中文摘要：

构件识别是软件复用中的一个关键问题，通过分析领域业务模型，得到一组具有较高复用价值和复用性能的业务构件，以支持复用。归纳了构件识别的分类，从识别过程、业务模型形态、识别目标、识别策略等角度探讨了业务构件识别的技术体系，把构件识别方法分为领域分析法、聚类分析法、创建-读取-写入-删除法和其他方法4类，并分析了各自的技术特征，比较了各自的优劣。探讨了业务构件识别方法研究中的不足之处及其原因。最后，探讨了该领域最有前途的发展趋势。

英文摘要：

By analyzing domain business models, a set of business components with high reuse value and good reuse performance was obtained to support reuse. Based on the concept and classification of component identification, its technical architecture was discussed from four views, i.e. , identification process, form of business models, identification goals, and identification strategies. Then, those component identification methods in literatures were classified as, domain-analysis-based methods, cohesion-coupling-based clustering methods, Create Read Update Delete （CRUD） matrix based methods, and other methods. And their respective technical characteristics were analyzed, as well as their advantages and disadvantages were also compared. In addition, deficiency of study on component identification was discussed and reasons were explained at the same time. Finally, the most promising perspective in this field was discussed.