Measuring The Contribution Of Developers In Open Source Software Community

Open source community and open source software is developing rapidly,more and more developers to participate in open source projects,many developers in a specific project performance and the role is different,the developer behavior,including code activities and Non-code-level socialization activities,their activities to promote the development of the project,a project's evolutionary process is the cumulative contribution of all the developers to promote.Research on their behavioral characteristics is a focus of the software engineering field.Through the study of the contributor characteristics of the developer,it can help the researcher discover the specific impact of the developer on the project and further understand the evolution process of the whole project and find some useful rules.In software engineering,the composition of the contribution of the developer can guide the project owner to understand the different developers of development preferences,so a reasonable staffing,and thus more efficient to promote the development of the project.This paper presents a general method for data collection of Github open source community,including data crawling and feature variable extraction,merging of heterogeneous data sources into structured databases,and finding a high accuracy combination of the same true identity but different An algorithm for submitting identity information.By introducing the contributions of radar maps,contribution to the evolution of ringshaped graph and parallel coordinate linear map and other visual mining technology,reducing the complexity of multidimensional variable mining research in the evolution of individual contributions to the summary of the open source project in the four Large class role,and found that long-term developers contribute to the characteristics of cross-version.In the existing related research,it dig deeper into developer contribution index very few research and explanation this paper focuses on the evaluation of the developer contribution to the indicators involved in research and development process of the various activities and contributions to the development of Finally,we select three significant variables to evaluate the contribution,which are the number of rows that the developer adds to the code,and the repair of the defect.In this paper,The number of participants and the number of comments in the research process also confirmed and found some laws.In this paper,we use software depot data mining technology combined with multivisual mining technology to study the contributions of developers in the open source environment and the evolution of the contribution of the law.The research includes the correlation and the importance ranking among the relevant metrics.Based on the measurement index,the contributions of the individual developers are divided according to the tag,and the multi-linear regression model is established by stepwise regression.,we screening the contribution metrics,and finally making a realistic interpretation of the model,which is useful for evaluating the contribution of many developers to large open source projects.