[英]Convert a plural noun to singular
這是使用 SimpleNLG Java API 完成的
我想將“精靈”轉換為精靈。 下面的代碼單數轉復數，如何修改復數轉單數？
final XMLLexicon xmlLexicon = new XMLLexicon();
final WordElement word = xmlLexicon.getWord("elves", LexicalCategory.NOUN);
final InflectedWordElement pluralWord = new InflectedWordElement(word);
pluralWord.setPlural(true);
final Realiser realiser = new Realiser(xmlLexicon);
System.out.println(realiser.realise(pluralWord));
在這個 API 中顯然沒有setSingular()方法（我真的很依賴那個，我認為這有點有趣，沒有這樣的東西。）而且從 V4 開始也沒有setPlural()方法.
[1] 請注意，在 SimpleNLG V4 中，沒有詞典方法可以直接獲取單詞的屈折變體； 換句話說，在 SimpleNLG V3 getPlural()、getPastParticiple() 等方法的 V4 中沒有等效項。 在 V4 中可以計算詞的屈折變體，但是過程比較復雜：基本上我們需要圍繞基本形式創建一個 InflectedWordElement，給這個 InflectedWordElement 添加適當的特征，然后實現它。
我認為這可能會奏效：（我沒有測試它，因為我現在沒有時間。）
final XMLLexicon xmlLexicon = new XMLLexicon();
final WordElement word = xmlLexicon.getWord("elves", LexicalCategory.NOUN);
final InflectedWordElement singularWord = new InflectedWordElement(word);
WordElement sw = singularWord.getBaseWord();
final Realiser realiser = new Realiser(xmlLexicon);
System.out.println(realiser.realise(sw));
這段代碼幫助了我：
/*
* JBoss DNA (http://www.jboss.org/dna)
* See the COPYRIGHT.txt file distributed with this work for information
* regarding copyright ownership. Some portions may be licensed
* to Red Hat, Inc. under one or more contributor license agreements.
* See the AUTHORS.txt file in the distribution for a full listing of
* individual contributors.
*
* JBoss DNA is free software. Unless otherwise indicated, all code in JBoss DNA
* is licensed to you under the terms of the GNU Lesser General Public License as
* published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* JBoss DNA is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this software; if not, write to the Free
* Software Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
* 02110-1301 USA, or see the FSF site: http://www.fsf.org.
*/
import java.util.HashSet;
import java.util.LinkedList;
import java.util.Set;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
/**
* Transforms words to singular, plural, humanized (human readable), underscore, camel case, or ordinal form. This is inspired by
* the <a href="http://api.rubyonrails.org/classes/Inflector.html">Inflector</a> class in <a
* href="http://www.rubyonrails.org">Ruby on Rails</a>, which is distributed under the <a
* href="http://wiki.rubyonrails.org/rails/pages/License">Rails license</a>.
*
* @author Randall Hauch
*/
public class Inflector {
protected static final Inflector INSTANCE = new Inflector();
public static final Inflector getInstance() {
return INSTANCE;
}
protected class Rule {
protected final String expression;
protected final Pattern expressionPattern;
protected final String replacement;
protected Rule( String expression,
String replacement ) {
this.expression = expression;
this.replacement = replacement != null ? replacement : "";
this.expressionPattern = Pattern.compile(this.expression, Pattern.CASE_INSENSITIVE);
}
/**
* Apply the rule against the input string, returning the modified string or null if the rule didn't apply (and no
* modifications were made)
*
* @param input the input string
* @return the modified string if this rule applied, or null if the input was not modified by this rule
*/
protected String apply( String input ) {
Matcher matcher = this.expressionPattern.matcher(input);
if (!matcher.find()) return null;
return matcher.replaceAll(this.replacement);
}
@Override
public int hashCode() {
return expression.hashCode();
}
@Override
public boolean equals( Object obj ) {
if (obj == this) return true;
if (obj != null && obj.getClass() == this.getClass()) {
final Rule that = (Rule)obj;
if (this.expression.equalsIgnoreCase(that.expression)) return true;
}
return false;
}
@Override
public String toString() {
return expression + ", " + replacement;
}
}
private LinkedList<Rule> plurals = new LinkedList<Rule>();
private LinkedList<Rule> singulars = new LinkedList<Rule>();
/**
* The lowercase words that are to be excluded and not processed. This map can be modified by the users via
* {@link #getUncountables()}.
*/
private final Set<String> uncountables = new HashSet<String>();
public Inflector() {
initialize();
}
protected Inflector( Inflector original ) {
this.plurals.addAll(original.plurals);
this.singulars.addAll(original.singulars);
this.uncountables.addAll(original.uncountables);
}
@Override
public Inflector clone() {
return new Inflector(this);
}
// ------------------------------------------------------------------------------------------------
// Usage functions
// ------------------------------------------------------------------------------------------------
/**
* Returns the plural form of the word in the string.
*
* Examples:
*
* <pre>
* inflector.pluralize("post") #=> "posts"
* inflector.pluralize("octopus") #=> "octopi"
* inflector.pluralize("sheep") #=> "sheep"
* inflector.pluralize("words") #=> "words"
* inflector.pluralize("the blue mailman") #=> "the blue mailmen"
* inflector.pluralize("CamelOctopus") #=> "CamelOctopi"
* </pre>
*
*
*
* Note that if the {@link Object#toString()} is called on the supplied object, so this method works for non-strings, too.
*
*
* @param word the word that is to be pluralized.
* @return the pluralized form of the word, or the word itself if it could not be pluralized
* @see #singularize(Object)
*/
public String pluralize( Object word ) {
if (word == null) return null;
String wordStr = word.toString().trim();
if (wordStr.length() == 0) return wordStr;
if (isUncountable(wordStr)) return wordStr;
for (Rule rule : this.plurals) {
String result = rule.apply(wordStr);
if (result != null) return result;
}
return wordStr;
}
public String pluralize( Object word,
int count ) {
if (word == null) return null;
if (count == 1 || count == -1) {
return word.toString();
}
return pluralize(word);
}
/**
* Returns the singular form of the word in the string.
*
* Examples:
*
* <pre>
* inflector.singularize("posts") #=> "post"
* inflector.singularize("octopi") #=> "octopus"
* inflector.singularize("sheep") #=> "sheep"
* inflector.singularize("words") #=> "word"
* inflector.singularize("the blue mailmen") #=> "the blue mailman"
* inflector.singularize("CamelOctopi") #=> "CamelOctopus"
* </pre>
*
*
*
* Note that if the {@link Object#toString()} is called on the supplied object, so this method works for non-strings, too.
*
*
* @param word the word that is to be pluralized.
* @return the pluralized form of the word, or the word itself if it could not be pluralized
* @see #pluralize(Object)
*/
public String singularize( Object word ) {
if (word == null) return null;
String wordStr = word.toString().trim();
if (wordStr.length() == 0) return wordStr;
if (isUncountable(wordStr)) return wordStr;
for (Rule rule : this.singulars) {
String result = rule.apply(wordStr);
if (result != null) return result;
}
return wordStr;
}
/**
* Converts strings to lowerCamelCase. This method will also use any extra delimiter characters to identify word boundaries.
*
* Examples:
*
* <pre>
* inflector.lowerCamelCase("active_record") #=> "activeRecord"
* inflector.lowerCamelCase("first_name") #=> "firstName"
* inflector.lowerCamelCase("name") #=> "name"
* inflector.lowerCamelCase("the-first_name",'-') #=> "theFirstName"
* </pre>
*
*
*
* @param lowerCaseAndUnderscoredWord the word that is to be converted to camel case
* @param delimiterChars optional characters that are used to delimit word boundaries
* @return the lower camel case version of the word
* @see #underscore(String, char[])
* @see #camelCase(String, boolean, char[])
* @see #upperCamelCase(String, char[])
*/
public String lowerCamelCase( String lowerCaseAndUnderscoredWord,
char... delimiterChars ) {
return camelCase(lowerCaseAndUnderscoredWord, false, delimiterChars);
}
/**
* Converts strings to UpperCamelCase. This method will also use any extra delimiter characters to identify word boundaries.
*
* Examples:
*
* <pre>
* inflector.upperCamelCase("active_record") #=> "SctiveRecord"
* inflector.upperCamelCase("first_name") #=> "FirstName"
* inflector.upperCamelCase("name") #=> "Name"
* inflector.lowerCamelCase("the-first_name",'-') #=> "TheFirstName"
* </pre>
*
*
*
* @param lowerCaseAndUnderscoredWord the word that is to be converted to camel case
* @param delimiterChars optional characters that are used to delimit word boundaries
* @return the upper camel case version of the word
* @see #underscore(String, char[])
* @see #camelCase(String, boolean, char[])
* @see #lowerCamelCase(String, char[])
*/
public String upperCamelCase( String lowerCaseAndUnderscoredWord,
char... delimiterChars ) {
return camelCase(lowerCaseAndUnderscoredWord, true, delimiterChars);
}
/**
* By default, this method converts strings to UpperCamelCase. If the <code>uppercaseFirstLetter</code> argument to false,
* then this method produces lowerCamelCase. This method will also use any extra delimiter characters to identify word
* boundaries.
*
* Examples:
*
* <pre>
* inflector.camelCase("active_record",false) #=> "activeRecord"
* inflector.camelCase("active_record",true) #=> "ActiveRecord"
* inflector.camelCase("first_name",false) #=> "firstName"
* inflector.camelCase("first_name",true) #=> "FirstName"
* inflector.camelCase("name",false) #=> "name"
* inflector.camelCase("name",true) #=> "Name"
* </pre>
*
*
*
* @param lowerCaseAndUnderscoredWord the word that is to be converted to camel case
* @param uppercaseFirstLetter true if the first character is to be uppercased, or false if the first character is to be
* lowercased
* @param delimiterChars optional characters that are used to delimit word boundaries
* @return the camel case version of the word
* @see #underscore(String, char[])
* @see #upperCamelCase(String, char[])
* @see #lowerCamelCase(String, char[])
*/
public String camelCase( String lowerCaseAndUnderscoredWord,
boolean uppercaseFirstLetter,
char... delimiterChars ) {
if (lowerCaseAndUnderscoredWord == null) return null;
lowerCaseAndUnderscoredWord = lowerCaseAndUnderscoredWord.trim();
if (lowerCaseAndUnderscoredWord.length() == 0) return "";
if (uppercaseFirstLetter) {
String result = lowerCaseAndUnderscoredWord;
// Replace any extra delimiters with underscores (before the underscores are converted in the next step)...
if (delimiterChars != null) {
for (char delimiterChar : delimiterChars) {
result = result.replace(delimiterChar, '_');
}
}
// Change the case at the beginning at after each underscore ...
return replaceAllWithUppercase(result, "(^|_)(.)", 2);
}
if (lowerCaseAndUnderscoredWord.length() < 2) return lowerCaseAndUnderscoredWord;
return "" + Character.toLowerCase(lowerCaseAndUnderscoredWord.charAt(0))
+ camelCase(lowerCaseAndUnderscoredWord, true, delimiterChars).substring(1);
}
/**
* Makes an underscored form from the expression in the string (the reverse of the {@link #camelCase(String, boolean, char[])
* camelCase} method. Also changes any characters that match the supplied delimiters into underscore.
*
* Examples:
*
* <pre>
* inflector.underscore("activeRecord") #=> "active_record"
* inflector.underscore("ActiveRecord") #=> "active_record"
* inflector.underscore("firstName") #=> "first_name"
* inflector.underscore("FirstName") #=> "first_name"
* inflector.underscore("name") #=> "name"
* inflector.underscore("The.firstName") #=> "the_first_name"
* </pre>
*
*
*
* @param camelCaseWord the camel-cased word that is to be converted;
* @param delimiterChars optional characters that are used to delimit word boundaries (beyond capitalization)
* @return a lower-cased version of the input, with separate words delimited by the underscore character.
*/
public String underscore( String camelCaseWord,
char... delimiterChars ) {
if (camelCaseWord == null) return null;
String result = camelCaseWord.trim();
if (result.length() == 0) return "";
result = result.replaceAll("([A-Z]+)([A-Z][a-z])", "$1_$2");
result = result.replaceAll("([a-z\\d])([A-Z])", "$1_$2");
result = result.replace('-', '_');
if (delimiterChars != null) {
for (char delimiterChar : delimiterChars) {
result = result.replace(delimiterChar, '_');
}
}
return result.toLowerCase();
}
/**
* Returns a copy of the input with the first character converted to uppercase and the remainder to lowercase.
*
* @param words the word to be capitalized
* @return the string with the first character capitalized and the remaining characters lowercased
*/
public String capitalize( String words ) {
if (words == null) return null;
String result = words.trim();
if (result.length() == 0) return "";
if (result.length() == 1) return result.toUpperCase();
return "" + Character.toUpperCase(result.charAt(0)) + result.substring(1).toLowerCase();
}
/**
* Capitalizes the first word and turns underscores into spaces and strips trailing "_id" and any supplied removable tokens.
* Like {@link #titleCase(String, String[])}, this is meant for creating pretty output.
*
* Examples:
*
* <pre>
* inflector.humanize("employee_salary") #=> "Employee salary"
* inflector.humanize("author_id") #=> "Author"
* </pre>
*
*
*
* @param lowerCaseAndUnderscoredWords the input to be humanized
* @param removableTokens optional array of tokens that are to be removed
* @return the humanized string
* @see #titleCase(String, String[])
*/
public String humanize( String lowerCaseAndUnderscoredWords,
String... removableTokens ) {
if (lowerCaseAndUnderscoredWords == null) return null;
String result = lowerCaseAndUnderscoredWords.trim();
if (result.length() == 0) return "";
// Remove a trailing "_id" token
result = result.replaceAll("_id$", "");
// Remove all of the tokens that should be removed
if (removableTokens != null) {
for (String removableToken : removableTokens) {
result = result.replaceAll(removableToken, "");
}
}
result = result.replaceAll("_+", " "); // replace all adjacent underscores with a single space
return capitalize(result);
}
/**
* Capitalizes all the words and replaces some characters in the string to create a nicer looking title. Underscores are
* changed to spaces, a trailing "_id" is removed, and any of the supplied tokens are removed. Like
* {@link #humanize(String, String[])}, this is meant for creating pretty output.
*
* Examples:
*
* <pre>
* inflector.titleCase("man from the boondocks") #=> "Man From The Boondocks"
* inflector.titleCase("x-men: the last stand") #=> "X Men: The Last Stand"
* </pre>
*
*
*
* @param words the input to be turned into title case
* @param removableTokens optional array of tokens that are to be removed
* @return the title-case version of the supplied words
*/
public String titleCase( String words,
String... removableTokens ) {
String result = humanize(words, removableTokens);
result = replaceAllWithUppercase(result, "\\b([a-z])", 1); // change first char of each word to uppercase
return result;
}
/**
* Turns a non-negative number into an ordinal string used to denote the position in an ordered sequence, such as 1st, 2nd,
* 3rd, 4th.
*
* @param number the non-negative number
* @return the string with the number and ordinal suffix
*/
public String ordinalize( int number ) {
int remainder = number % 100;
String numberStr = Integer.toString(number);
if (11 <= number && number <= 13) return numberStr + "th";
remainder = number % 10;
if (remainder == 1) return numberStr + "st";
if (remainder == 2) return numberStr + "nd";
if (remainder == 3) return numberStr + "rd";
return numberStr + "th";
}
// ------------------------------------------------------------------------------------------------
// Management methods
// ------------------------------------------------------------------------------------------------
/**
* Determine whether the supplied word is considered uncountable by the {@link #pluralize(Object) pluralize} and
* {@link #singularize(Object) singularize} methods.
*
* @param word the word
* @return true if the plural and singular forms of the word are the same
*/
public boolean isUncountable( String word ) {
if (word == null) return false;
String trimmedLower = word.trim().toLowerCase();
return this.uncountables.contains(trimmedLower);
}
/**
* Get the set of words that are not processed by the Inflector. The resulting map is directly modifiable.
*
* @return the set of uncountable words
*/
public Set<String> getUncountables() {
return uncountables;
}
public void addPluralize( String rule,
String replacement ) {
final Rule pluralizeRule = new Rule(rule, replacement);
this.plurals.addFirst(pluralizeRule);
}
public void addSingularize( String rule,
String replacement ) {
final Rule singularizeRule = new Rule(rule, replacement);
this.singulars.addFirst(singularizeRule);
}
public void addIrregular( String singular,
String plural ) {
//CheckArg.isNotEmpty(singular, "singular rule");
//CheckArg.isNotEmpty(plural, "plural rule");
String singularRemainder = singular.length() > 1 ? singular.substring(1) : "";
String pluralRemainder = plural.length() > 1 ? plural.substring(1) : "";
addPluralize("(" + singular.charAt(0) + ")" + singularRemainder + "$", "$1" + pluralRemainder);
addSingularize("(" + plural.charAt(0) + ")" + pluralRemainder + "$", "$1" + singularRemainder);
}
public void addUncountable( String... words ) {
if (words == null || words.length == 0) return;
for (String word : words) {
if (word != null) uncountables.add(word.trim().toLowerCase());
}
}
/**
* Utility method to replace all occurrences given by the specific backreference with its uppercased form, and remove all
* other backreferences.
*
* The Java {@link Pattern regular expression processing} does not use the preprocessing directives <code>\l</code>,
* <code>\u</code>, <code>\L</code>, and <code>\U</code>. If so, such directives could be used in the replacement string
* to uppercase or lowercase the backreferences. For example, <code>\L1</code> would lowercase the first backreference, and
* <code>\u3</code> would uppercase the 3rd backreference.
*
*
* @param input
* @param regex
* @param groupNumberToUppercase
* @return the input string with the appropriate characters converted to upper-case
*/
protected static String replaceAllWithUppercase( String input,
String regex,
int groupNumberToUppercase ) {
Pattern underscoreAndDotPattern = Pattern.compile(regex);
Matcher matcher = underscoreAndDotPattern.matcher(input);
StringBuffer sb = new StringBuffer();
while (matcher.find()) {
matcher.appendReplacement(sb, matcher.group(groupNumberToUppercase).toUpperCase());
}
matcher.appendTail(sb);
return sb.toString();
}
/**
* Completely remove all rules within this inflector.
*/
public void clear() {
this.uncountables.clear();
this.plurals.clear();
this.singulars.clear();
}
protected void initialize() {
Inflector inflect = this;
inflect.addPluralize("$", "s");
inflect.addPluralize("s$", "s");
inflect.addPluralize("(ax|test)is$", "$1es");
inflect.addPluralize("(octop|vir)us$", "$1i");
inflect.addPluralize("(octop|vir)i$", "$1i"); // already plural
inflect.addPluralize("(alias|status)$", "$1es");
inflect.addPluralize("(bu)s$", "$1ses");
inflect.addPluralize("(buffal|tomat)o$", "$1oes");
inflect.addPluralize("([ti])um$", "$1a");
inflect.addPluralize("([ti])a$", "$1a"); // already plural
inflect.addPluralize("sis$", "ses");
inflect.addPluralize("(?:([^f])fe|([lr])f)$", "$1$2ves");
inflect.addPluralize("(hive)$", "$1s");
inflect.addPluralize("([^aeiouy]|qu)y$", "$1ies");
inflect.addPluralize("(x|ch|ss|sh)$", "$1es");
inflect.addPluralize("(matr|vert|ind)ix|ex$", "$1ices");
inflect.addPluralize("([m|l])ouse$", "$1ice");
inflect.addPluralize("([m|l])ice$", "$1ice");
inflect.addPluralize("^(ox)$", "$1en");
inflect.addPluralize("(quiz)$", "$1zes");
// Need to check for the following words that are already pluralized:
inflect.addPluralize("(people|men|children|sexes|moves|stadiums)$", "$1"); // irregulars
inflect.addPluralize("(oxen|octopi|viri|aliases|quizzes)$", "$1"); // special rules
inflect.addSingularize("s$", "");
inflect.addSingularize("(s|si|u)s$", "$1s"); // '-us' and '-ss' are already singular
inflect.addSingularize("(n)ews$", "$1ews");
inflect.addSingularize("([ti])a$", "$1um");
inflect.addSingularize("((a)naly|(b)a|(d)iagno|(p)arenthe|(p)rogno|(s)ynop|(t)he)ses$", "$1$2sis");
inflect.addSingularize("(^analy)ses$", "$1sis");
inflect.addSingularize("(^analy)sis$", "$1sis"); // already singular, but ends in 's'
inflect.addSingularize("([^f])ves$", "$1fe");
inflect.addSingularize("(hive)s$", "$1");
inflect.addSingularize("(tive)s$", "$1");
inflect.addSingularize("([lr])ves$", "$1f");
inflect.addSingularize("([^aeiouy]|qu)ies$", "$1y");
inflect.addSingularize("(s)eries$", "$1eries");
inflect.addSingularize("(m)ovies$", "$1ovie");
inflect.addSingularize("(x|ch|ss|sh)es$", "$1");
inflect.addSingularize("([m|l])ice$", "$1ouse");
inflect.addSingularize("(bus)es$", "$1");
inflect.addSingularize("(o)es$", "$1");
inflect.addSingularize("(shoe)s$", "$1");
inflect.addSingularize("(cris|ax|test)is$", "$1is"); // already singular, but ends in 's'
inflect.addSingularize("(cris|ax|test)es$", "$1is");
inflect.addSingularize("(octop|vir)i$", "$1us");
inflect.addSingularize("(octop|vir)us$", "$1us"); // already singular, but ends in 's'
inflect.addSingularize("(alias|status)es$", "$1");
inflect.addSingularize("(alias|status)$", "$1"); // already singular, but ends in 's'
inflect.addSingularize("^(ox)en", "$1");
inflect.addSingularize("(vert|ind)ices$", "$1ex");
inflect.addSingularize("(matr)ices$", "$1ix");
inflect.addSingularize("(quiz)zes$", "$1");
inflect.addIrregular("person", "people");
inflect.addIrregular("man", "men");
inflect.addIrregular("child", "children");
inflect.addIrregular("sex", "sexes");
inflect.addIrregular("move", "moves");
inflect.addIrregular("stadium", "stadiums");
inflect.addUncountable("equipment", "information", "rice", "money", "species", "series", "fish", "sheep");
}
}
SimpleNLG - 是否有可能讓SimpleNLG自動檢測名詞是單數還是復數？
[英]SimpleNLG - Is it possible to have SimpleNLG automatically detect if a noun is singular or plural?
WordNet和JWNL是否能夠將復數形式轉換為單數形式和將名詞轉換為名詞形式？
[英]Are WordNet and JWNL capable of converting plural-to-singular and ver(ing)-to-noun form?
聲明:本站的技術帖子網頁，遵循CC BY-SA 4.0協議，如果您需要轉載，請注明本站網址或者原文地址。任何問題請咨詢:yoyou2525@163.com.