Abstract: User responses to prompts or queries are classified or scored using explainable artificial intelligence (XAI) according to levels of a defined proficiency scale for selected subject matter. Recommendations for improving user proficiency are generated using XAI based on the proficiency classifications or scores and underlying rationales leading to those classifications or scores. XAI can be employed for identifying user responses that are consistent with predetermined cheating patterns. Prompts and queries can be presented, user responses classified or scored, and recommendation generated in a single session, i.e., in near real time. Disclosed methods can be advantageously employed for assessing user proficiency, and providing recommendations for improving proficiency, in a wide variety of subject areas, including spoken, written, or signed human languages.