Natural language generation

Let's learn about natural language generation (NLG):

  • NLG is considered the second component of NLP.
  • NLG is defined as the process of generating NL by a machine as output.
  • The output of the machine should be in a logical manner, meaning, whatever NL is generated by the machine should be logical.
  • In order to generate logical output, many NLG systems use basic facts or knowledge-based representation.
  • Let's take an example. You have a system that writes an essay on a particular topic. If I am instructing my machine to generate 100 words on the topic of The Cows, and my machine generates 100 words on the topic of cows, then the output (here, 100 words about cows) generated by the machine should be in form of valid sentences, all sentences should be logically correct, and the context should also make sense.