中文摘要：

标注技术广泛应用于标注Web2.0应用的对象中。标签能支持Web服务更好地被理解分类和发现，这在面向对象的软件系统中是很重要的。论文提出一种新颖的方法，描述文本和标签都用来帮助API服务聚类。此外还提出一种标签推荐策略，为了验证论文方法的有效性，在API服务数据集上进行了实验，实验结果表明进行了标签推荐后的服务更好改进了聚类的性能。

英文摘要：

Tagging technique is widely used to annotate objects in Web2.0applications.Tags can support web service under?standing,categorizing and discovering,which are important tasks in a service-oriented software system.In this paper,a novel ap?proach is proposed that both description and tags are utilized for API service clustering in this approach.In addition,a tag recom?mendation strategy is presented.In order to verify the validity of this method,experiments are carried out on the API service datasets.The experimental results show that the service after the tag recommendation better improve the clustering performance.