基于网络智能的群体软件开发行为建模与分析方法

针对云计算时代超量软件服务的迫切需求，以群体软件工程为研究取向，采用网络智能方法，在实证数据分析和挖掘的基础上，研究群体软件开发交互行为建模与分析方法。重点研究内容包括：构建软件开发者网络与软件网络为基础的社会-技术网络模型，结合具体情境分析基于该模型的多粒度协同演化对交互行为的影响与作用，提出交互行为的度量指标，发现典型交互行为模式，提供群体软件开发过程的优化原则。同时建设大规模群体交互行为库，提供交互行为建模与分析工具。研究成果将阐明社会-技术网络中大众参与、自组织、分布式知识型生产中交互行为的内在机制，为理解、预测和影响群体软件开发中的个体、群体和组织等行为提供理论依据和方法支持，有利于优化群体软件开发过程，为提高群体软件开发效率与质量水平奠定基础。

In the emerging cloud computing era, collaborative software engineering has become an increasingly important approach to serve the urgent needs for mass software services. This proposal focuses on modeling and analytical methods for interactive behaviors of collaborative software development based on empirical data analysis and data mining. Network intelligence is the enabling technology for this study. .The major research mechanisms are as follows. A socio-technical network model will be constructed based on developer networks and software networks. Then, how multi-granularity coevolution impacts on interactive behaviors will be studied, while contexts being considered in the socio-technical network model. Some metrics will be proposed to measure the interactive behaviors; typical behavior interaction patterns will be identified; and optimization principles supporting collaborative software development will established. In addition, a large-scale interactive behaviors repository will be established and a suite of software tools will be developed to support this research. .The results of this study will elucidate the inherent mechanism of interactive behaviors of mass participation，self-organizing and distributed knowledge-based peer production in socio-technical networks, and will provide theoretical bases for understanding, predicting and affecting the behaviors of individuals, groups and organizations. Our approach will optimize the development process, and will lay a solid foundation to improve the productivity and quality of software through collaborative development.