Merge branch '201250084' into 'master'

更新了数据挖掘部分

See merge request 0010-qingyun/backend!20

pom.xml

@@ -22,7 +22,7 @@
 			<artifactId>sentistrength_backend</artifactId>
 			<version>1.0</version>
 			<scope>system</scope>
-			<systemPath>${project.basedir}/lib/sentiStrength-v1.0.19.jar</systemPath>
+			<systemPath>${project.basedir}/lib/sentiStrength-v1.0.20.jar</systemPath>
 		</dependency>
 		<dependency>
 			<groupId>org.springframework.boot</groupId>

src/main/java/com/sentistrength/model/vo/dataMining/ClassifyAllArffVO.java

+package com.sentistrength.model.vo.dataMining;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+import java.util.List;
+
+@Data
+@AllArgsConstructor
+@NoArgsConstructor
+@Builder
+public class ClassifyAllArffVO {
+    /**
+     * 类成员变量inFileName, String类型
+     * 输入文件名
+     */
+    String inFileName;
+    /**
+     * 类成员变量inFile, String类型List
+     * 输入文件若干行
+     */
+    List<String> inFile;
+    /**
+     * 类成员变量classifierName, String类型
+     * 分类器名
+     */
+    String classifierName;
+    /**
+     * 类成员变量classifierExclude, String类型
+     * 排除的分类器名
+     */
+    String classifierExclude;
+    /**
+     * 类成员变量features, int类型
+     * 特征数
+     */
+    int features;
+}

src/main/java/com/sentistrength/model/vo/dataMining/ClassifyAllArffWithoutSelectionVO.java

+package com.sentistrength.model.vo.dataMining;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+import java.util.List;
+
+@Data
+@AllArgsConstructor
+@NoArgsConstructor
+@Builder
+public class ClassifyAllArffWithoutSelectionVO {
+    /**
+     * 类成员变量inFileName, String类型
+     * 输入文件名
+     */
+    String inFileName;
+    /**
+     * 类成员变量inFile, String类型List
+     * 输入文件
+     */
+    List<String> inFile;
+    /**
+     * 类成员变量classifierName, String类型
+     * 分类器名
+     */
+    String classifierName;
+    /**
+     * 类成员变量classifierExclude, String类型
+     * 排除的分类器名
+     */
+    String classifierExclude;
+}

src/main/java/com/sentistrength/model/vo/dataMining/DirectClassifyAllArffVO.java

+package com.sentistrength.model.vo.dataMining;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+import java.util.List;
+
+@Data
+@NoArgsConstructor
+@AllArgsConstructor
+@Builder
+public class DirectClassifyAllArffVO {
+    /**
+     * 类成员变量trainFileName String类型
+     * 输入的训练集文件名
+     */
+    String trainFileName;
+    /**
+     * 类成员变量trainFile, String类型List
+     * 输入的训练集文件若干行
+     */
+    List<String> trainFile;
+    /**
+     * 类成员变量evalFile, String类型List
+     * 输入的测试集文件若干行
+     */
+    List<String> evalFile;
+    /**
+     * 类成员变量classifierName, String类型
+     * 分类器名
+     */
+    String classifierName;
+    /**
+     * 类成员变量classifierExclude, String类型
+     * 排除的分类器名
+     */
+    String classifierExclude;
+}

src/main/java/com/sentistrength/model/vo/dataMining/PredictArffClassVO.java

+package com.sentistrength.model.vo.dataMining;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+import java.util.List;
+
+@Data
+@AllArgsConstructor
+@NoArgsConstructor
+@Builder
+public class PredictArffClassVO {
+    /**
+     * 类成员变量inFile, String类型List
+     * 用作训练集的输入文件
+     */
+    List<String> inFle;
+    /**
+     * 类成员变量classifier, String类型
+     * 分类器名
+     */
+    String classifier;
+    /**
+     * 类成员变量unlabelledFile, String类型List
+     * 要分类的输入文件
+     */
+    List<String> unlabelledFile;
+    /**
+     * 类成员变量classFor0, int类型
+     * 标签的最大值
+     */
+    int classFor0;
+}

src/main/java/com/sentistrength/service/Impl/dataMining/DataMiningServiceImpl.java

@@ -10,10 +10,7 @@ import uk.ac.wlv.sentistrength.ClassificationOptions;
 import uk.ac.wlv.sentistrength.ClassificationResources;
 import uk.ac.wlv.sentistrength.TextParsingOptions;
 
-import java.util.Arrays;
-import java.util.HashMap;
-import java.util.List;
-import java.util.Map;
+import java.util.*;
 import java.util.stream.Collectors;
 
 @Service
@@ -44,9 +41,13 @@ public class DataMiningServiceImpl implements DataMiningService {
     public List<String> convertSentimentTextToArff(ConvertSentimentTextToArffVO convertSentimentTextToArffVO) {
         List<String> outFile;
         textParsingOptions.igNgramSize = convertSentimentTextToArffVO.getNgram();
-        outFile = uk.ac.wlv.wkaclass.Arff.convertSentimentTextToArff(convertSentimentTextToArffVO.getInFileName(),convertSentimentTextToArffVO.getInFile(),
-                convertSentimentTextToArffVO.isHeaderLine(),textParsingOptions,classificationOptions,classificationResources,map.get(ClassificationMode.valueOf(convertSentimentTextToArffVO.getMode())),
-                convertSentimentTextToArffVO.getMinFeatureFrequency(),null);
+        try {
+            outFile = uk.ac.wlv.wkaclass.Arff.convertSentimentTextToArff(convertSentimentTextToArffVO.getInFileName(), convertSentimentTextToArffVO.getInFile(),
+                    convertSentimentTextToArffVO.isHeaderLine(), textParsingOptions, classificationOptions, classificationResources, map.get(ClassificationMode.valueOf(convertSentimentTextToArffVO.getMode())),
+                    convertSentimentTextToArffVO.getMinFeatureFrequency(), null);
+        }catch (Exception e) {
+            throw new MyServiceException("DS002", "将情感文本转换为Arff文件出错");
+        }
         return outFile;
     }
 
@@ -58,7 +59,11 @@ public class DataMiningServiceImpl implements DataMiningService {
     @Override
     public List<String> combineTwoArffs(CombineTwoArffsVO combineTwoArffsVO) {
         List<String> outFile;
-        outFile = uk.ac.wlv.wkaclass.Arff.combineTwoArffs(combineTwoArffsVO.getInFile1(),combineTwoArffsVO.getInFile2(),false);
+        try {
+            outFile = uk.ac.wlv.wkaclass.Arff.combineTwoArffs(combineTwoArffsVO.getInFile1(), combineTwoArffsVO.getInFile2(), false);
+        }catch (Exception e) {
+            throw new MyServiceException("DS003", "合并两个Arff文件出错");
+        }
         return outFile;
     }
 
@@ -76,7 +81,7 @@ public class DataMiningServiceImpl implements DataMiningService {
         }
         outFile = uk.ac.wlv.wkaclass.Arff.deleteColumnFromArff(inFile,removeDataColumnVO.getColToRemove());
         if (outFile == null){
-            throw new MyServiceException("DS002", "Arff文件删除列出错");
+            throw new MyServiceException("DS004", "Arff文件删除列出错");
         }
         return outFile;
     }
@@ -93,7 +98,11 @@ public class DataMiningServiceImpl implements DataMiningService {
         if(inFile.size() == 0){
             throw new MyServiceException("DS001", "Arff文件为空");
         }
-        outFile = uk.ac.wlv.wkaclass.Arff.moveColumnToEndOfArff(inFile,moveColumnVO.getColToMove());
+        try {
+            outFile = uk.ac.wlv.wkaclass.Arff.moveColumnToEndOfArff(inFile, moveColumnVO.getColToMove());
+        }catch (Exception e){
+            throw new MyServiceException("DS005", "将Arff文件某一列移动到最后出错");
+        }
         return outFile;
     }
 
@@ -109,7 +118,11 @@ public class DataMiningServiceImpl implements DataMiningService {
         if(inFile.size() == 0){
             throw new MyServiceException("DS001", "Arff文件为空");
         }
-        outFile = uk.ac.wlv.wkaclass.Arff.deleteColAndMoveRemainingFirstColToEnd(inFile,deleteColAndMoveVO.getColToDelete());
+        try {
+            outFile = uk.ac.wlv.wkaclass.Arff.deleteColAndMoveRemainingFirstColToEnd(inFile, deleteColAndMoveVO.getColToDelete());
+        }catch (Exception e){
+            throw new MyServiceException("6", "删除某一列并将第一列移到最后出错");
+        }
         return outFile;
     }
 
@@ -130,8 +143,12 @@ public class DataMiningServiceImpl implements DataMiningService {
         int[][] iData = new int[iAttributeCount + 1][iDataCount + 1];
         double[] fColIG = new double[iAttributeCount + 1];
         String[] sAttributes = new String[iAttributeCount + 1];
-        uk.ac.wlv.wkaclass.Arff.readArffAttributesAndData(inFile, iAttributeCount, iDataCount, sAttributes, iData);
-        uk.ac.wlv.wkaclass.Arff.calculateInformationGainOfData(iData,iAttributeCount,iDataCount,fColIG);
+        try {
+            uk.ac.wlv.wkaclass.Arff.readArffAttributesAndData(inFile, iAttributeCount, iDataCount, sAttributes, iData);
+            uk.ac.wlv.wkaclass.Arff.calculateInformationGainOfData(iData, iAttributeCount, iDataCount, fColIG);
+        }catch (Exception e){
+            throw new MyServiceException("DS007", "计算信息增益出错");
+        }
         return Arrays.stream(fColIG).boxed().collect(Collectors.toList());
     }
 
@@ -147,7 +164,11 @@ public class DataMiningServiceImpl implements DataMiningService {
         if(inFile.size() == 0){
             throw new MyServiceException("DS001", "Arff文件为空");
         }
-        outFile = uk.ac.wlv.wkaclass.Arff.makeArffWithTopNAttributes(inFile,selectTopNFeaturesVO.getTopN());
+        try {
+            outFile = uk.ac.wlv.wkaclass.Arff.makeArffWithTopNAttributes(inFile, selectTopNFeaturesVO.getTopN());
+        }catch (Exception e) {
+            throw new MyServiceException("DS008", "用信息增益最高的N个Attributes新建Arff文件出错");
+        }
         return outFile;
     }
 
@@ -170,9 +191,27 @@ public class DataMiningServiceImpl implements DataMiningService {
     @Override
     public List<String> mergeLabelledAndUnlabelled(MergeLabelledAndUnlabelledVO mergeLabelledAndUnlabelledVO) {
         List<String> outFile;
-        outFile = uk.ac.wlv.wkaclass.Arff.mergeLabelledAndUnlabelledTextFiles(mergeLabelledAndUnlabelledVO.getLabelledFile(),mergeLabelledAndUnlabelledVO.getUnlabelledFile());
+        List<String> labelledFile = handleList(mergeLabelledAndUnlabelledVO.getLabelledFile());
+        if(labelledFile.size() == 0){
+            throw new MyServiceException("DS001", "Arff文件为空");
+        }
+        List<String> unlabelledFile = handleList(mergeLabelledAndUnlabelledVO.getUnlabelledFile());
+        if(unlabelledFile.size() == 0){
+            throw new MyServiceException("DS001", "Arff文件为空");
+        }
+        try {
+            outFile = uk.ac.wlv.wkaclass.Arff.mergeLabelledAndUnlabelledTextFiles(mergeLabelledAndUnlabelledVO.getLabelledFile(), mergeLabelledAndUnlabelledVO.getUnlabelledFile());
+        }catch(Exception e){
+            throw new MyServiceException("DS009", "合并标记好的和未标记好的文件出错");
+        }
         return outFile;
     }
+
+    /**
+     * 删除arff文件中的空行
+     * @param inFile 输入的Arff文件若干行
+     * @return List<String> 删除后的Arff文件若干行
+     */
     public List<String> handleList(List<String> inFile){
         for(int i = 0; i < inFile.size(); i++){
             if(inFile.get(i).equals("")){
@@ -182,4 +221,72 @@ public class DataMiningServiceImpl implements DataMiningService {
         }
         return inFile;
     }
+
+    /**
+     * 根据训练集和分类器预测未标记文件
+     * @param predictArffClassVO 根据训练集和分类器预测未标记文件VO
+     * @return List<String> 结果文件
+     */
+    @Override
+    public List<String> predictArffClass(PredictArffClassVO predictArffClassVO) {
+        List<String> outFile;
+        try {
+            outFile = uk.ac.wlv.wkaclass.PredictClass.predictArffClass(predictArffClassVO.getInFle(), predictArffClassVO.getClassifier(), predictArffClassVO.getUnlabelledFile(), predictArffClassVO.getClassFor0());
+        }catch (Exception e){
+            throw new MyServiceException("DS010", "根据训练集预测未分类文件出错");
+        }
+        return outFile;
+    }
+
+    /**
+     * 根据输入的训练文件和测试文件训练分类器并测试，生成结果文件和总结文件
+     * @param directClassifyAllArffVO 据输入的训练文件和测试文件训练分类器并测试VO
+     * @return Map<List<String>,List<String>> 结果文件和总结文件
+     */
+    @Override
+    public Map<List<String>, List<String>> directClassifyAllArff(DirectClassifyAllArffVO directClassifyAllArffVO) {
+        Map<List<String>, List<String>> outFile;
+        try {
+            outFile = uk.ac.wlv.wkaclass.WekaDirectTrainClassifyEvaluate.directClassifyAllArff(directClassifyAllArffVO.getTrainFileName(), directClassifyAllArffVO.getTrainFile(), directClassifyAllArffVO.getEvalFile(), directClassifyAllArffVO.getClassifierName(), directClassifyAllArffVO.getClassifierExclude());
+        }catch (Exception e) {
+            throw new MyServiceException("DS011", "根据输入的训练文件和测试文件训练分类器并测试出错");
+        }
+        return outFile;
+    }
+
+    /**
+     * 有选择的信息增益交叉验证
+     * @param classifyAllArffVO 有选择的信息增益交叉验证VO
+     * @return Map<List<String>, List<String>> 生成的结果和总结文件
+     */
+    @Override
+    public Map<List<String>, List<String>> classifyAllArff(ClassifyAllArffVO classifyAllArffVO) {
+        Map<List<String>, List<String>> outFile;
+        Random random = new Random();
+        int randomSeed = random.nextInt();
+        try{
+            outFile = uk.ac.wlv.wkaclass.WekaCrossValidateInfoGain.classifyArff(classifyAllArffVO.getInFileName(),classifyAllArffVO.getInFile(), classifyAllArffVO.getClassifierName(), classifyAllArffVO.getClassifierExclude(), randomSeed,classifyAllArffVO.getFeatures());
+        }catch (Exception e){
+            throw new MyServiceException("DS012", "有选择的信息增益交叉验证出错");
+        }
+        return outFile;
+    }
+
+    /**
+     * 无选择的信息增益交叉验证
+     * @param classifyAllArffWithoutSelectionVO 无选择的信息增益交叉验证VO
+     * @return Map<List<String>, List<String>> 生成的结果和总结文件
+     */
+    @Override
+    public Map<List<String>, List<String>> classifyAllArffWithoutSelection(ClassifyAllArffWithoutSelectionVO classifyAllArffWithoutSelectionVO) {
+        Map<List<String>, List<String>> outFile;
+        Random random = new Random();
+        int randomSeed = random.nextInt();
+        try{
+            outFile = uk.ac.wlv.wkaclass.WekaCrossValidateNoSelection.classifyAllArff(classifyAllArffWithoutSelectionVO.getInFileName(), classifyAllArffWithoutSelectionVO.getInFile(), classifyAllArffWithoutSelectionVO.getClassifierName(), classifyAllArffWithoutSelectionVO.getClassifierExclude(), randomSeed);
+        } catch (Exception e){
+            throw new MyServiceException("DS013", "无选择的信息增益交叉验证出错");
+        }
+        return outFile;
+    }
 }

src/main/java/com/sentistrength/service/Interface/dataMining/DataMiningService.java

@@ -4,7 +4,7 @@ package com.sentistrength.service.Interface.dataMining;
 import com.sentistrength.model.vo.dataMining.*;
 
 import java.util.List;
-import java.util.ListIterator;
+import java.util.Map;
 
 public interface DataMiningService {
     /**
@@ -56,8 +56,36 @@ public interface DataMiningService {
     void setOptions(SetOptionsVO setOptionsVO);
     /**
      * 合并标记好的文件和未标记好的文件
-     * @param mergeLabelledAndUnlabelledVO 合并标记好的和未标记好的文件
+     * @param mergeLabelledAndUnlabelledVO 合并标记好的和未标记好的文件VO
      * @return List<String> 合并结果文件的若干行
      */
     List<String> mergeLabelledAndUnlabelled(MergeLabelledAndUnlabelledVO mergeLabelledAndUnlabelledVO);
+
+    /**
+     * 根据训练集和分类器预测未标记文件
+     * @param predictArffClassVO 根据训练集和分类器预测未标记文件VO
+     * @return List<String> 生成结果文件若干行
+     */
+    List<String> predictArffClass(PredictArffClassVO predictArffClassVO);
+
+    /**
+     * 根据输入的训练文件和测试文件训练分类器并测试，生成结果文件和总结文件
+     * @param directClassifyAllArffVO 据输入的训练文件和测试文件训练分类器并测试VO
+     * @return 生成的结果文件和测试文件
+     */
+    Map<List<String>,List<String>> directClassifyAllArff(DirectClassifyAllArffVO directClassifyAllArffVO);
+
+    /**
+     * 有选择的信息增益交叉验证
+     * @param classifyAllArffVO 有选择的信息增益交叉验证VO
+     * @return Map<List<String>, List<String>> 生成的结果和总结文件
+     */
+    Map<List<String>,List<String>> classifyAllArff(ClassifyAllArffVO classifyAllArffVO);
+
+    /**
+     * 无选择的信息增益交叉验证
+     * @param classifyAllArffWithoutSelectionVO 无选择的信息增益交叉验证VO
+     * @return Map<List<String>, List<String>> 生成的结果和总结文件
+     */
+    Map<List<String>,List<String>> classifyAllArffWithoutSelection(ClassifyAllArffWithoutSelectionVO classifyAllArffWithoutSelectionVO);
 }

src/main/java/com/sentistrength/web/controller/dataMining/DataMiningController.java

@@ -108,4 +108,44 @@ public class DataMiningController {
     public Response mergeUnlabelledAndLabelled(@RequestBody MergeLabelledAndUnlabelledVO mergeLabelledAndUnlabelledVO){
         return Response.buildSuccess(dataMiningService.mergeLabelledAndUnlabelled(mergeLabelledAndUnlabelledVO));
     }
+
+    /**
+     * 根据训练集来预测未标记的文件
+     * @param predictArffClassVO 根据训练集来预测未标记的文件VO
+     * @return Response 返回前端的信息
+     */
+    @PostMapping("/predictArffClass")
+    public Response predictArffClass(@RequestBody PredictArffClassVO predictArffClassVO) {
+        return Response.buildSuccess(dataMiningService.predictArffClass(predictArffClassVO));
+    }
+
+    /**
+     * 根据输入的训练文件和测试文件训练分类器并测试，生成结果文件和总结文件
+     * @param directClassifyAllArffVO 据输入的训练文件和测试文件训练分类器并测试VO
+     * @return Response 返回前端的信息
+     */
+    @PostMapping("directClassifyAllArff")
+    public Response directClassifyAllArff(@RequestBody DirectClassifyAllArffVO directClassifyAllArffVO){
+        return Response.buildSuccess(dataMiningService.directClassifyAllArff(directClassifyAllArffVO));
+    }
+
+    /**
+     * 有选择的信息增益交叉验证
+     * @param classifyAllArffVO 有选择的信息增益交叉验证VO
+     * @return Response 返回前端的信息
+     */
+    @PostMapping("classifyAllArff")
+    public Response classifyAllArff(@RequestBody ClassifyAllArffVO classifyAllArffVO){
+        return Response.buildSuccess(dataMiningService.classifyAllArff(classifyAllArffVO));
+    }
+
+    /**
+     * 无选择的信息增益交叉验证
+     * @param classifyAllArffWithoutSelectionVO 无选择的信息增益交叉验证VO
+     * @return Response 返回前端的信息
+     */
+    @PostMapping("classifyAllArffWithoutSelection")
+    public Response classifyAllArffWithoutSelection(@RequestBody ClassifyAllArffWithoutSelectionVO classifyAllArffWithoutSelectionVO) {
+        return Response.buildSuccess(dataMiningService.classifyAllArffWithoutSelection(classifyAllArffWithoutSelectionVO));
+    }
 }