Package 'tilt.company.match'

Reports duplicates from manual matching outcome

Description

Function throws a descriptive error if a company from the loanbook is matched to > 1 company in the tilt db or reverse.

Usage

check_duplicated_relation(manually_matched)

Arguments

manually_matched

Tibble holding the result of the matching process, after the user has manually verified and matched the results

Value

Input manually_matched

---

Checks your loanbook is as we expect

Description

Checks your loanbook is as we expect

Usage

check_loanbook(loanbook)

Arguments

loanbook

A loanbook dataframe like demo_loanbook.

Value

Called for it's side effects. Returns loanbook invisibly.

Examples

library(vroom)
library(dplyr, warn.conflicts = FALSE)

loanbook <- vroom(example_file("demo_loanbook.csv"), show_col_types = FALSE)
check_loanbook(loanbook)

# Do you have the expected columns?
bad_name <- rename(loanbook, ids = id)
try(check_loanbook(bad_name))

# Do you have any duplicates in the column `id`?
bad_id <- bind_rows(loanbook, slice(loanbook, 1))
try(check_loanbook(bad_id))

# Do you have missing values (`NA`s) in non-nullable columns?
# styler: off
missing_id <- tribble(
  ~id,            ~company_name, ~postcode,  ~country, ~misc_info,
   NA, "John Meier's Groceries",   "55555", "germany",        "Y",
   11, "John Meier's Groceries",   "55555",  "norway",        "Y"
)
# styler: on
try(check_loanbook(missing_id))

---

Get the path to an example file

Description

Get the path to an example file

Usage

example_file(file)

Arguments

file	Name of the file.

Value

A path.

Examples

example_file("demo_loanbook.csv")

example_file("demo_tilt.csv")

example_file("demo_matched.csv")

---

Reports companies that were not matched in the loanbook

Description

Reports companies that were not matched in the loanbook

Usage

report_no_matches(loanbook, manually_matched)

Arguments

loanbook	Loanbook data set
manually_matched	Tibble holding the result of the matching process, after the user has manually selected and matched the companies in the loanbook with the tilt data set.

Value

not_matched_companies Tibble holding id and company name of the companies not matched by the tilt data set.

Examples

library(tibble)

loanbook <- tibble(id = 1:2, company_name = letters[id], irrelevant = "xyz")
loanbook

accepted <- tibble(id = 1:2, accept_match = c(TRUE, FALSE))
accepted

report_no_matches(loanbook, accepted)

# It's rigurous but fails with informative messages:
# The names of crucial columns must be as documented.
try(report_no_matches(loanbook, tibble(ids = 1, accept_match = TRUE)))

# The type of `accept_match` must be as documented.
try(report_no_matches(loanbook, tibble(id = 1, accept_match = "TRUE")))

---

Suggest matching companies in a loanbook and tilt datasets

Description

This function suggests that a company in your loanbook is the same as a company in the tilt dataset when the similarity between their names meets all of these conditions:

• It's the highest among all other candidates.

• It's above the value set in the argument suggestion_threshold.

• It's the only such highest value in the group defined by a combination of company_name x postcode – to avoid duplicates.

Usage

suggest_match(
  loanbook,
  tilt,
  eligibility_threshold = 0.75,
  suggestion_threshold = 0.9
)

Arguments

loanbook	A loanbook dataframe like demo_loanbook.
tilt	A tilt dataframe like demo_tilt.
eligibility_threshold	Minimum value of similarity to keep a candidate match. Values under it are most likely false positives and thus dropped. This drastically reduce the number of candidates you'll need to validate manually. We believe this benefit outweighs the potential loss of a few true positives.
suggestion_threshold	Value of similarity above which a match may be suggested.

Details

This function calculates the similarity between a standardized alias of the company_name from the loanbook and tilt datasets. The standardized alias makes real matches more likely by applying common best practices in names matching. Complete similarity corresponds to 1, and complete dissimilarity corresponds to 0.

The columns postcode and country affect the quality of the matches and the amount of manual-validation work ahead:

• If your loanbook has both postcode and country we match companies in that specific postcode and that specific country. You will likely match companies that are really the same (true positives) because it's unlikely that two companies with similar name will be located close to each other. This will cost you the minimum amount of manual-validation work ahead.

• If your loanbook lacks postcode but has country we match companies in that specific country but across every postcode. You will possibly match companies that are not really the same (false positives) but happen to have a similar name and are located in the same country. This will cost you additional manual-validation work ahead.

• If your loanbook has postcode but lacks country we match companies with the same postcode but across every country. You will possibly match companies that are not really the same (false positives) but happen to have a similar name and the same postcode. This will cost you additional manual-validation work ahead.

• If your loanbook lacks both postcode and country we match companies across the entire dataset. You will most likely match companies that are not really the same (false positives). This will cost you the greatest amount of additional manual-validation work ahead.

Value

A dataframe with:

• All the columns from the loanbook dataset.

• All the columns from the tilt dataset butthe columns id, company_name, postcode and country all get the suffix "_tilt".

• New columns:

  • company_alias

  • company_alias_tilt

  • similarity

  • suggest_match

  • accept_match. For each company in the loanbook matching candidates are arranged by descending similarity.

Examples

library(vroom)
loanbook <- vroom(example_file("demo_loanbook.csv"), show_col_types = FALSE)
tilt <- vroom(example_file("demo_tilt.csv"), show_col_types = FALSE)

suggest_match(loanbook, tilt)

---