Natural Language Processing (NLP) is a field of computer science that aims at having a computer parse, interpret, generate, and derive meaning from our natural spoken and written language. From simple beginnings in the early 1950s, with a paper summarizing the work taken by its’ authors to mechanize translation1; to the modern utilizations found in Apple’s Siri, Amazon’s Alexa, and IBM’s Watson the field has accomplished major advances.
In its’ current state NLP has far reaching effects across industries such as commerce, news, and advertising. They enable automatic summarization, concierge like service following the trend of conversational commerce, and deeper analysis of customers and users through opinion mining and sentiment analysis.
Not so recently, a buzzword, conversational commerce, was spawned by a blog entry from Chris Messina. He defines it, saying:
“Conversational commerce is about delivering convenience, personalization, and decision support while people are on the go, with only partial attention to spare.”
In short, conversational commerce is the trend of interacting with our devices through a very familiar interface: our voices. Powered by NLP the underlying interfaces, services, and bots that allow us to use the machine with our own language, are proliferating. As they do we have the opportunity to build applications that connect with the users in a way that no keyboard can mimic. This presents new spaces to innovate in; look at your industry for ways to incorporate language as an interface.
Beyond The Device
While changing the way we interact with our devices is a major impact from the field of NLP, it is far from the only. On the other side of the screen it’s enabling powerful tools and analyses such as: summarization (http://smmry.com) of written articles, news sources, and essays; and sentiment analysis which allows businesses, marketers, and hobbyists to explore the general opinion of a large group of users by way of statistics.
This field offers advances and trends that will bring more power to your applications, and more flexibility to your interfaces. I encourage you to dig deeper, familiarize yourself with the field and some of the tooling to have come from it. See how you can apply a computational understanding of natural language to better your domain.
As we are just beginning to see the possibilities, the field will continue to grow and change in the coming months and years. Follow these resources below as a starting place in learning more and keeping up with the potential with Natural Language Processing.
- Does sentiment analysis work? A tidy analysis of Yelp reviews http://varianceexplained.org/r/yelp-sentiment/
- Stanford CoreNLP – a suite of core NLP tools
- Natural Language Toolkit (Python)
- Apache OpenNLP
- Wikipedia – Natural Language Processing
1- Richens, R. H., & Booth, A. D. (n.d.). SOME METHODS OF MECHANIZED TRANSLATION