A novel method for software bug report assignment

During the development of software and electronic devices, it is inevitable to make mistakes. In large, developed companies, assigning a request to the right development team or even a department is not an easy task. Often, the creation of software bug reports and assignment to groups is also formalized by appropriate processes. However, there is an evident lack of approaches based on machine learning algorithms using specific contexts and/or additional insights coming from known hierarchical structure; for example, combination of models related to different levels especially with usage of thresholds tuning related to confidence level. Objective: The paper reviews state of the art solutions for software bug report assignment with the focus on targets or usage of the organization structure. The paper presents a novel method using specific organization structure for bug assignment. Method: There were conducted studies on company internal data contains report of software bugs. Experimental results comes from both machine learning predictions of simulations as well as analyzed data from introduced software bug report assignment solution in company. The work also covers conclusive research about potential benefits of the application of automated assignment of bug reports. Results: The results have been obtained from both simulations and real production uses in company application. Both scenarios showed the benefits of the proposed method of software bug report assignment. The paper presents results of human predictions and compares them to the results generated by machine learning models. Conclusion: The paper presents a novel method for software bug report assignment. The approach is based on outputs from multiple models with the specific context of organisation structure. The proposed method shows its effectiveness in both simulations as well as in real usage in the company.