To improve the efficiency, developers tend to use APIs to avoid reinventing wheels in the development of Apps. However, there are thousands of APIs for various purposes, so it is difficult for developers to identify suitable APIs according to the functionalities to be realized. App stores manage millions of products, which embody the experience and wisdom of developers, and they provide valuable data resource for solving this problem. By summarizing the API usage for the same or similar functionalities in Apps, reusable knowledge can be mined for the API recommendation. In this paper, we utilize the data resource in App stores and provide an API recommendation method for the development of Android Apps. Firstly, by using UI elements as the bridge, we establish mapping relationships between functionalities and APIs. Secondly, we build a framework to describe functionalities of Apps in the same category, and utilize relationships between functionalities and APIs to construct the API knowledge for each node in the framework. Thirdly, we identify nodes according to queried features and show the API knowledge to developers by giving recommendation lists. We conducted experiments based on Google Play, and the result shows that our method has a good recommendation performance.

为了提高效率，开发人员倾向于使用API​​来避免在Apps开发中重新发明轮子。但是，有成千上万的各种用途的API，因此开发人员很难根据要实现的功能来确定合适的API。应用商店管理着数百万种产品，这些产品体现了开发人员的经验和智慧，它们为解决此问题提供了宝贵的数据资源。通过汇总应用程序中相同或相似功能的API使用情况，可以挖掘可重复使用的知识以用于API建议。在本文中，我们利用App Store中的数据资源，并为开发Android Apps提供了API推荐方法。首先，通过使用UI元素作为桥梁，我们在功能和API之间建立映射关系。其次，我们构建了一个框架来描述同一类别中的Apps功能，并利用功能和API之间的关系来构建框架中每个节点的API知识。第三，我们根据查询的特征识别节点，并通过给出推荐列表向开发人员展示API知识。我们基于Google Play进行了实验，结果表明我们的方法具有良好的推荐性能。