Learning Machines: In Search of a Concept Oriented Language
This work addresses foundational questions about machine intelligence and knowledge discovery, but it appears incremental as it builds on existing ideas without presenting new empirical results.
The paper tackles the challenge of defining and enabling next-generation intelligent machines by proposing a general framework for a concept-oriented language, based on historical contributions and analogies to human intelligence.
What is the next step after the data/digital revolution? What do we need the most to reach this aim? How machines can memorize, learn or discover? What should they be able to do to be qualified as "intelligent"? These questions relate to the next generation "intelligent" machines. Probably, these machines should be able to handle knowledge discovery, decision-making and concepts. In this paper, we will take into account some historical contributions and discuss these different questions through an analogy to human intelligence. Also, a general framework for a concept oriented language will be proposed.