自主学习进行PCA识别,自动处理词库优化.20190427

yaoguangluo · Apr 27, 2019 · 1feb189 · 1feb189
1 parent 5c6f0eb
commit 1feb189
Show file tree

Hide file tree

Showing 6 changed files with 48 additions and 2 deletions.
diff --git a/wordSegment/org/tinos/engine/analysis/Analyzer.java b/wordSegment/org/tinos/engine/analysis/Analyzer.java
@@ -26,4 +26,5 @@ public interface Analyzer {
 	List<String> parserMixedString(String mixedString);
 	Map<String,WordFrequency> parserMixStringByReturnFrequencyMap(String key);
 	void studyNewWord(String study, String token, String posStudy);
+	Map<String, String> getStudyPos();
 }
diff --git a/wordSegment/org/tinos/engine/analysis/imp/AnalyzerImp.java b/wordSegment/org/tinos/engine/analysis/imp/AnalyzerImp.java
@@ -462,4 +462,9 @@ public void studyNewWord(String study, String token, String posStudy) {
 		//learn new pos
 		fHMMList.studyNewPos(study+token, posStudy);
 	}
+
+	@Override
+	public Map<String, String> getStudyPos() {
+		return fHMMList.getStudyPos();
+	}
 }
diff --git a/wordSegment/org/tinos/ortho/fhmm/FHMMList.java b/wordSegment/org/tinos/ortho/fhmm/FHMMList.java
@@ -44,5 +44,6 @@ public interface FHMMList {
 	Map<String, String> getFullPositive();
 	List<String> englishStringToWordsList(String string);
 	Map<Long, Map<String, String>> getWordsForests();
-	void studyNewPos(String string, String posStudy); 
+	void studyNewPos(String string, String posStudy);
+	Map<String, String> getStudyPos(); 
 }
diff --git a/wordSegment/org/tinos/ortho/fhmm/imp/FMHMMListImp.java b/wordSegment/org/tinos/ortho/fhmm/imp/FMHMMListImp.java
@@ -276,4 +276,9 @@ public void studyNewPos(String string, String posStudy) {
 		// TODO Auto-generated method stub
 
 	}
+
+	@Override
+	public Map<String, String> getStudyPos() {
+		return null;
+	}
 }
diff --git a/wordSegment/org/tinos/ortho/fhmm/imp/FMHMMListOneTimeImp.java b/wordSegment/org/tinos/ortho/fhmm/imp/FMHMMListOneTimeImp.java
@@ -19,6 +19,7 @@
 //I will build a collection class for managing this maps. at the next version.
 @SuppressWarnings("unchecked")
 public class FMHMMListOneTimeImp implements FHMMList {
+	private Map<String, String> studyPos;
 	private Map<String, String> posCnToCn;
 	private Map<String, String> posEnToEn;
 	private Map<String, String> posEnToCn;
@@ -75,6 +76,7 @@ public Map<Long, FMHMMNode>[] getMaps() {
 	}
 
 	public void indexMixed() throws IOException {
+		studyPos= new ConcurrentHashMap<>();
 		posCnToCn= new ConcurrentHashMap<>();
 		linkedHashMap= new ConcurrentHashMap<>();
 		listCn= new CopyOnWriteArrayList<>();
@@ -477,6 +479,8 @@ public void indexMixed() throws IOException {
 								.split(StableData.NLP_SYMBO_SLASH)[StableData.INT_ONE]);
 					}
 				}
+				studyPos.put(cInputString.split(StableData.NLP_SYMBO_SLASH)[StableData.INT_ZERO], cInputString
+						.split(StableData.NLP_SYMBO_SLASH)[StableData.INT_ONE]);
 				posCnToCn.put(cInputString.split(StableData.NLP_SYMBO_SLASH)[StableData.INT_ZERO], cInputString
 						.split(StableData.NLP_SYMBO_SLASH)[StableData.INT_ONE]);
 				linkedHashMap = loopLoadForest(cInputString);
@@ -1058,4 +1062,9 @@ public Map<Long, Map<String, String>> getWordsForests() {
 	public void studyNewPos(String string, String posStudy) {
 		posCnToCn.put(string, posStudy);
 	}
+
+	@Override
+	public Map<String, String> getStudyPos() {
+		return this.studyPos;
+	}
 }
diff --git a/wordSegment/org/tinos/ortho/fhmm/imp/poscc.lyg b/wordSegment/org/tinos/ortho/fhmm/imp/poscc.lyg
@@ -26373,4 +26373,29 @@ z/字母
 　　/段落标识标点
 同桌/名词代词
 睡梦/名词
-反差/名词
+反差/名词
+相传/动词
+瑰宝/名词
+盾盾/形谓词
+又怎/副词
+去接/动词
+力棒/名词
+州娃/名词
+食疗/名词
+”，/标点
+潮头/名词
+带动/动词
+也是/副词
+看到/动词
+，“/标点
+流变/动词
+带来/动词
+惊喜/形谓词
+启超/动词
+：“/标点
+。”/标点
+担负/动词
+负起/动词
+传到/动词
+去关/动词
+血脉/名词