One question which frequently arises within the context of artifacts stored in a bug tracking repository is: "who should work on this bug report?" A number of approaches exist to semi-automatically identify and recommend developers, e.g. using machine learning techniques and social networking analysis. In this work, we propose a new approach for assignee recommendation leveraging user activities in a bug tracking repository. Within the bug tracking repository, an activity profile is created for each user from the history of all his activities (i.e. review, assign, and resolve). This profile, to some extent, indicates the user's role, expertise, and involvement in this project. These activities influence and contribute to the identification and ranking of suitable assignees. In order to evaluate our work, we apply it to bug reports of three different projects. Our results indicate that the proposed approach is able to achieve an average hit ratio of 88%. Comparing this result to the LDA-SVM-based assignee recommendation technique, it was found that the proposed approach performs better.
Model-based CASE tools provide mechanisms to capture and store heterogeneous artifacts produced during the software development process. These tools incorporate a metamodel describing artifact types and traceability links. Although model-based CASE tools provide required means to create and link different artifact types, still the process of linking artifacts is primarily manual resulting in missing or broken traceability links. This paper proposes a novel approach to create and utilize a project-specific ontology derived from the textual and structural information available in the development artifacts to assist the traceability link creation process. We discuss the benefits and challenges of incorporating the proposed approach in a model-based CASE tool.
Developing mathematical models for reliable approximation of epidemic spread on a network is a challenging task, which becomes even more difficult when a wireless network is considered, because there are a number of inherent physical properties and processes which are apparently invisible. The aim of this paper is to explore the impact of several abstract features including trust, selfishness and collaborative behavior on the course of a network epidemic, especially when considered in the context of a wireless network. A five-component differential epidemic model has been proposed in this work. The model also includes a latency period, with a possibility of switching epidemic behavior. Bilinear incidence has been considered for the epidemic contacts. An analysis of the long term behavior of the system reveals the possibility of an endemic equilibrium point, in addition to an infection-free equilibrium. The paper characterizes the endemic equilibrium in terms of its existence conditions. The system is also seen to have an epidemic threshold which marks a well-defined boundary between the two long-term epidemic states. An expression for this threshold is derived and stability conditions for the equilibrium points are also established in terms of this threshold. Numerical simulations have further been used to show the behavior of the system using four different experimental setups. The paper concludes with some interesting results which can help in establishing an interface between epidemic spread and collaborative behavior in wireless networks.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.