Natural Language Understanding Wiki

Knowledge certainly play an important role in human language comprehension. Hence, it is presumably an very important ingredient for NLU.

TODO: Kendeou and O'Brien, Noordman et al.


Based on representation[]

Tacit vs. explicit knowledge[]

Tacit knowledge is the kind of knowledge that is difficult to transfer to another person by means of writing it down or verbalizing it. It is opposed to explicit knowledge which can be formalized easily.

Common-sense vs. common knowledge[]

According to Cambria and White, 2014[1]:

"common-sense knowledge (which humans normally acquire during the formative years of their lives) and common knowledge (which people continue to accrue in their everyday life) in a re-usable knowledge base for machines. Common knowledge includes general knowledge about the world, e.g., a chair is a type of furniture, while common-sense knowledge comprises obvious or widely accepted things that people normally know about the world but which are usually left unstated in discourse, e.g., that things fall downwards (and not upwards) and people smile when they are happy. The difference between common and common-sense knowledge can be expressed as the difference between knowing the name of an object and understanding the same object’s purpose. For example, you can know the name of all the different kinds or brands of ‘pipe’, but not its purpose nor the method of usage. In other words, a ‘pipe’ is not a pipe unless it can be used (Magritte, 1929)."

The distinction is made along the verbalizability axis as with tacit and explicit knowledge. However the meaning of common-sense and common knowledge is restricted to knowledge of somewhat general nature (as opposed to domain-specific knowledge).

Based on acquisition[]

"Knowledge by acquaintance" and "knowledge by description"[]

Two fundamentally different types of knowledge according to Russell.

Procedual knowledge[]

Procedual knowledge or imperative knowledge is the knowledge formed while doing some task. It is very task- and agent-specific.

Based on reference[]


In short, metaknowledge is knowledge about knowledge.


About the information-processing agent itself.

Bloom’s Taxonomy[]

  • Factual Knowledge
  • Conceptual Knowledge
  • Procedural Knowledge
  • Meta‐Cognitive Knowledge

What every language user knows[]

This list is taken from Winograd (1983; p. 4-6)[2]:

  • Word order rules
  • Vocabulary and word structure
  • Semantic features
  • Reference
  • Time
  • Discourse structure
  • Attitude message
  • Prosodic conventions
  • Style conventions
  • World knowledge


  1. Cambria, Erik, and Bebo White. "Jumping NLP curves: A review of natural language processing research." IEEE Computational Intelligence Magazine 9.2 (2014): 48-57.
  2. Terry Winograd. 1983. Language as a cognitive process. Volumn I - Syntax.