Fuzzyjoin

fuzzyjoin

Join tables by inexact matching.

String Matching

library(fuzzyjoin)

# Regex join
regex_left_join(df1, df2, by = c("name" = "pattern"))
regex_inner_join(df1, df2, by = "name")

# Fuzzy string matching (stringdist)
stringdist_left_join(df1, df2, by = "name", max_dist = 2)
stringdist_inner_join(df1, df2, by = "name", method = "jw", max_dist = 0.1)

# Methods: "osa", "lv", "dl", "hamming", "lcs", "qgram", "cosine", "jaccard", "jw"

Numeric Matching

# Difference join (within tolerance)
difference_left_join(df1, df2, by = "value", max_dist = 5)

# Distance join
distance_left_join(df1, df2, by = c("x", "y"), max_dist = 10)

Interval Matching

# Interval join (overlapping ranges)
interval_left_join(df1, df2, by = c("start", "end"))

# Genome-style intervals
genome_left_join(df1, df2, by = c("chr", "start", "end"))

Geographic Matching

# Geo join (within distance)
geo_left_join(df1, df2,
  by = c("lat", "lon"),
  max_dist = 10,
  unit = "km"
)

Custom Matching

# Fuzzy join with custom function
fuzzy_left_join(df1, df2,
  by = c("x" = "y"),
  match_fun = function(x, y) abs(x - y) `)
)

Semi and Anti Joins

# Fuzzy semi join (filter matches)
stringdist_semi_join(df1, df2, by = "name", max_dist = 2)

# Fuzzy anti join (filter non-matches)
stringdist_anti_join(df1, df2, by = "name", max_dist = 2)

All Join Types

# Available for all fuzzy methods:
# _inner_join, _left_join, _right_join, _full_join
# _semi_join, _anti_join

stringdist_inner_join(df1, df2, by = "name")
stringdist_full_join(df1, df2, by = "name")
regex_semi_join(df1, df2, by = "name")
difference_anti_join(df1, df2, by = "value")

Examples

# Match company names with typos
companies =`, `<=`)
)

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: LeoLin990405
• Source: LeoLin990405/r-analytics-skill
• License: MIT

Install and usage instructions live in the source repository linked above.