Skip to content

Commit

Permalink
延迟初始化
Browse files Browse the repository at this point in the history
  • Loading branch information
ysc committed May 23, 2015
1 parent 0715190 commit 1fb15cc
Showing 1 changed file with 9 additions and 1 deletion.
10 changes: 9 additions & 1 deletion src/main/java/org/apdplat/word/analysis/TextSimilarity.java
Original file line number Diff line number Diff line change
Expand Up @@ -43,10 +43,14 @@ public abstract class TextSimilarity implements Similarity{
protected static final Logger LOGGER = LoggerFactory.getLogger(TextSimilarity.class);

//默认分词器
protected Segmentation segmentation = SegmentationFactory.getSegmentation(SegmentationAlgorithm.MaxNgramScore);
private Segmentation segmentation = null;
//是否忽略停用词
protected boolean filterStopWord = false;

public void setSegmentationAlgorithm(SegmentationAlgorithm segmentationAlgorithm){
segmentation = SegmentationFactory.getSegmentation(segmentationAlgorithm);
LOGGER.info("设置分词算法为:"+segmentationAlgorithm.getDes());
}
/**
* 文本1和文本2的相似度分值
* @param text1 文本1
Expand Down Expand Up @@ -112,6 +116,10 @@ public double similarScore(List<Word> words1, List<Word> words2) {
* @return 分词结果
*/
private List<Word> seg(String text){
if(segmentation == null){
//延迟初始化
segmentation = SegmentationFactory.getSegmentation(SegmentationAlgorithm.MaxNgramScore);
}
List<Word> words = segmentation.seg(text);
if(filterStopWord) {
//停用词过滤
Expand Down

0 comments on commit 1fb15cc

Please sign in to comment.