Answer the question
In order to leave comments, you need to log in
How to train a neural network to recognize formal logical errors in human speech when writing?
The question for me is both interesting and important, however, to my shame, it is worth noting that ANN is not my subject area, so I am interested in nothing less than an assessment of the realities of such a project today (and in the future), but also problems associated with it - for example, huge computing resources that may be needed in the process of forming such complex patterns, etc.
Briefly and in general terms, the "task" can be described as follows: to implement a self-learning critical analysis of text content based on well-known logical errors and cognitive distortions. Something like this.
Answer the question
In order to leave comments, you need to log in
No way. In this formulation, this is a task from the series of how to teach a neural network to think. Sometimes not every person is able to understand human logic in the text, with his computing resources, there can be no talk of a computer.
entity-property-relationship => noun-adjective (participle and forms)-verb
........start thinking.............
Didn't find what you were looking for?
Ask your questionAsk a Question
731 491 924 answers to any question