Description
Cap prefixLength at four and add usedPrefix*scaling*(1-jaro) only when jaro is strictly above threshold.
A similarity term depends on the comparison universe, missing-value policy, treatment of repeated features, and normalization. A score can also have a caller-selected value when its mathematical denominator vanishes. Those conventions matter before interpreting a threshold or comparing methods.
Cap prefixLength at four and add usedPrefix*scaling*(1-jaro) only when jaro is strictly above threshold.
When to use this tool
Use Jaro Winkler Prefix Boost Cap when you need this particular comparison term, normalization, domain condition, or empty-result policy in a larger matching workflow.
How to use the tool and how it works
- Enter jaro, prefixLength, scaling, threshold. Use the units and conventions below.
- Run the tool. Correct any format, range, or cross-field input error before interpreting the output.
- Read score, usedPrefix. Use the output convention below.
Cap prefixLength at four and add usedPrefix*scaling*(1-jaro) only when jaro is strictly above threshold. Require bounded scores, nonnegative integer prefix length, and scaling from zero through one quarter.1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| jaro | Jaro | Required. Finite number; minimum 0; maximum 1. |
| prefixLength | Prefix Length | Required. Safe integer; minimum 0. |
| scaling | Scaling | Required. Finite number; minimum 0; maximum 0.25. |
| threshold | Threshold | Required. Finite number; minimum 0; maximum 1. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| score | Score | Required. Finite number. |
| usedPrefix | Prefix length capped at four, even if boosting is inactive. | Required. Safe integer. |
{
"input": {
"jaro": 0.8,
"prefixLength": 6,
"scaling": 0.1,
"threshold": 0.7
},
"output": {
"score": 0.88,
"usedPrefix": 4
}
}Limitations and assumptions
- UsedPrefix is returned even when no boost applies. No prefix or Jaro score is computed from strings, and the threshold is an explicit caller convention.
- The field tables state structural bounds; the mechanics also specify cross-field constraints. Malformed inputs and unsupported domains cause errors. A successful false flag, empty branch value, or zero placeholder has the meaning stated above and is distinct from an input error.
- Numeric calculations use binary floating point unless exact decimal-integer arithmetic is explicitly stated. No hidden measurement tolerance or decimal rounding is added; nonfinite or out-of-contract results are rejected. Signed zero is exported as zero.
Alternative or Complementary methods
Compare methods only after aligning feature definitions, missingness, empty policies, multiplicities, and normalizers. Use a complete comparison when the tool returns only one contribution or count check.
References
Similar or alternative tools
- Jaro Match Window One String Length
Return window=max(0,floor(max(firstLength,secondLength)/2)-1) for nonnegative integer lengths.
- Ranking Shared Prefix Credit Term
Take the first depth elements from each list, require each selected prefix unique, and return the number shared and credit=shared/depth.