Detect and Flag Duplicates

Overview

Duplicate specimens are common in herbarium collections due to specimen exchange among institutions. Identifying and flagging them is crucial for avoiding data inflation and ensuring analytical accuracy. The function barroso_flag_duplicates() provides a fast and flexible way to detect potential duplicates based on collector name, collection number, taxon, and collection date.

Function: barroso_flag_duplicates()

Purpose

This function flags duplicate specimens by comparing fields that typically indicate specimen identity and collection event.

Fields used in comparison:

  • recordedBy (collector name)
  • recordNumber (collection number)
  • year, month, day
  • family, genus, specificEpithet

You can run this function as a standalone or as part of barroso_std().

Example

library(barRoso)

# Load sample specimen dataset
df <- read.csv("raw_herbarium_data.csv")

# Detect duplicates and add flag column
df_flagged <- barroso_flag_duplicates(df,
                                      rm_duplicates = FALSE)

# View flagged rows
subset(df_flagged, duplicate == TRUE)

Arguments

  • df: A data frame of herbarium records
  • rm_duplicates: Logical; if TRUE, duplicates will be removed (default: FALSE)

Output

Returns the same data frame with an additional column: - duplicate: Logical column indicating whether a row is a suspected duplicate

Best Practices

  • Run this step after standardizing fields using barroso_std()
  • Check for misspellings in recordedBy and inconsistencies in dates before trusting flags

Integration Tip

Use barroso_flag_duplicates() to visually inspect duplicates before removing them.

# To keep only unique records:
df_clean <- barroso_flag_duplicates(df, rm_duplicates = TRUE)

See Also