// @Bean
public boolean doconfig() {
WordConfTools.set("dic.path", "classpath:dic.txt，classpath:config/mydic.txt");
DictionaryFactory.reload(); // 更改词典路径之后，重新加载词典
// String path = WordConfTools.get("dic.path");
// log.info("dic.path {}" ,path);
return true;
}
public static void reload() {
AutoDetector.loadAndWatch(
new ResourceLoader() {
@Override
public void clear() {
quantifiers.clear();
}
@Override
public void load(List<String> lines) {
LOGGER.info("初始化数量词");
for (String line : lines) {
if (line.length() == 1) {
char _char = line.charAt(0);
if (quantifiers.contains(_char)) {
LOGGER.info("配置文件有重复项：" + line);
} else {
quantifiers.add(_char);
}
} else {
LOGGER.info("忽略不合法数量词：" + line);
}
}
LOGGER.info("数量词初始化完毕，数量词个数：" + quantifiers.size());
}
@Override
public void add(String line) {
if (line.length() == 1) {
char _char = line.charAt(0);
quantifiers.add(_char);
} else {
LOGGER.info("忽略不合法数量词：" + line);
}
}
@Override
public void remove(String line) {
if (line.length() == 1) {
char _char = line.charAt(0);
quantifiers.remove(_char);
} else {
LOGGER.info("忽略不合法数量词：" + line);
}
}
},
WordConfTools.get("quantifier.path", "classpath:quantifier.txt"));
}
public static void reload() {
AutoDetector.loadAndWatch(
new ResourceLoader() {
@Override
public void clear() {
DOUBLE_ARRAY_GENERIC_TRIE.clear();
}
@Override
public void load(List<String> lines) {
LOGGER.info("初始化trigram");
Map<String, Integer> map = new HashMap<>();
for (String line : lines) {
try {
addLine(line, map);
} catch (Exception e) {
LOGGER.error("错误的trigram数据：" + line);
}
}
int size = map.size();
DOUBLE_ARRAY_GENERIC_TRIE.putAll(map);
LOGGER.info("trigram初始化完毕，trigram数据条数：" + size);
}
@Override
public void add(String line) {
throw new RuntimeException("not yet support menthod!");
}
private void addLine(String line, Map<String, Integer> map) {
String[] attr = line.split("\\s+");
int frequency = Integer.parseInt(attr[1]);
if (frequency > maxFrequency) {
maxFrequency = frequency;
}
map.put(attr[0], frequency);
}
@Override
public void remove(String line) {
throw new RuntimeException("not yet support menthod!");
}
},
WordConfTools.get("trigram.path", "classpath:trigram.txt"));
}
/**
* 三元语法模型
*
* @author 杨尚川
*/
public class Trigram {
private static final Logger LOGGER = LoggerFactory.getLogger(Trigram.class);
private static final DoubleArrayGenericTrie DOUBLE_ARRAY_GENERIC_TRIE =
new DoubleArrayGenericTrie(WordConfTools.getInt("trigram.double.array.trie.size", 9800000));
private static int maxFrequency = 0;
static {
reload();
}
public static void reload() {
AutoDetector.loadAndWatch(
new ResourceLoader() {
@Override
public void clear() {
DOUBLE_ARRAY_GENERIC_TRIE.clear();
}
@Override
public void load(List<String> lines) {
LOGGER.info("初始化trigram");
Map<String, Integer> map = new HashMap<>();
for (String line : lines) {
try {
addLine(line, map);
} catch (Exception e) {
LOGGER.error("错误的trigram数据：" + line);
}
}
int size = map.size();
DOUBLE_ARRAY_GENERIC_TRIE.putAll(map);
LOGGER.info("trigram初始化完毕，trigram数据条数：" + size);
}
@Override
public void add(String line) {
throw new RuntimeException("not yet support menthod!");
}
private void addLine(String line, Map<String, Integer> map) {
String[] attr = line.split("\\s+");
int frequency = Integer.parseInt(attr[1]);
if (frequency > maxFrequency) {
maxFrequency = frequency;
}
map.put(attr[0], frequency);
}
@Override
public void remove(String line) {
throw new RuntimeException("not yet support menthod!");
}
},
WordConfTools.get("trigram.path", "classpath:trigram.txt"));
}
public static int getMaxFrequency() {
return maxFrequency;
}
/**
* 一次性计算多种分词结果的三元模型分值
*
* @param sentences 多种分词结果
* @return 分词结果及其对应的分值
*/
public static Map<List<Word>, Float> trigram(List<Word>... sentences) {
Map<List<Word>, Float> map = new HashMap<>();
// 计算多种分词结果的分值
for (List<Word> sentence : sentences) {
if (map.get(sentence) != null) {
// 相同的分词结果只计算一次分值
continue;
}
float score = 0;
// 计算其中一种分词结果的分值
if (sentence.size() > 2) {
for (int i = 0; i < sentence.size() - 2; i++) {
String first = sentence.get(i).getText();
String second = sentence.get(i + 1).getText();
String third = sentence.get(i + 2).getText();
float trigramScore = getScore(first, second, third);
if (trigramScore > 0) {
score += trigramScore;
}
}
}
map.put(sentence, score);
}
return map;
}
/**
* 计算分词结果的三元模型分值
*
* @param words 分词结果
* @return 三元模型分值
*/
public static float trigram(List<Word> words) {
if (words.size() > 2) {
float score = 0;
for (int i = 0; i < words.size() - 2; i++) {
score +=
Trigram.getScore(
words.get(i).getText(), words.get(i + 1).getText(), words.get(i + 2).getText());
}
return score;
}
return 0;
}
/**
* 获取三个词前后紧挨着同时出现在语料库中的分值 分值被归一化了： 完全没有出现分值为0 出现频率最高的分值为1
*
* @param first 第一个词
* @param second 第二个词
* @param third 第三个词
* @return 同时出现的分值
*/
public static float getScore(String first, String second, String third) {
int frequency = getFrequency(first, second, third);
float score = frequency / (float) maxFrequency;
if (LOGGER.isDebugEnabled()) {
if (score > 0) {
LOGGER.debug("三元模型 " + first + ":" + second + ":" + third + " 获得分值：" + score);
}
}
return score;
}
public static int getFrequency(String first, String second, String third) {
Integer value = DOUBLE_ARRAY_GENERIC_TRIE.get(first + ":" + second + ":" + third);
if (value == null) {
return 0;
}
return value;
}
}
