Lossan Bonde2025-05-142025-05-142019-062349-5162https://irepository.aua.ac.ke/handle/123456789/529Full text ArticleEmail-based IT support refers to IT support systems whereby an email address is set to receive requests from users. Upon receipt of an email at that address, a ticketing system triggers an incident ticket generation. Then the ticket is assigned to an IT support person who processes it manually and responds to the user. This process is subject to long delays since the assignment and processing require human intervention. The manual operation is not only time-consuming but also leads to a lot of workloads to IT support personnel and produces frustrations on the users’ side. As chatbots technology gets more and more reliable, IT support is one of the areas where their use is much needed and appreciated. This research proposes to elaborate on a conceptual design of a framework for the production of chatbots for IT support. The solution operates in three steps: (1) mails in the mailbox are read, analysed and incorporated into a knowledge base (KB). (2) A machine learning algorithm (MLA) is applied to enrich it and make it ready for natural language processing. (3) The last step of the process produces the chatbot. The design and implementation of the proposed framework is a heavy project that is just starting. The scope of this paper is about the overall conceptual design of the framework. Our future work will address the details and implementations of the componentsen- Chatbotmachine learningartificial intelligenceontologynatural language processingArticle