James,
In re: Rousseau’s Systemic Semantics (Systemic Semantics: A Systems Approach to Building Ontologies and Concept Maps) in developing an ontology for systems engineering and science, there has been a historical problem inherent in over-reliance on the “systems thinking” approach. This problem, identified by Rousseau’s “confusion” characterization relates to vast differences between an individual “authority” as an expert and the dispersion of concepts in a groups > 7 (i.e. the INCOSE Fellows attempts at the definition of a system). We found similar dispersion in working with a BFO implementation WRT experts.
One promising solution may be to apply OpenAI Five (or something similar, such as ES-HyperNEAT) in NLP to relate and refine concepts in linguistic terms (the conceptual encoding is not directly language driven, but offers indirect representations to languages). The general strategy is to use competitive co-evolution between the machine’s knowledge base (encoded as a meta-neural network that generates other neural networks in various contexts) and the diverse human knowledge bases. The result can be empirically compared with alternative strategies. Furthermore, a meta-ontology may be developed. This strategy is well known and (for AI and NLP at least) reasonably mature. For example, see https://blog.openai.com/openai-five/ in relation to the game Dota.
K A Lloyd
--
The SysSciWG wiki is at https://sites.google.com/site/syssciwg/.
Contributions to the discussion are licensed by authors under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
---
You received this message because you are subscribed to the Google Groups "Sys Sci Discussion List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+u...@googlegroups.com.
Visit this group at https://groups.google.com/group/syssciwg.
For more options, visit https://groups.google.com/d/optout.