Natural Language Processing (NLP)

Overview

NLP is a field of artificial intelligence and linguistics concerned with the interactions between computers and human (natural) languages. The goal of NLP is to effectively bridge natural language to a form of representation interpretable by intelligent software. Although this bridge can go both ways, it is significantly more difficult to go from what is an essentially grammatically-structured collection of terms to a semantically-rich representation (i.e., an ontology or other type of semantic model). Not only must terms and phrases be processed into the concepts that they represent but this must be achieved within a naturally ambiguous paradigm. As human beings, we have no problem understanding the phrase "Time flies like an arrow." Yet this simple example of natural language poses a significant challenge for software, which lacks the extensive knowledge of the world and linguistics that we humans so readily take for granted.

Considering the amount of potentially valuable content existing in natural language form (i.e., media, forums, blogs, and so on), the significance of equipping a decision-making environment with an effective NLP capability becomes clear. Once intelligent agents can readily interpret content sourced from various media, their ability to identify and expand their knowledge of relevant subject matters is dramatically improved.

Research Area(s)

• Model-based approach to NLP, where the NLP engine is equipped with a specific model describing the domains of the expected content.

• Discovery-based approach to NLP, in which the NLP capability attempts to process natural language without any preconceived notion of the content's domain.