Classic Artificial Intelligence: Tools for Autonomous Reasoning
Stephen S. Mwanje, Marton Kajo, Benedek Schultz, Kimmo Hätönen, Ilaria Malanchini · 2020
Autonomous reasoning requires that the cognitive entity perceives relationships amongst data elements and makes inferences about these elements and their relations to subsequently select the appropriate action. This chapter summarizes some of the available techniques for achieving autonomous reasoning. Essentially, it presents a toolbox of classic artificial intelligence techniques that could be useful for automating inference tasks in networks. For each tool or technique, the presentation includes an evaluation of the accorded degree of cognition based on the cognitive decision-making model. The chapter provides the basis on which the specific tool may be selected for application towards specific network challenges. It focuses on the most common methods that are also likely to have wider usage in network management: expert systems, closed-loop control systems, case-based reasoning, and fuzzy inference systems. These methods are presented in an order that highlights ever more cognitive capability beyond simple inference.