eu.openaire.common.author

Class OrderedTokenAndAbbreviationsMatcher

java.lang.Object

eu.openaire.common.author.OrderedTokenAndAbbreviationsMatcher

public class OrderedTokenAndAbbreviationsMatcher
extends Object

Utility class for comparing author names using token-based matching and abbreviation handling.

This class provides methods to tokenize author names and compare them based on abbreviation recognition. It is useful for identifying name variations where part of the full name might be reordered or abbreviated.

Field Summary

Fields

Modifier and Type	Field and Description
static int	NUM_TOKEN_MAX_DIFF Maximum allowed difference in the number of tokens between two names for them to be comparable.
static Pattern	SPLIT_REGEX Regular expression pattern used to split names into tokens.

Constructor Summary

Constructors

Constructor and Description
OrderedTokenAndAbbreviationsMatcher()

Method Summary

Modifier and Type	Method and Description
static Optional<Double>	compare(String a1, String a2) Compares two author names using token-based matching and abbreviation handling.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

SPLIT_REGEX

public static final Pattern SPLIT_REGEX

Regular expression pattern used to split names into tokens.

The pattern matches spaces, punctuation symbols, and dashes, ensuring that names are split into meaningful components.

NUM_TOKEN_MAX_DIFF

public static int NUM_TOKEN_MAX_DIFF

Maximum allowed difference in the number of tokens between two names for them to be comparable.

Constructor Detail

OrderedTokenAndAbbreviationsMatcher

public OrderedTokenAndAbbreviationsMatcher()

Method Detail

compare

public static Optional<Double> compare(String a1,
                                       String a2)

Compares two author names using token-based matching and abbreviation handling.

The comparison follows these rules:

• Both names must have at least two tokens to be comparable.
• The number of tokens between the two names should not differ by more than NUM_TOKEN_MAX_DIFF.
• Matching considers both full-token matches and abbreviation-based matches.

The method returns an Optional containing a confidence score if a match is found, or an empty Optional if no match is identified.

Parameters:

a1 - The first author name.

a2 - The second author name.

Returns:

An Optional<Double> with a confidence score (1.0 if a match is found), or empty if no match.