The latest advancements in hardware, software, and cyberware technologies have made it possible to develop intelligent systems that use sensors, detectors, mining, reasoning, inference, exploration, heuristics, etc. to generate and aggregate knowledge that had not been designed and embedded by the designers and implementers of these systems. What is the (science philosophical) status of this run-time obtained, (maybe system-independent and decontextualized, therefore partly or fully shareable) formal knowledge (not big data!)? This is in the background of the principal question: If epistemology is the theory of scientific knowledge, and gnoseology is the theory of human general knowledge, what is the science philosophical theory that investigates, describes, categorizes, justifies and validates the knowledge possessed, obtained, explored, mined, inferred, and produced and exploited in any way by intelligent systems (e.g. social-cyber-physical systems)? Actually, what is the current state of investigating the ontological, "epistemology-like", methodological and/or axiological characteristics of this knowledge?
Dear Dr. Wong, thank you for your answer. Over a period of time, due to technological possibilities and limitations, the knowledge that knowledge-intensive systems possessed and processed was inputted mainly by the developers of these systems and/ or extended by simple formal mechanisms such as rule-based inference engines in the case of some expert systems. However, nowadays, thanks to the much more sophisticated software and knowledge engineering technologies and methodologies, but also, for instance, to sensor and detector technologies, and mining, analytics, and abstraction algorithms, systems are also able to extend their own knowledge and share it with other systems. As far as I know, the philosophical status of this knowledge is yet not addressed in the literature. In other words, system synthesized knowledge does not have its 'epistemology' or 'gnoseology'. I wonder if there has any step been made in this direction, that is towards a theory of systems knowledge, which would provide a rigorous explanation and disposition of this knowledge, as epistemology does it for scientific knowledge. Thank you for your consideration of my question. Regards, Imre Horvath
Dear Dr. Wong, though there are references to certain reasoning and learning mechanisms in my previous clarification, I am more interested in the results of the operation of these mechanisms, that is, in the knowledge synthesized and possessed by knowledge intensive systems, no matter what category of systems they belong to. For example, in the case of an incremental decision tree algorithm, my concern is the knowledge that this algorithm outputs, rather than the operation or any other characteristics of the algorithm. A relevant question is, for instance: How the chunks of knowledge generated by this kind of (multiple different) algorithms come together and form a body of knowledge. I am interested in these type of questions since the relational knowledge derived from big data, the synthetic knowledge produced by self-learning systems, and the tacit knowledge shared in social places are gradually merging and creating (a need for) a new 'epistemology', which is probably radically different than the traditional science philosophical epistemology. Does this new 'epistemology' or 'gnoseology' or 'whatever' exist already? How is it called? What are its principles, aspects of investigation, and how does it relate to the development of complex knowledge-intensive systems? Kind regards, Imre Horvath
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Artificial Intelligence