Organizations currently possess a vast and rapidly growing amount of information--much of it residing in corporate databases, generated as a by-product of transaction automation. Although this information is a potentially rich source of knowledge about production processes, few companies have begun to implement learning systems to leverage the value of this stored data and information (Bohn 1994). At the same time there is a broad and expanding array of technologies, tools, and models to assist in deriving knowledge from data. These include technological areas such as knowledge discovery in databases, machine learning, statistics, neural networks, expert systems, and case-based reasoning (Piatetsky-Shapiro and Frawley 1991). These technologies and tools possess a broad range of capabilities for inductive, deductive and analogical reasoning approaches to the creation and validation of knowledge. The literature of computer science, information systems, and statistics contains a vast number of studies comparing the most similar types of learning algorithms from a very technical perspective (Curram and Mingers 1994; Kodratoff 1988; Weiss and Kulikowski 1991) and specialists exist in each of the technology areas. However organizations faced with planning learning systems cannot normally assemble a team with expertise in each of the potentially important technology areas. They must start by identifying critical questions and learning goals, which are driven by business context and unconstrained by the capabilities of particular technologies. The premise of the research is that there is a growing need for guiding frameworks and methods to help organizations assess their learning needs--starting from the broadened perspective of business knowledge requirements --and match them with the most suitable categories of learning technologies, models, tools and specialists (Keen 1994). This report describes research that is underway to develop a classification theory to serve as the foundation for technology selection stages of such a method