NLPIR分词器javademo2016改编---非开源

itace

浏览: 181430 次
性别:
来自: 北京

最近访客更多访客>>

u012363178

wangyy

tianshiguishu

Sharpleo

博主相关

博客

微博

相册

留言

关于我

文章分类

社区版块

存档分类

博客分类：

分词

该项目并非开源项目：fail!E:\data\NLPIR\Data\NLPIR.user Not valid license or your license expired! Please feel free to contact pipy_zhang@msn.com!

官网下载压缩包，http://ictclas.nlpir.org/downloads

java版运行需要jna-4.0.0.jar，Data文件夹下的数据，NLPIR.dll动态链接库文件（linux是NLPIR.so，另外有32位和64位之分）

NLPIRLibrary类详细

package com.chipmunk.analyzer;

import java.util.Map;

import com.sun.jna.Library;
import com.sun.jna.Native;

public interface NLPIRLibrary extends Library {

	// 定义并初始化接口的静态变量 这一个语句是来加载 dll 的， 注意 dll 文件的路径
	// 可以是绝对路径也可以是相对路径，只需要填写 dll 的文件名，不能加后缀
	NLPIRLibrary Instance = (NLPIRLibrary) Native.loadLibrary(
			NLPIRPath.PATH_LIB, NLPIRLibrary.class);

	/**
	 *  初始化函数声明
	 * @param sDataPath:Initial Directory Path, where file Configure.xml and Data directory
			stored. the default value is 0, it indicates the initial directory is current
			working directory path
	 * @param encoding:encoding of input string, default is GBK_CODE (GBK encoding), and it
			can be set with UTF8_CODE (UTF8 encoding) and BIG5_CODE (BIG5 encoding).
	 * @param sLicenceCode:license code, special use for some commercial users. Other users
			ignore the argument
	 * @return 1-success,0-fail
	 */
	public int NLPIR_Init(String sDataPath, int encoding, String sLicenceCode);

	/**
	 * 执行分词函数声明
	 * The NLPIR_ParagraphProcess function works properly only if NLPIR_Init succeeds.
	 * @param sParagraph: The source paragraph
	 * @param bPOStagged: Judge whether need POS tagging, 0 for no tag; 1 for tagging; default:1.
	 * @return
	 */
	public String NLPIR_ParagraphProcess(String sSrc, int bPOSTagged);

	/**
	 *  提取关键词函数声明
	 * @param sLine, the input text.
	 * @param nMaxKeyLimit, the maximum number of key words.
	 * @param bWeightOut: whether the keyword weight output or not
	 * @return the keywords list if excute succeed. otherwise return NULL.
	 */
	public String NLPIR_GetKeyWords(String sLine, int nMaxKeyLimit,
			boolean bWeightOut);

	/**
	 * Extract keyword from a text file.
	 * @param sTextFile, the input text filename. 
	 * @param nMaxKeyLimit, the maximum number of key words. 
	 * @param bWeightOut: whether the keyword weight output or not 
	 * @return Return the keywords list if excute succeed. otherwise return NULL.
	 */
	public String NLPIR_GetFileKeyWords(String sLine, int nMaxKeyLimit,
			boolean bWeightOut);

	/**
	 *  添加用户词典声明Add a word to the user dictionary.
	 * @param sWord
	 * @return Return 1 if add succeed. Otherwise return 0.
	 */
	public int NLPIR_AddUserWord(String sWord);// add by qp 2008.11.10
	/**
	 *  删除用户词典声明
	 * @param sWord
	 * @return Return -1, the word not exist in the user dictionary; else, the handle of the word deleted
	 */
	public int NLPIR_DelUsrWord(String sWord);// add by qp 2008.11.10

	//错误信息
	public String NLPIR_GetLastErrorMsg();

	/**
	 *  退出函数声明
	 *  Exit the program and free all resources and destroy all working buffer used in NLPIR.
	 */
	public void NLPIR_Exit();

	/**
	 *  文件分词声明 The NLPIR_FileProcess function works properly only if NLPIR_Init succeeds.
	 * @param sSourceFilename: The source file name to be analysized; 
	 * @param sResultFilename: The result file name to store the results. 
	 * @param bPOStagged: Judge whether need POS tagging, 0 for no tag; 1 for tagging; default:1. 
	 * Return the processing speed if processing succeed. Otherwise return false.
	 */
	public boolean NLPIR_FileProcess(String utf8File, String utf8FileResult, int i);
	
	//--------------------------------plus----------------------------------------//
	/**
	 * 引入用户词典 Import user-defined dictionary from a text file.
	 * @param sFilename: Text filename for user dictionary 
	 * @param bOverwrite: true(default), overwrite the existing dictionary ,false, add to the existing dictionary
	 * @return 
	 */
	public int NLPIR_ImportUserDict(String sFilename,boolean bOverwrite);
	
	/**
	 * 
	 * @param sParagraph: The source paragraph 
	 * @param pResultCount: pointer to result vector size 
	 * @param bUserDict:whether use UserDict 
	 * @return
	 *  the pointer of result vector, it is managed by system, user cannot alloc and free it
		struct result_t{
			int start; //start position,词语在输入句子中的开始位置
			int length; //length,词语的长度
			char sPOS[POS_SIZE];//word type，词性ID值，可以快速的获取词性表
			int iPOS;//词性
			int word_ID; //如果是未登录词，设成或者-1
			int word_type; //区分用户词典;1，是用户词典中的词；，非用户词典中的词
			int weight;// word weight
		};
	 */
//	public result_t NLPIR_ParagraphProcessA(String sParagraph,int pResultCount,boolean bUserDict);
	public Map<String, Object> NLPIR_ParagraphProcessA(String sParagraph,int pResultCount,boolean bUserDict);
	
	/**
	 * The output format is customized in NLPIR configure.
	 * @param sParagraph
	 * @return
	 */
	public int NLPIR_GetParagraphProcessAWordCount(String sParagraph);
	
	/**
	 * 
	 * @param nCount: the paragraph word count. 
	 * @param result: Pointer to structure to store results. 
	 */
//	public void NLPIR_ParagraphProcessAW(int nCount,result_t result);
	public void NLPIR_ParagraphProcessAW(int nCount,Map<String, Object> result);
	
	/**
	 * Save the user dictionary to disk.
	 * @return Return 1 if save succeed. Otherwise return 0.
	 */
	public int NLPIR_SaveTheUsrDic();
	
	/**
	 * 关键词黑名单（永远不作为关键词输出）
	 * Import blacklist keyword dictionary from a text file.
	 * @param sFilename: Text filename for user dictionary
	 * @return The number of lexical entry imported successfully
	 */
	public int NLPIR_ImportKeyBlackList(String sFilename);
	
	/**
	 * Extract new words from paragraph.
	 * @param sLine, the input text.
	 * @param nMaxKeyLimit, the maximum number of key words. 
	 * @param bWeightOut: whether the keyword weight output or not 
	 * @return
	 */
	public String NLPIR_GetNewWords(String sLine,int nMaxKeyLimit,boolean bWeightOut);
	/**
	 * Extract new words from a text file.
	 * @param sTextFile, the input text filename. 
	 * @param nMaxKeyLimit, the maximum number of key words. 
	 * @param bWeightOut: whether the keyword weight output or not 
	 * @return Return the keywords list if excute succeed. otherwise return NULL.
	 */
	public String NLPIR_GetFileNewWords(String sTextFile,int nMaxKeyLimit,boolean bWeightOut);
	/**
	 * Extract a finger print from the paragraph .
	 * @param sLine
	 * @return  0, failed; else, the finger print of the content
	 */
	public long NLPIR_FingerPrint(String sLine);
	/**
	 * select which pos map will use
	 * @param nPOSmap
	 * @return
	 */
	public int NLPIR_SetPOSmap(int nPOSmap);
	/**
	 * 
	 * @return true:success, false:fail
	 */
	public boolean NLPIR_NWI_Start();
	/**
	 * 需要在运行NLPIR_NWI_Start()之后，才有效
	 * @param sFilename
	 * @return
	 */
	public int NLPIR_NWI_AddFile(String sFilename);
	/**
	 * 往新词识别系统中添加一段待识别新词的内存
	 * 需要在运行NLPIR_NWI_Start()之后，才有效
	 * @param sText
	 * @return
	 */
	public boolean NLPIR_NWI_AddMem(String sText);
	/**
	 * 新词识别添加内容结束
	 * 需要在运行NLPIR_NWI_Start()之后，才有效
	 * @return
	 */
	public boolean NLPIR_NWI_Complete();//新词
	/**
	 * 获取新词识别的结果
	 * 需要在运行NLPIR_NWI_Complete()之后，才有效
	 * @param bWeightOut 是否需要输出每个新词的权重参数
	 * @return
	 */
	public String NLPIR_NWI_GetResult(boolean bWeightOut);//输出新词识别结果
	/**
	 * 将新词识别结果导入到用户词典中
	 * 需要在运行NLPIR_NWI_Complete()之后，才有效
	 * 如果需要将新词结果永久保存，建议在执行NLPIR_SaveTheUsrDic
	 * @return
	 */
	public int NLPIR_NWI_Result2UserDict();//新词识别结果转为用户词典,返回新词结果数目
	/**
	 * //NLPIR
		NLPIR_NWI_Start();//启动新词发现功能
		NLPIR_NWI_AddFile(sInputFile); //添加新词训练的文件，可反复添加
		NLPIR_NWI_Complete();//添加文件或者训练内容结束
		const char *pNewWordlist=NLPIR_NWI_GetResult();//输出新词识别结果
		printf("识别出的新词为：%s\n",pNewWordlist);
		strcpy(sResultFile,sInputFile);
		strcat(sResultFile,"_result1.txt");
		NLPIR_FileProcess(sInputFile,sResultFile);
		NLPIR_NWI_Result2UserDict();//新词识别结果导入到用户词典
		strcpy(sResultFile,sInputFile);
		strcat(sResultFile,"_result2.txt");
		NLPIR_FileProcess(sInputFile,sResultFile);
		NLPIR_Exit();
	 */
	/**
	 * 功能：当前的切分结果过大时，如“中华人民共和国”
		需要执行该函数，将切分结果细分为“中华人民共和国”
		细分粒度最大为三个汉字
		返回：返回细粒度分词，如果不能细分，则返回为空字符串""
	 * @param sLine
	 * @return
	 */
	public String NLPIR_FinerSegment(String sLine);//最大细粒化分词
	/**
	 * 功能：获取各类英文单词的原型，考虑了过去分词、单复数等情况
		返回：返回的词原型形式
		driven->drive drives->drive drove-->drive
	 * @param sWord
	 * @return
	 */
	public String NLPIR_GetEngWordOrign(String sWord);//获取各类英文单词的原型，考虑了过去分词、单复数等情况
	/**
	 * 功能：获取输入文本的词，词性，频统计结果，按照词频大小排序
		返回：返回的是词频统计结果形式如下：
		张华平/nr/10#博士/n/9#分词/n/8
	 * 
	 * @param sText
	 * @return
	 */
	public String NLPIR_WordFreqStat(String sText);//获取输入文本的词，词性，频统计结果，按照词频大小排序
	/**
		功能：获取输入文本的词，词性，频统计结果，按照词频大小排序
		参数：sFilename 文本文件的全路径
		返回： 返回的是词频统计结果形式如下：
		张华平/nr/10#博士/n/9#分词/n/8
	 * @param sFilename
	 * @return
	 * 
	 */
	public String NLPIR_FileWordFreqStat(String sFilename);//获取输入文本的词，词性，频统计结果，按照词频大小排序
}

Demo

public static void main(String[] args) {
		String argu = NLPIRPath.PATH_DATA;
		int charset_type = 1;
		int init_flag = NLPIRLibrary.Instance.NLPIR_Init(argu, charset_type,"0");
		if (init_flag==1) {
			System.out.println("init success!");
		}else if (init_flag==0) {
			String message = NLPIRLibrary.Instance.NLPIR_GetLastErrorMsg();
			System.out.println("init fail!"+message);
		}
		String word = NLPIRLibrary.Instance.NLPIR_GetEngWordOrign("wanted");
		System.out.println(word);
		NLPIRLibrary.Instance.NLPIR_AddUserWord("强降雨 n");
		String word2 = NLPIRLibrary.Instance.NLPIR_WordFreqStat("南方多地出现强降雨的同时，高温天气也在南方蔓延。据中国天气网21日消息，在副热带高压控制下，19日开始，华南一带出现高温天气，影响范围逐步扩大。预计，未来10天，高温继续蔓延，江南中南部、华南将出现日最高气温为35-38℃的持续高温晴热天气。");
		System.out.println(word2);
		String keywords =NLPIRLibrary.Instance.NLPIR_GetFileKeyWords("E:/temp/abc.txt", 10, true);
		System.out.println(keywords);
		
//		NLPIRLibrary.Instance.NLPIR_NWI_Start();//新词识别开始
//		NLPIRLibrary.Instance.NLPIR_NWI_AddFile("E:/temp/def.txt");//批量增加输入文件，可以不断循环调用NLPIR_NWI_AddFile或者NLPIR_NWI_AddMem
//		
//		NLPIRLibrary.Instance.NLPIR_NWI_Complete();//新词识别导入文件结束
//		
//		String t= NLPIRLibrary.Instance.NLPIR_NWI_GetResult(false);//获取本次批量导入文本文件中识别的新词结果
//		System.out.println("新词识别结果 " + t);//打印输出新词识别结果
//		System.out.println("============");
		
		try {
			String aaa = NLPIRLibrary.Instance.NLPIR_ParagraphProcess(FileUtil.read(new File("E:/temp/def.txt"), "UTF-8"), 0);
			System.out.println(aaa.replaceAll(" ", "，"));
		} catch (Exception e) {
			e.printStackTrace();
		}
		String abc =NLPIRLibrary.Instance.NLPIR_GetFileNewWords("E:/temp/def.txt", 10, false);
		for (String a : abc.split("#")) {
			NLPIRLibrary.Instance.NLPIR_AddUserWord(a+" n");
			System.out.println(a);
		}
		System.out.println("abc:"+abc);
		NLPIRLibrary.Instance.NLPIR_AddUserWord("暴雨黄色预警 n");
		try {
			String ddd = NLPIRLibrary.Instance.NLPIR_ParagraphProcess(FileUtil.read(new File("E:/temp/def.txt"), "UTF-8"), 0);
			System.out.println(ddd.replaceAll(" ", "，"));
		} catch (Exception e) {
			e.printStackTrace();
		}
		NLPIRLibrary.Instance.NLPIR_Exit();
		
	}