应用程序编程接口（Application Programming Interface, API）是开源软件复用是主要模式。当前大量研究聚焦在复用API或是API模式上，开发者查找、调用API实现特定任务的编码主要难点在于需求表述、API适配和参数填充等问题。为此，本项目提出了自然语言编程接口（Natural Language Interface, NLI）的概念，把API这种程序代码层面的复用单元提升到功能层次的复用。项目研究NLI的自动构造及辅助代码生成技术，包括面向NLI的软件功能特征提取方法、功能特征与API模式的关联方法、NLI快速构造与管理机制、基于NLI的交互式代码生成技术等。此研究，可使开发者在功能层次精确描述复用需求，精准匹配需求对应的API模式，提高代码生成中API适配和参数填充的自动化程度。项目还将研发相应系统原型，在产业实践中以现实案例验证研究的有效性和实用性。

Open-source software typically evolves into the application programming interface (API) for software reuse. Existing research focuses on searching and invoking related APIs or API patterns. Given a specific programming task, developers need to deal with a lot of issues such as requirement understanding, API adaptation and automatically parameter filling. This project proposes a new concept of Natural Language Interface (NLI), which aims to promote software reuse from code level (i.e. APIs) to functional level. In this project, we do search on software reuse from API to NLI, and try to automatically construct NLI and generate code based on NLI. We will propose NLI-oriented software functional feature extraction approach, traceability recovery approach between functional features and API patterns, rapid NLI building mechanism and smart code completion approach based on NLI. As a result, NLI will enable developers to accurately describe their reuse requirements at the functional level, precisely match their requirements to API patterns, and improve the automation of API adaptation and parameter filling. We will also develop corresponding system prototypes so as to verify the validity and practicality of our research with real-world cases from industrial practice.