As data in the telecommunications industry becomes more voluminous and complex, extracting insightful information requires efficient and scalable systems that can effectively link and manage this data. This paper introduces a novel, multi-layered approach to managing interlinked data for Cloud Radio Access Network (CloudRAN) at Ericsson, utilizing Knowledge Graphs (KGs). Our system is structured into six distinct layers, each focusing on a specific aspect of managing interlinked data. This division enhances clarity and manageability, and promotes effective teamwork and collaborative development. A cornerstone of our architecture is its modularity, which enables the flexible exchange of components, such as the triple store, with minimal impact on the system's operations, ensuring longevity and adaptability to evolving technological trends. Moreover, we introduce novel applications in knowledge graph summarization and semantic search, specifically engineered for industrial decision-making. These innovations provide concise insights and actionable intelligence, fostering rapid and informed decision-making processes crucial for industry professionals. Finally, we discuss the lessons learned from deploying and utilizing this six-layer framework.

This paper presents our experience developing a Llama-based chatbot for question answering about continuous integration and continuous delivery (CI/CD) at Ericsson, a multinational telecommunications company. Our chatbot is designed to handle the specificities of CI/CD documents at Ericsson, employing a retrieval-augmented generation (RAG) model to enhance accuracy and relevance. Our empirical evaluation of the chatbot on industrial CI/CD-related questions indicates that an ensemble retriever, combining BM25 and embedding retrievers, yields the best performance. When evaluated against a ground truth of 72 CI/CD questions and answers at Ericsson, our most accurate chatbot configuration provides fully correct answers for 61.11% of the questions, partially correct answers for 26.39%, and incorrect answers for 12.50%. Through an error analysis of the partially correct and incorrect answers, we discuss the underlying causes of inaccuracies and provide insights for further refinement. We also reflect on lessons learned and suggest future directions for further improving our chatbot's accuracy.

Today, software developers work on complex and fast-moving projects that often require instant assistance from other domain and subject matter experts. Chat servers such as Discord facilitate live communication and collaboration among developers all over the world. With numerous topics discussed in parallel, mining and analyzing the chat data of these platforms would offer researchers and tool makers opportunities to develop software tools and services such as automated virtual assistants, chat bots, chat summarization techniques, Q&A thesaurus, and more.

Knowledge and experience are touted as both the necessary and sufficient conditions to make a person an expert. This paper attempts to investigate this issue in the context of software development by studying software developer's expertise based on their activity and experience on GitHub and Stack Overflow platforms. We study how developers themselves define the notion of an "expert", as well as why or why not developers contribute to online collaborative platforms. We conducted an exploratory survey with 73 software developers and applied a mixed methods approach to analyze the survey results. The results provided deeper insights into how an expert in the field could be defined. Further, the study provides a better understanding of the underlying factors that drive developers to contribute to GitHub and Stack Overflow, and the challenges they face when participating on either platform.The quantitative analysis showed that JavaScript remains a popular language, while knowledge and experience are the key factors driving expertise. On the other hand, qualitative analysis showed that soft skills such as effective and clear communication, analytical thinking are key factors defining an expert. We found that both knowledge and experience are only necessary but not sufficient conditions for a developer to become an expert, and an expert would necessarily have to possess adequate soft skills. Lastly, an expert's contribution to GitHub seems to be driven by personal factors, while contribution to Stack Overflow is motivated more by professional drivers (i.e., skills and expertise). Moreover, developers seem to prefer contributing to GitHub as they face greater challenges while contributing to Stack Overflow.