所以我有这个包装程序，可以让我从方法中返回两个数量。
**包装类**
public class Words
{
private String leftWords;
private String rightWords;
public Words(String leftWords, String rightWords) {
this.leftWords = leftWords;
this.rightWords = rightWords;
}
public String getLeftWords() {
return leftWords;
}
public String getRightWords() {
return rightWords;
}
@Override
public int hashCode()
{
final int prime = 31;
int result = 1;
result = prime * result
+ ((leftWords == null) ? 0 : leftWords.hashCode());
result = prime * result
+ ((rightWords == null) ? 0 : rightWords.hashCode());
return result;
}
@Override
public boolean equals(Object obj)
{
if (this == obj)
return true;
if (obj == null)
return false;
if (getClass() != obj.getClass())
return false;
Words other = (Words) obj;
if (leftWords == null)
{
if (other.leftWords != null)
return false;
}
else if (!leftWords.equals(other.leftWords))
return false;
if (rightWords == null)
{
if (other.rightWords != null)
return false;
}
else if (!rightWords.equals(other.rightWords))
return false;
return true;
}
}
我想与之相关的方法是：
private static Map <Set<String>,Set<Words>> getLeftRightWords(LinkedHashMap<Set<String>,Set<Integer>> nnpIndexTokens, NLChunk chunk) throws FileNotFoundException
{
// Map <Set<String>,Set<Integer>> nnpMap = new LinkedHashMap<Set<String>, Set<Integer>>();
Map <Set<String>,Set<Words>> contextMap = new LinkedHashMap<Set<String>, Set<Words>>();
Set<Words> leftRightWords = new HashSet<Words>();
//for(NLChunk chunk : sentence.getChunks()){
if(chunk.getStrPostags().contains("NNP")){
String leftWords = "";
String rightWords = "";
int chunkStartIndex = chunk.getStartIndex();
int chunkEndIndex = chunk.getEndIndex();
//nnpMap = getNNPs(chunk);
String previous = null;
int previousNnpEndIndex = 0;
int previousNnpStartIndex = 0;
for (Map.Entry<Set<String>, Set<Integer>> entry : nnpIndexTokens.entrySet()){
for (Iterator<String> i = entry.getKey().iterator(); i.hasNext();){
Set<Integer> entryIndex = null;
int nnpStartIndex = 0;
int nnpEndIndex = 0;
String currentElement = i.next();
//Deriving values for beginning and ending of chunk
//and beginning and ending of NNP
if (!(entry.getValue().isEmpty())){
if (currentElement.trim().split(" ").length > 1){
entryIndex = entry.getValue();
nnpStartIndex = entryIndex.iterator().next();
nnpEndIndex = getLastElement(entryIndex);
}
else {
entryIndex = entry.getValue();
nnpStartIndex = entryIndex.iterator().next();
nnpEndIndex = nnpStartIndex;
}
}
if(!(chunkStartIndex<=nnpStartIndex && chunkEndIndex>=nnpEndIndex)){
continue;
}
//Extracting LEFT WORDS of the NNP
//1)If another NNP is present in left words, left words of current NNP start from end index of previous NNP
if (previous != null && chunk.toString().substring(chunkStartIndex, nnpStartIndex).contains(previous)){
int leftWordsEndIndex = nnpStartIndex;
int leftWordsStartIndex = previousNnpEndIndex;
for (NLWord nlword : chunk.getTokens())
{
if(nlword.getIndex()>=leftWordsStartIndex
&& nlword.getIndex()<leftWordsEndIndex )
leftWords+=nlword.getToken() +" ";
}
System.out.println("LEFT WORDS:" + leftWords+ "OF:"+ currentElement);
}
//2) If no left words are present
if (chunkStartIndex == nnpStartIndex){
System.out.println("NO LEFT WORDS");
}
//3) Normal case where left words consist of all the words left of the NNP starting from the beginning of the chunk
else {
for (NLWord nlword : chunk.getTokens())
{
if(nlword.getIndex()>=chunkStartIndex
&& nlword.getIndex()<nnpStartIndex )
leftWords+=nlword.getToken() +" ";
}
System.out.println("LEFT WORDS:" + leftWords+ "OF:"+ currentElement);
}
//Extracting RIGHT WORDS of NNP
if (entry.getKey().iterator().hasNext()){// entry.getKey().iterator().hasNext()){
String nextElement = entry.getKey().iterator().next();
//1)If another NNP is present in right words, right words of current NNP start from end index of current NNP to beginning of next NNP
if (nextElement !=null && nextElement != currentElement && chunk.toString().substring(entry.getValue().iterator().next(), chunkEndIndex).contains(nextElement)){
int rightWordsStartIndex = entryIndex.iterator().next();
int rightWordsEndIndex = entry.getValue().iterator().next();
//String rightWord="";
for (NLWord nlword : chunk.getTokens())
{
if(nlword.getIndex()>=rightWordsStartIndex
&& nlword.getIndex()<rightWordsEndIndex )
rightWords+=nlword.getToken() +" ";
}
System.out.println("LEFT WORDS:" + leftWords+ "OF:"+ currentElement);
}
}
//2) If no right words exist
if(nnpEndIndex == chunkEndIndex){
System.out.println("NO RIGHT WORDS");
//continue;
}
//3) Normal case where right words consist of all the words right of the NNP starting from the end of the NNP till the end of the chunk
else {
for (NLWord nlword : chunk.getTokens())
{
if(nlword.getIndex()>=nnpEndIndex+1
&& nlword.getIndex()<=chunkEndIndex )
rightWords+=nlword.getToken() +" ";
}
System.out.println("RIGHT WORDS:" + rightWords+ "OF:"+ currentElement);
}
if (previous == null){
previous = currentElement;
previousNnpStartIndex = nnpStartIndex;
previousNnpEndIndex = nnpEndIndex;
}
Words contextWords = new Words(leftWords.toString(), rightWords.toString());
leftRightWords.add(contextWords);
}
contextMap.put(entry.getKey(), leftRightWords);
}//nnps set
}
System.out.println(contextMap);
return contextMap;
}
正如你可以看到我在这个方法中尝试做的是采用专有名词并提取该专有名词的左右词.Eg为一个块“罗德岛同胞解决方案提供商”我的输出是：
左词：OF：罗德岛 正确的词：解决方案提供商OF：罗德岛
现在我想将这些放在罗德岛是关键的地图中，其价值是解决方案提供者和研究员。
当我尝试打印此地图时，输出get为：
{[罗德岛] = [com.gyan.siapp.nlp.test.Words@681330f0]}
如何获得正确的输出？
答案 0 :(得分：0)
我不知道这是否是唯一的问题，但您的班级Words没有覆盖
toString()方法。
不确定您的Java技能水平。很抱歉，如果我发布您熟悉的内容。
System.out.println(...)调用toString()方法获取对象的消息。
通过使用您自己的实现覆盖默认值
@Override
public String toString(){
return "leftWords: "+leftWords+", rightWords: "+rightWords;
}
您将com.gyan.siapp.nlp.test.Words@681330f0更改为您自己的输出。