital voucher

Source: R/set_digital_voucher.R
set_digital_voucher.Rd

Grouping Duplicates and Selecting the Digital Voucher

1. Handling Complete Collection Event Keys

  • Unique collection events often generate multiple duplicate GBIF records. One duplicate is designated as the master digital voucher, integrating data from other duplicates.

  • When the collection event key is complete, duplicates are grouped and parsed based on record completeness and geospatial quality:

    • Record Completeness : Assessed using data-quality scores for fields: recordedBy, recordNumber, year, institutionCode, catalogNumber, locality, municipality, countryCode, stateProvince, and fieldNotes.

    • Geospatial Quality : Ranked via a score derived from geospatial issues in EnumOccurrenceIssue (GBIF table).

  • The duplicate with the highest total score (sum of record completeness + geospatial quality) is assigned as the master voucher. Missing data from other duplicates are merged into this voucher.

2. Handling Incomplete Collection Event Keys

  • When the collection event key is incomplete, duplicates cannot be parsed. Each record is treated as a unique collection event, with no duplicates identified.

  • Record completeness and geospatial quality are still evaluated as described below to assess data integrity.

3. Quality Score Calculation

UltraGBIF_digital_voucher : The duplicate with the highest total score, calculated as

Total Score = Record Completeness + Quality of Geospatial Information

4. Record Completeness Calculation

  • Measured as the sum of binary flags (TRUE = 1, FALSE = 0) for the following fields:

    • Is recordedBy present?

    • Is recordNumber present?

    • Is year present?

    • Is institutionCode present?

    • Is catalogNumber present?

    • Is locality present?

    • Is municipality present?

    • Is countryCode present?

    • Is stateProvince present?

    • Is fieldNotes present?

5. Geospatial Quality Calculation

  • Based on GBIF geospatial issues in EnumOccurrenceIssue, classified into three categories:

    • No Impact : Issues not affecting coordinate accuracy (selection_score = -1).

    • Potential Impact : Issues that may affect coordinate accuracy (selection_score = -3).

    • Exclusion : Records with severe issues (selection_score = -9) are excluded.

This optimized version eliminates O(n²) complexity by using vectorized data.table operations.

Usage

set_digital_voucher(occ_import = NA, names.checked = NA, collection_key = NA)

Arguments

occ_import

imported GBIF records

names.checked

your checked taxon names from check_occ_name

collection_key

your collection mark from generate_collection_mark

Value

A list with duration and 2 data.table: "occ_digital_voucher" for all data processing fields and "occ_results" for only result fields.

Details

UltraGBIF_duplicates_grouping_status :

  • groupable: Complete key, duplicates parsed.

  • not groupable: no recordedBy and no recordNumber: Missing both fields.

  • not groupable: no recordNumber: Missing recordNumber.

  • not groupable: no recordedBy: Missing recordedBy.

UltraGBIF_num_duplicates: Integer count of duplicate records.

UltraGBIF_duplicates: TRUE if duplicates exist, FALSE otherwise.

UltraGBIF_non_groupable_duplicates: TRUE if duplicates cannot be grouped, FALSE otherwise.

Examples

# \donttest{
help(set_digital_voucher)
#> ℹ Rendering development documentation for "set_digital_voucher"
# }