NAME

Gender Guesser

SYNOPSIS

include_once 'genderguesser.php';

$name = '王杰';

$GenderGuesser = new GenderGuesser();
$GenderGuesser->loadLexicon('charlex_gender_full.lex');

$male_prob = $GenderGuesser->getMaleProbability($name);

echo "$name: ";
if ($male_prob === false) {
    echo 'Error';
} elseif ($male_prob >= 0.5) {
    echo 'Male, probability: ' . sprintf('%2.2f', $male_prob * 100) . '%';
} else {
    echo 'Female, probability: ' . sprintf('%2.2f', (1.0 - $male_prob) * 100) . '%';
}

DESCRIPTION

This class can guess the gender by name.

Generally speaking, name is related to gender. For example, the name contained "杰", "志", or "宏" is male name generally, the name contained "琬", "佩", or "琳" is female name generally. But it's difficult to guess the gender by the neutral character such as "文", "安", or "清".

NOTICE

The charset of data is UTF-8.

CLASS METHODS

__construct ( [ mixed options [, string lexicon_path ] ] )

Constructor. If the optional parameters options and lexicon_path are given, they will be used to initialize options and load lexicon.

bool setOptions ( mixed options )

Sets options. The argument can be an array or a string (space separated).

To switch off a option, add the prefix "-". Options:

s - The argument name contains family name.

bool loadLexicon ( string path )

Loads a lexicon. Returns true on success, or false on failure.

string getLexiconComment ( )

Returns the comment of the lexicon loaded.

float getMaleProbability ( string name )

Returns the probability that the specified name is a male name.

The value greater than 0.5 means it is a male name; the value less than 0.5 means it is a female name; and an exact 0.5 means it is a neutral name or the character cannot be found in the lexicon.

HISTORY

v0.10.0 (2014-10-08)

• Removed getOffset(), getTendencyByOffset() and getTendency() methods
• Added getMaleProbability() method by which to get the calculated probability
• Replaced getLexiconName() method by getLexiconComment()
• The default lexicon came from 17 million name-gender records (20 thousand records in previous version).
• The default lexicon has two versions, full and essential, containing different number of characters.

v0.05.0 (2012-03-18)

• Supports PHP 5 environment only, and the mbstring extension need to be enabled.
• Lexicon data stores in PHP serialization format now.
• The default lexicon came from 20,933 name-gender records (5,561 records in previous version).
• The default lexicon contains characters of both Simplified Chinese and Tranditional Chinese.

v0.02.0 (2005-11-12)

• The first public alpha version.

AUTHOR

2005-2014, Wudi <wudi@wudilabs.org>, Wudi Labs