双名词拆分细节 前缀优化 辨别上海南 为 上海 南 20190804 罗瑶光

behaviorMap/org/tinos/behavior/test/LiterarinessLevelTest.java

@@ -8,8 +8,8 @@ public class LiterarinessLevelTest{
 	double[] literarinessLevel;
 	double literarinessDuration;
 	public void getEducationLevel(double[][] measurement) {
-		literarinessLevel = new double[measurement.length];
-		for(int i = StableData.INT_ZERO; i < measurement.length; i++) {
+		literarinessLevel= new double[measurement.length];
+		for(int i= StableData.INT_ZERO; i< measurement.length; i++) {
 			literarinessLevel[i]=(measurement[i][StableData.INT_ZERO]*measurement[i][StableData.INT_THREE])
 					/measurement[i][StableData.INT_ONE];
 			literarinessDuration+=literarinessLevel[i];

wordSegment/org/tinos/engine/pos/imp/POSControllerImp.java

@@ -124,6 +124,13 @@ public int chuLiMingCiOfThree(Map<String, String> wordsForest, List<String> outp
 					return countInputStringLength;
 				}
 			}
+			if(StableData.INT_ZERO== fixWord[StableData.INT_ZERO].length()
+				|| StableMaps.jieCi.containsKey(fixWord[StableData.INT_ZERO].toString())){
+				if (StableMaps.CiTwo.containsKey(strings[StableData.INT_ONE])){
+					countInputStringLength= parserFirstTwoCharOfThree(countInputStringLength, outputList, strings, fixWord);
+					return countInputStringLength;
+				}
+			}
 			countInputStringLength= parserFirstCharOfThree(countInputStringLength, outputList, strings, fixWord);
 			return countInputStringLength;
 		}

wordSegment/org/tinos/test/DemoPOSforSpecial.java

@@ -9,29 +9,29 @@
 
 public class DemoPOSforSpecial {
 	public static void main(String[] args) throws IOException {
-		Analyzer analyzer = new CogsBinaryForestAnalyzerImp();
+		Analyzer analyzer= new CogsBinaryForestAnalyzerImp();
 		analyzer.initMixed();
-		Map<String, String> nlp = analyzer.getPosCnToCn();
-		List<String> sets = new ArrayList<>();
-		String[] ss = new String[1];
-		String[] ss1 = new String[1];
-		ss[0] = "存在这是非常常是非常愚蠢的为主要求知不断的提高自身的知识的不锻炼改变化";
-		ss1[0] = "나는 일을 준비 하고 있다 ";
-		for (int i = 0; i < ss.length; i++) {
-			System.out.println("超级变态复杂病句-->" + ss[i]);
-			sets = analyzer.parserMixedString(ss[i].replace("  ", " "));//词性分析
+		Map<String, String> nlp= analyzer.getPosCnToCn();
+		List<String> sets= new ArrayList<>();
+		String[] ss= new String[1];
+		String[] ss1= new String[1];
+		ss[0]= "存在这是非常常是非常愚蠢的为主要求知不断的提高自身的知识的不锻炼改变化나는일을준비하고있다";
+		ss1[0]= "나는 일을 준비 하고 있다 ";
+		for (int i= 0; i< ss.length; i++) {
+			System.out.println("超级变态复杂病句-->"+ ss[i]);
+			sets= analyzer.parserMixedString(ss[i].replace("  ", " "));//词性分析
 			out.print("分析处理真实结果-->");
-			for (int j = 0; j < sets.size(); j++) {
+			for (int j= 0; j< sets.size(); j++) {
 				if (!sets.get(j).replaceAll("\\s+", "").equals("")) {
-					out.print(sets.get(j) + " ");
+					out.print(sets.get(j)+ " ");
 				}
 			}
 			out.println();
 			out.println("期望得到分词效果-->" + ss1[i]);
-			for (int k = 0; k < sets.size(); k++) {
+			for (int k= 0; k< sets.size(); k++) {
 				if (!sets.get(k).replaceAll("\\s+", "").equals("")) {
 					nlp.get(sets.get(k));
-					out.println(sets.get(k) + "/" + nlp.get(sets.get(k)) + "  ");
+					out.println(sets.get(k)+ "/"+ nlp.get(sets.get(k))+ "  ");
 				}
 			}
 			out.println("");