利用文本挖掘技术来找出《天龙八部》中的“小鲜词”

package grid.common;

import java.io.Serializable;
import java.util.HashMap;


public class CountMap<T> extends HashMap<T, Integer> implements Serializable {

    private static final long serialVersionUID = 6097963798841161750L;

    public void increase(T t) {
        Integer count = get(t);
        if (null == count) {
            put(t, 1);
        } else {
            put(t, ++count);
        }
    }

    public int count() {   
        int count = 0;
        for (T t : keySet()) {
            count += get(t);
        }
        return count;
    }

    public int get(char c) {
        Integer count = super.get(c);
        return null == count ? 0 : count;
    }
}

package grid.common;

import java.util.ArrayList;
import java.util.Collections;
import java.util.List;

public class Node<T> {
    protected List<Node<T>> children;

    protected Node<T> parent;

    protected T value;

    Node(T value) {
        this.value = value;
    }

    public Node<T> add(T value) {
        if (null == children) {
            children = new ArrayList<Node<T>>();
        }
        Node<T> child = new Node<T>(value);
        child.setParent(this);
        children.add(child);
        return child;
    }

    public T getValue() {
        return value;
    }

    public Node<T> getParent() {
        return parent;
    }

    public void setParent(Node<T> parent) {
        this.parent = parent;
    }

    private void recurseChildren(List<Node<T>> list, Node<T> parent) {
        if (null == parent.children) {
            list.add(parent);
        } else {
            for (Node<T> node : parent.children) {
                recurseChildren(list, node);
            }
        }
    }

    public List<Node<T>> getLeaves() {
        List<Node<T>> list = new ArrayList<Node<T>>();
        recurseChildren(list, this);
        return list;

    }

    public List<T> getBranchPath() {
        List<T> list = new ArrayList<T>();
        Node<T> node = this;
        do {
            list.add(node.getValue());
            node = node.parent;
        } while (null != node && !(node instanceof Tree<?>));
        Collections.reverse(list);
        return list;
    }

    private void append(StringBuilder builder, int deep, Node<T> node) {
        for (int i = 0; i < deep; i++) {
            builder.append("  ");
        }
        builder.append("|--");
        builder.append(node.getValue());
        builder.append("\n");
        if (null != node.children) {
            for (Node<T> child : node.children) {
                append(builder, deep + 1, child);
            }
        }
    }

    public String dump() {
        StringBuilder builder = new StringBuilder();
        append(builder, 0, this);
        return builder.toString();
    }

    public String toString() {
        return value.toString();
    }
}

package grid.common;

import java.io.BufferedReader;
import java.io.File;
import java.io.FileInputStream;
import java.io.FileOutputStream;
import java.io.IOException;
import java.io.InputStreamReader;

public class TextDatReader {

    @SuppressWarnings("resource")
    public static String read(String path) throws IOException {
        File file = new File(path);
        FileInputStream s = new FileInputStream(file);

    FileReader fr = new FileReader(file);
        BufferedReader reader = new BufferedReader(new InputStreamReader(s,
                "utf8"));
        char buffer[] = new char[(int) file.length()];
        reader.read(buffer);
        return new String(buffer);
    }


    public static void createDir() {
        File file = new File("./dat");
        if (!file.exists() && !file.isDirectory()) {
            file.mkdir();
        }
    }

    public static final String SUFFIX = ".dat"; 


    public static void divide(String name, long size) throws Exception {
        File file = new File(name);
        if (!file.exists() || (!file.isFile())) {
            throw new Exception("指定文件不存在！");
        }

        long fileLength = file.length();
        if (size <= 0) {
            size = fileLength / 2;
        }

        int num = (fileLength % size != 0) ? (int) (fileLength / size + 1)
                : (int) (fileLength / size);

        String[] fileNames = new String[num];

        FileInputStream in = new FileInputStream(file);

        long end = 0;
        int begin = 0;
        createDir();

        for (int i = 1; i <= num; i++) {

            File outFile = new File("./dat", "text" + i + SUFFIX);

            FileOutputStream out = new FileOutputStream(outFile);

            end += size;
            end = (end > fileLength) ? fileLength : end;

            for (; begin < end; begin++) {
                out.write(in.read());
            }
            out.close();
            fileNames[i] = outFile.getAbsolutePath();
            System.out.println("第"+i+"个子文件生成……");

        }
        in.close();
    }



}

package grid.common;


public class TextUtils {

    public static boolean isCnLetter(char c) {
        return c >= 0x4E00 && c <= 0x9FCB;
    }

    public static boolean isNumeric(char c) {
            return c >= '0' && c <= '9';
    }

    public static boolean isEnLetter(char c) {
        return (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z');
    }

    public static boolean match(String src, int off, String dest) {
        int len = dest.length();
        int srcLen = src.length();
        for (int i = 0; i < len; i++) {
            if (srcLen <= off + i) {
                return false;
            }
            if (dest.charAt(i) != src.charAt(off + i)) {
                return false;
            }
        }
        return true;
    }

    public static boolean isBlank(String str) {
        return null == str || str.isEmpty() || str.trim().isEmpty();
    }
}

package grid.common;


public class Tree<T> extends Node<T> {

    public Tree(T value) {
        super(value);
    }

}

package grid.text.dic;

import grid.common.CountMap;
import grid.common.TextDatReader;
import grid.common.TextUtils;

import java.io.IOException;
import java.util.HashSet;
import java.util.Set;


public class CnDictionary {

    private final String COMMON_WORD_DIC_PATH = "common.dic";


    private final String COMMON_LETTER_RESOURCE_PATH = "text.dat";

    private Set<String> dictionary = new HashSet<String>();

    private CountMap<Character> letterCountMap = new CountMap<Character>();

    private int totalLetterCount;

    private static CnDictionary instance;

    public static CnDictionary Instance() {
        if (null == instance) {
            try {
                instance = new CnDictionary();
            } catch (IOException e) {
                e.printStackTrace();
            }
        }
        return instance;
    }

    private CnDictionary() throws IOException {
        initWordDic();
        initLetterCountMap();
    }

    private void initLetterCountMap() throws IOException {
        String letterResource = TextDatReader.read(COMMON_LETTER_RESOURCE_PATH);
        final int len = letterResource.length();
        char c;
        for (int i = 0; i < len; i++) {
            c = letterResource.charAt(i);
            if (TextUtils.isCnLetter(c)) {
                letterCountMap.increase(c);
            }
        }
        totalLetterCount = letterCountMap.count();

    }

    private void initWordDic() throws IOException {

        String bytes = TextDatReader.read(COMMON_WORD_DIC_PATH);
        final int len = bytes.length();
        String s = "";
        char c;
        for (int i = 0; i < len; i++) {
            c = bytes.charAt(i);

            if ('\n' == c || '\r' == c || 0 == c) {
                if (!TextUtils.isBlank(s)) {
                    dictionary.add(s.trim());
                }
                s = "";
            } else {
                s += c;
            }
            if (0 == c) {
                break;
            }
        }
    }

    public boolean contains(String word) {
        return dictionary.contains(word);
    }

    public double rate(char c) {
        return (double) letterCountMap.get(c) / totalLetterCount;
    }

    public int size() {
        return dictionary.size();
    }
}

package grid.text.evolution;

import grid.common.CountMap;
import grid.common.TextUtils;
import grid.text.index.Pos;
import grid.text.index.TextIndexer;

public class EntropyJudger {

    private TextIndexer indexer;


    private static int LEAST_COUNT_THRESHOLD = 5;


    private static double SOLID_RATE_THRESHOLD = 0.018;


    private static double ENTROPY_THRESHOL = 1.92;

    public EntropyJudger(TextIndexer indexer) {
        this.indexer = indexer;
    }

    public boolean judge(String candidate) {
        double solidRate = getSolidRate(candidate);

        if (solidRate < SOLID_RATE_THRESHOLD) {
            return false;
        }

        double entropy = getEntropy(candidate);

        if (entropy < ENTROPY_THRESHOL) {
            return false;
        }
        return true;
    }

    private double getEntropy(String candidate) {
        Pos pos = new Pos(candidate);
        CountMap<Character> frontCountMap = new CountMap<Character>();
        CountMap<Character> backCountMap = new CountMap<Character>();
        final int candidateLen = candidate.length();
        int off = 0;
        char c;
        double rate, frontEntropy = 0, backEntropy = 0;

        while (indexer.find(pos).isFound()) {
            off = pos.getPos();

            c = indexer.charAt(off - 1);
            if (TextUtils.isCnLetter(c)) {
                frontCountMap.increase(c);
            }
            c = indexer.charAt(off + candidateLen);
            if (TextUtils.isCnLetter(c)) {
                backCountMap.increase(c);
            }

        }

        for (char key : frontCountMap.keySet()) {
            rate = (double) frontCountMap.get(key) / frontCountMap.count();
            frontEntropy -= rate * Math.log(rate);
        }
        for (char key : backCountMap.keySet()) {
            rate = (double) backCountMap.get(key) / backCountMap.count();
            backEntropy -= rate * Math.log(rate);
        }

        return frontEntropy > backEntropy ? backEntropy : frontEntropy;

    }


    public double getSolidRate(String candidate) {

        final int candidateLen = candidate.length();

        if (candidateLen < 2) {
            return 1;
        }

        final int count = indexer.count(candidate);
        double rate = 1;

        if (count < LEAST_COUNT_THRESHOLD) {
            return 0;
        }

        for (int i = 0; i < candidateLen; i++) {
            rate *= (double) count / indexer.count("" + candidate.charAt(i));
        }

        return Math.pow(rate, 1D / candidateLen) * Math.sqrt(candidateLen);
    }

    public void setIndexer(TextIndexer indexer) {
        this.indexer = indexer;
    }

}

package grid.text.evolution;

import grid.common.TextUtils;
import grid.text.dic.CnDictionary;
import grid.text.index.CnPreviewTextIndexer;
import grid.text.index.TextIndexer;
import grid.text.selector.CnTextSelector;
import grid.text.selector.TextSelector;

import java.util.HashSet;
import java.util.Set;

public class NewWordDiscover {

    private CnDictionary dictionary;


    private final static int MIN_CANDIDATE_LEN = 2;


    private final static int MAX_CANDIDATE_LEN = 6;

    private static Set<Character> structuralLetterSet = new HashSet<Character>();

    private static char[] structuralLetters = { '我', '你', '您', '他', '她', '谁',
            '哪', '那', '这', '的', '了', '着', '也', '是', '有', '不', '在', '与', '呢',
            '啊', '呀', '吧', '嗯', '哦', '哈', '呐' };

    static {
        for (char c : structuralLetters) {
            structuralLetterSet.add(c);
        }
    }

    public NewWordDiscover() {
        dictionary = CnDictionary.Instance();
    }


    public Set<String> discover(String document) {

        Set<String> set = new HashSet<String>();
        TextIndexer indexer = new CnPreviewTextIndexer(document);
        TextSelector selector = new CnTextSelector(document, MIN_CANDIDATE_LEN,
                MAX_CANDIDATE_LEN);
        EntropyJudger judger = new EntropyJudger(indexer);
        String candidate;
        while (!selector.end()) {
            candidate = selector.next();
            if (TextUtils.isBlank(candidate)) {
                continue;
            }
            if (structuralLetterSet.contains(candidate.charAt(0))
                    || structuralLetterSet.contains(candidate.charAt(candidate
                            .length() - 1))) {
                continue;
            }

            if (dictionary.contains(candidate) || set.contains(candidate)) {
                selector.select();
            } else if (judger.judge(candidate)) {
                set.add(candidate);
            }
        }
        return set;
    }
}

package grid.text.index;

import grid.common.TextUtils;

import java.util.HashMap;
import java.util.Map;
import java.util.Vector;

public class CnPreviewTextIndexer implements TextIndexer {

    private final static int CN_LETTER_COUNT = 5021;

    private String document;

    private Map<Character, Vector<Integer>> posMap;

    public CnPreviewTextIndexer(String document) {
        this.document = document;
        init();
    }

    private void init() {
        final int len = document.length();

        final int supposedMinCount = 1 + (int) Math.log(len / CN_LETTER_COUNT
                + 1);

        char c;

        Vector<Integer> posVector;

        posMap = new HashMap<Character, Vector<Integer>>(CN_LETTER_COUNT);

        for (int i = 0; i < len; i++) {
            c = document.charAt(i);
            if (!TextUtils.isCnLetter(c)) {
                continue;
            }
            posVector = posMap.get(c);
            if (null == posVector) {
                posVector = new Vector<Integer>(supposedMinCount);
                posMap.put(c, posVector);
            }
            posVector.add(i);
        }
    }

    @Override
    public int count(String text) {

        if (TextUtils.isBlank(text)) {
            return 0;
        }

        Vector<Integer> vector = posMap.get(text.charAt(0));

        if (null == vector) {
            return 0;
        }

        if (1 == text.length()) {
            return vector.size();
        }

        final int size = vector.size();
        int count = 0;

        for (int i = 0; i < size; i++) {
            if (TextUtils.match(document, vector.get(i), text)) {
                count++;
            }
        }

        return count;
    }

    @Override
    public Pos find(Pos pos) {
        String text = pos.getTarget();

        pos.setFound(false);

        if (TextUtils.isBlank(text)) {
            return pos;
        }

        Vector<Integer> vector = posMap.get(text.charAt(0));

        if (null == vector) {
            return pos;
        }

        final int arraySize = vector.size();
        final int arrayIndex = pos.arrayIndex + 1;

        for (int i = arrayIndex; i < arraySize; i++) {
            if (TextUtils.match(document, vector.get(i), text)) {
                pos.setFound(true);
                pos.setPos(vector.get(i));
                pos.arrayIndex = i;
                break;
            }
        }

        return pos;
    }

    @Override
    public int len() {
        return document.length();
    }

    @Override
    public String sub(int off, int len) {
        if (off < 0 || off + len >= document.length()) {
            return "";
        }
        return document.substring(off, off + len);
    }

    @Override
    public char charAt(int index) {
        if (index < 0 || index >= document.length()) {
            return 0;
        }
        return document.charAt(index);
    }
}

package grid.text.index;


public class Pos {
    private String target;

    /**
     * Pos for current matched full target text
     */
    private int pos = -1;

    /**
     * Index in position array for current matched full target text
     */
    int arrayIndex = -1;

    private boolean found = false;

    public Pos(String target) {
        this.target = target;
    }

    public String getTarget() {
        return target;
    }

    public int getPos() {
        return pos;
    }

    public boolean isFound() {
        return found;
    }

    void setPos(int pos) {
        this.pos = pos;
    }

    void setFound(boolean found) {
        this.found = found;
    }
}

package grid.text.index;


public class SimpleTextIndexer implements TextIndexer {

    private String document;

    public SimpleTextIndexer(String document) {
        this.document = document;
    }

    @Override
    public int count(String text) {
        int off = 0;
        int count = 0;
        final int len = text.length();
        while ((off = document.indexOf(text, off)) > -1) {
            count++;
            off += len;
        }
        return count;
    }

    @Override
    public Pos find(Pos pos) {
        final String text = pos.getTarget();
        final int len = text.length();
        int off = pos.getPos() + len;
        if (pos.getPos() < 0)
            off = 0;

        pos.setFound(false);

        if ((off = document.indexOf(text, off)) > -1) {
            pos.setFound(true);
            pos.setPos(off);
        }
        return pos;
    }

    @Override
    public int len() {
        return document.length();
    }

    @Override
    public String sub(int off, int len) {
        return document.substring(off, off + len);
    }

    @Override
    public char charAt(int index) {
        if (index < 0 || index >= document.length()) {
            return 0;
        }
        return document.charAt(index);
    }
}

package grid.text.index;


public interface TextIndexer {

    /**
     * @param text
     * @return count for specific text
     */
    public int count(String text);

    /**
     * @param pos
     * @return next position for current pos
     */
    public Pos find(Pos pos);

    /**
     * @return original document length
     */
    public int len();

    /**
     * @param off
     * @param len
     * @return the sub string start from <b>off</b> and with a length with
     *         <b>len</b>
     */
    public String sub(int off, int len);

    /**
     * @param index
     * @return return the character in the specified index
     */
    public char charAt(int index);
}

package grid.text.participle;

import grid.text.dic.CnDictionary;

import java.util.List;


public class Chunk implements Comparable<Chunk> {

    private List<String> list;

    private int len = 0;

    private double avg = 0;

    private double variance = 0;

    public Chunk(List<String> list) {
        this.list = list;
        init();
    }

    private void init() {

        for (String s : list) {
            len += s.length();
        }
        avg = (double) len / list.size();

        for (String s : list) {
            variance += Math.pow(avg - s.length(), 2);
        }
        variance = Math.sqrt(variance);
    }

    public int getLen() {
        return len;
    }

    public double getAvg() {
        return avg;
    }

    public double getVariance() {
        return variance;
    }

    public String getHead() {
        if (null == list || list.isEmpty()) {
            return "";
        }
        return list.get(0);
    }

    private int compareDouble(double d1, double d2) {
        if (d1 - d2 < -0.0000001D) {
            return 1;
        } else if (d1 - d2 > 0.0000001D) {
            return -1;
        }
        return 0;
    }

    @Override
    public int compareTo(Chunk o) {

        if (len != o.len) {
            return o.len - len;
        }

        int d = compareDouble(avg, o.avg);
        if (0 != d) {
            return d;
        }

        d = compareDouble(variance, o.variance);
        if (0 != d) {
            return d;
        }

        CnDictionary dictionary = CnDictionary.Instance();

        double rateSrc = 0, rateDest = 0;
        for (String s : list) {
            if (1 == s.length()) {
                rateSrc += dictionary.rate(s.charAt(0));
            }
        }
        for (String s : o.list) {
            if (1 == s.length()) {
                rateDest += dictionary.rate(s.charAt(0));
            }
        }
        return compareDouble(rateSrc, rateDest);
    }

    public String toString() {
        return list.toString();
    }
}

package grid.text.participle;

import grid.common.Node;
import grid.common.TextUtils;
import grid.common.Tree;
import grid.text.dic.CnDictionary;

import java.util.ArrayList;
import java.util.Collections;
import java.util.List;

public class ChunkStream {

    /**
     * Define the max supposed word length
     * 
     * You could shorten the value if you don't need too long participle result
     */
    private static final int MAX_WORD_LEN = 7;

    /**
     * Define the predict level while execute participle.
     * 
     * Negligible accuracy will be promoted if you increase this value
     */
    private static final int PREDICT_LEVEL = 3;

    private static CnDictionary dictionary = CnDictionary.Instance();

    public String next(String text, int off) {
        Tree<String> root = new Tree<String>("ROOT");
        recurse(root, off, text, 0);
        List<Node<String>> list = root.getLeaves();
        List<Chunk> chunkList = new ArrayList<Chunk>();
        for (Node<String> node : list) {
            chunkList.add(new Chunk(node.getBranchPath()));
        }
        Collections.sort(chunkList);
        return chunkList.get(0).getHead();

    }

    private void recurse(Node<String> node, int off, String text,
            int predictDeep) {
        int len = MAX_WORD_LEN + off > text.length() ? text.length() - off
                : MAX_WORD_LEN;

        while (predictDeep < PREDICT_LEVEL) {
            if (len < 1) {
                return;
            }

            String s = text.substring(off, off + len);
            if (len < 2) {
                if (!TextUtils.isCnLetter(text.charAt(off))) {
                    break;
                }
                recurse(node.add(s), off + 1, text, predictDeep + 1);
            } else if (dictionary.contains(s)) {
                recurse(node.add(s), off + s.length(), text, predictDeep + 1);
            }
            len--;
        }
    }
}

package grid.text.participle;

import grid.common.TextUtils;

import java.util.Vector;


public class MechanicalParticiple {

    public Vector<String> partition(String document) {
        Vector<String> vector = new Vector<String>();
        final int docLen = document.length();
        int off = 0;
        char c;
        String seg = "";
        ChunkStream stream = new ChunkStream();

        while (off < docLen) {
            c = document.charAt(off);
            if (TextUtils.isEnLetter(c) || TextUtils.isNumeric(c)) {
                seg += c;
                off++;
            } else if (TextUtils.isCnLetter(c)) {
                if (!TextUtils.isBlank(seg)) {
                    vector.add(seg);
                    seg = "";
                }
                String word = stream.next(document, off);
                if (!TextUtils.isBlank(word)) {
                    vector.add(word);
                    off += word.length();
                }
            } else {
                if (!TextUtils.isBlank(seg)) {
                    vector.add(seg);
                    seg = "";
                }

                /**
                 * TODO: Uncomment the "ELSE IF" clause if you would like to
                 * reserve punctuations
                 */

                // else if (!TextUtils.isBlank("" + c)) { vector.add("" + c); }

                off++;
            }
        }
        if (!TextUtils.isBlank(seg)) {
            vector.add(seg);
        }
        return vector;

    }
}

package grid.text.selector;

import grid.common.TextUtils;


public class CnTextSelector extends CommonTextSelector {

    public CnTextSelector(String document, int minSelectLen, int maxSelectLen) {
        super(document, minSelectLen, maxSelectLen);
    }

    protected void adjustCurLen() {
        while (pos < docLen && !TextUtils.isCnLetter(document.charAt(pos))) {
            pos++;
        }
        for (int i = 0; i < maxSelectLen && pos + i < docLen; i++) {
            if (!TextUtils.isCnLetter(document.charAt(pos + i))) {
                curLen = i;
                if (curLen < minSelectLen) {
                    pos++;
                    adjustCurLen();
                }
                return;
            }
        }

        curLen = pos + maxSelectLen > docLen ? docLen - pos : maxSelectLen;
    }
}

package grid.text.selector;


public class CommonTextSelector implements TextSelector {

    protected String document;

    protected int pos = 0;

    protected int maxSelectLen = 5;

    protected int minSelectLen = 2;

    protected int curLen;

    protected final int docLen;

    public CommonTextSelector(String document, int minSelectLen,
            int maxSelectLen) {
        this.document = document;
        this.minSelectLen = minSelectLen;
        this.maxSelectLen = maxSelectLen;
        docLen = document.length();
        adjustCurLen();
    }

    public void select() {
        pos += ++curLen;
        adjustCurLen();
    }

    protected void adjustCurLen() {
        curLen = pos + maxSelectLen > docLen ? docLen - pos : maxSelectLen;
    }

    public String next() {
        if (curLen < minSelectLen) {
            pos++;
            adjustCurLen();
        }

        if (pos + curLen <= docLen && curLen >= minSelectLen) {
            return document.substring(pos, pos + curLen--);
        } else {
            curLen--;
            // return document.substring(pos, docLen);
            return "";
        }
    }

    public boolean end() {
        return curLen < minSelectLen && curLen + pos >= docLen - 1;
    }

    @Override
    public int getCurPos() {
        return pos;
    }
}

package grid.text.selector;


public interface TextSelector {
    public boolean end();

    public void select();

    public String next();

    public int getCurPos();

}

package grid.test;

import grid.common.TextDatReader;
import grid.text.evolution.NewWordDiscover;
import grid.text.index.CnPreviewTextIndexer;

import java.io.BufferedWriter;
import java.io.File;
import java.io.FileWriter;
import java.io.IOException;
import java.util.Scanner;
import java.util.Set;

public class NewWordDiscoverTest {
    public static void writefile(String m) {

        try {
            File file = new File("result.txt");
            if (!file.exists()) {
                file.createNewFile();
            }
            FileWriter fileWritter = new FileWriter(file.getName(), true);
            BufferedWriter bufferWritter = new BufferedWriter(fileWritter);
            bufferWritter.write(m);
            bufferWritter.close();

        } catch (IOException e) {
            e.printStackTrace();
        }
    }

    @SuppressWarnings("resource")
    public static void main(String[] args) throws Exception {

        File filere = new File("result.txt");
        filere.delete();

        Scanner scan = new Scanner(System.in);
        System.out.println("请输入您要处理的文件名称:\n");
        String path = scan.next();
        File file = new File(path);
        if (!file.exists() || (!file.isFile())) {
            throw new Exception("指定文件不存在！");
        }
        long maxsize = 1024 * 1024 * 1024;
        long size = 1024 * 1024 * 5; 
        long fileLength = file.length();
        if (size <= 0) {
            size = fileLength / 2;
        }

        int num = (fileLength % size != 0) ? (int) (fileLength / size + 1)
                : (int) (fileLength / size);
        if (file.length() >= maxsize) {
            System.out.println("文件大小超出1G，是否开始进行文件切割？1:是 0:否\n");

            int t = scan.nextInt();
            if (t == 1) {
                TextDatReader.divide(path, size);
                System.out.println("切割完成\n");
                System.out.println("结果保存在当前目录下的dat文件夹中\n");

            }


            for (int m = 1; m <= num; m++) {
                String pathdived = "./dat/text" + m + ".dat";
                System.out.println("开始提取第" + m + "个文件……");
                discovrWord(pathdived);
            }

        } else {
            System.out.println("开始提取文件……");
            discovrWord(path);
        }
    }

    private static void discovrWord(String path) throws IOException {
        String document = TextDatReader.read(path);
        NewWordDiscover discover = new NewWordDiscover();
        Set<String> words = discover.discover(document);
        CnPreviewTextIndexer ci = new CnPreviewTextIndexer(document);

        System.out.println("新词个数: " + words.size());
        System.out.println("发现的新词:" + "\n");
        for (String newword : words) {
            System.out.println(newword + "," + ci.count(newword) + "\n");
            writefile(newword + "," + ci.count(newword) + "\n");
        }
    }
}

package grid.test;

import grid.text.participle.MechanicalParticiple;

import java.util.Vector;


public class ParticipleTest {

    private static String document = "我是中国人";

    public static void main(String args[]) {
        MechanicalParticiple participle = new MechanicalParticiple();
        Vector<String> vec = participle.partition(document);
        System.out.println(vec);
    }
}