面向大众生产的软件信息网络挖掘及其应用研究

项目名称： 面向大众生产的软件信息网络挖掘及其应用研究

项目编号： No.61472430

项目类型： 面上项目

立项/批准年度： 2015

项目学科： 自动化技术、计算机技术

项目作者： 尹刚

作者单位： 中国人民解放军国防科技大学

项目金额： 83万元

中文摘要： 大众化软件生产模式正快速形成一种基于互联网的新型软件生产力，在软件开发和应用各个环节发挥了巨大作用。大众化软件生产的数据规模日趋庞大、协同范围不断扩展、项目管理高度精简，这些全球化特征使其在软件复用、协同开发、知识管理等环节面临诸多挑战，迫切需要新的理论指导和工具支持。本项目发现，大众化软件生产活动广泛活跃于软件协同开发社区和软件知识分享社区，前者主要托管结构化的软件过程制品，后者富含语义化的软件知识，两者具有很强的互补性和关联性，对其综合利用有助于在更大范围内系统的考察上述挑战的研究方案。据此，本项目首次提出软件信息网络挖掘问题，将异质信息网络分析系统的引入软件仓库挖掘，基于软件协同开发数据和知识分享数据，研究高质量软件信息网络的构造和分析面临的一系列基础理论问题，突破面向可信资源定位、开发任务预测、社区知识推荐等软件信息网络挖掘关键技术，为大众化软件生产提供新的技术理论和支撑工具。

中文关键词： 软件开发；大众生产；信息网络；软件仓库挖掘

英文摘要： Crowd-based software production models are rapidly becoming a kind of new software productivity, and have great impacts on many stages of software development and applications. Crowd-based software production generates large amounts of software data, continuously expands its collaboration scopes, and highly simplifies its project management. These globalization features make crowd-based software production face many challenges in software reuse, collaboration development, and knowledge management, and urgently requires new theories and tools. We observe that, crowd-based software developers and users are active in software collaboration development communities and knowledge sharing communities, in which the former manages structural software artifacts while the latter consists semantical software knowledge, and they have strong complementarity and associations between each other. To make comprehensive utilization of these two kinds of data will help study the above challenges systematically in a broader extent. Hereby, this project for the first time proposes a new problem named software information network mining, systematically introducing the theory of heterogeneous network analysis into software repository mining, based on the data from above two kinds of software communities, conducts the research on a series of basic theories for construction and analysis of software information network, and breaks through the key technologies of software information network mining for trustworthy resource location, development task prediction and community knowledge recommendation. This project will provide new technology theories and supporting tools for crowd-based software production.

英文关键词： Software Development;Crowd-based Production;Information Network;Mining of Software Repository