A Perspective on Natural Language Understanding Capability: An Interview with Sam Bowman

Prof. Bowman: I can only give a very high-level overview of the earlier part of this history, but there are three broad families of approaches to language understanding in natural language processing (NLP). From the 1960s through right around 1990, the predominant mode of research was rule and template based and very strictly symbolic. This kind of work on language understanding involved quite a lot of knowledge engineering, of trying to build systems around expert-designed representations for the semantics of various kinds of situations and the meanings of words. The CYC project seems like a good example of the kind of work that was at the center of a lot of this type of thinking. This was an attempt to build a symbolic knowledge base of sort of all of human commonsense knowledge, such that when you hear a new sentence and you want to build some sort of representation of it, you can reference it against a large knowledge base. The word ‘‘dog’’ (i.e., the symbol ‘‘dog’’) has attached to it all of this information about what dogs are and what they do. It allows you to ground your meaning of symbols around such representations.