java 简易词语提取程序

需求简述:

从句子中提取标准词库出现的词语(稍作调整即可记录位置/词性等信息)

示例:

输入: 
text: 问渠那得清如许, 为有源头活水来, 问渠
dictionary: ['问渠', '问渠那', '清如许', '活水']
输出: 
['问渠那', '清如许', '活水', '问渠']

package kim.nzxy;

import java.util.*;

/**
 * @author ly-chn
 */
public class LyKeywordsExtractor {

    private final Trie trie;
    private final int maxLength;

    public LyKeywordsExtractor(Set<String> dictionary) {
        this.trie = new Trie();
        int maxLength = 0;
        for (String word : dictionary) {
            maxLength = Math.max(maxLength, word.length());
            trie.insert(word);
        }
        this.maxLength = maxLength;
    }

    public static void main(String[] args) {
        Set<String> standardKeywords = new HashSet<>();
        standardKeywords.add("问渠");
        standardKeywords.add("渠那");
        standardKeywords.add("清如许");

        String text = "问渠那得清如许, 为有源头活水来, 问渠那得清如许, 为有源头活水来";

        LyKeywordsExtractor extractor = new LyKeywordsExtractor(standardKeywords);
        Set<String> extractedKeywords = extractor.segment(text);
        System.out.println(extractedKeywords);
    }

    public Set<String> segment(String text) {
        Set<String> words = new LinkedHashSet<>();
        int i = 0;
        while (i < text.length()) {
            String maxWord = "";
            int maxWordLength = Math.min(text.length(), i + maxLength);
            for (int j = i + 1; j <= maxWordLength; j++) {
                String word = text.substring(i, j);
                if (trie.find(word) && word.length() > maxWord.length()) {
                    maxWord = word;
                }
            }
            if (!maxWord.isEmpty()) {
                words.add(maxWord);
                i += maxWord.length();
            } else {
                i++;
            }
        }
        return words;
    }

    static class TrieNode {
        Map<Character, TrieNode> children;
        boolean isWord;

        TrieNode() {
            children = new HashMap<>();
            isWord = false;
        }
    }

    static class Trie {
        TrieNode root;

        Trie() {
            root = new TrieNode();
        }

        void insert(String word) {
            TrieNode node = root;
            for (char c : word.toCharArray()) {
                if (!node.children.containsKey(c)) {
                    node.children.put(c, new TrieNode());
                }
                node = node.children.get(c);
            }
            node.isWord = true;
        }

        boolean find(String word) {
            TrieNode node = root;
            for (char c : word.toCharArray()) {
                if (!node.children.containsKey(c)) {
                    return false;
                }
                node = node.children.get(c);
            }
            return node.isWord;
        }
    }
}

简述一下流程:

首先构建一下字典树, 然后从给定字符串的每个索引位置往后检索, 判断是否在字典中出现, 如果出现(如果出现多个, 则仅提取最长的那个), 则列入结果集