KeywordExtractor

All Implemented Interfaces:


public interface KeywordExtractor

Implementations can extract keywords from text

Nested Class Summary

Nested Classes
Modifier and Type	Class	Description

Field Summary

Fields
Modifier and Type	Field	Description

Constructor Summary

Constructors
Constructor	Description

Enum Constant Summary

Enum Constants
Enum Constant	Description

Method Summary

Modifier and Type	Method	Description
`abstract Set<String>`	`extractKeywords(String text)`	Extract keywords from the given text
`Double`	`matchCountToScore(Integer matchCount)`	Converts a match count to a similarity score between 0 and 1.
`abstract Set<String>`	`getKeywords()`	All known keywords

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Constructor Detail
- Method Detail
  - extractKeywords
```
 abstract Set<String> extractKeywords(String text)
```
    Extract keywords from the given text
    
    Parameters:
    
    text - the text to extract keywords from
    
    Returns:
    
    the set of extracted keywords
  - matchCountToScore
```
 Double matchCountToScore(Integer matchCount)
```
    Converts a match count to a similarity score between 0 and 1.
    This default implementation uses an exponent of 0.4 to provide nonlinear scoring that is generous to partial matches while still giving diminishing returns. For example, matching 2 out of 15 keywords yields ~0.45 rather than 0.13, reflecting that even partial matches can be quite valuable.
    The formula is: (matchCount / totalKeywords)^0.4
    This approach aligns with information retrieval principles where early matches are most significant, but avoids being overly harsh on documents that match only a subset of keywords.
    
    Parameters:
    
    matchCount - the number of keywords that matched
    
    Returns:
    
    a similarity score from 0.0 to 1.0
  - getKeywords
```
 abstract Set<String> getKeywords()
```
    All known keywords