Menu
July 25, 2018 | By

AI in Your Workday: A Quick Guide to Natural Language Processing

Artificial Intelligence is all over the news and many experts predict its application will greatly impact the future of work. Natural Language Processing (NLP) is one of the most commonly used types of AI today. From translator apps to voice remotes to virtual assistants, NLP is becoming part of our everyday lives. Have you ever stopped to think how these machines are able to understand us when we do not speak Python, Java, or C++? In this quick guide to NLP, we’ll unravel some of the mystery behind this technology. You’ll get a better understanding of why customer service is improving and why science seems to be getting smarter.

Let’s start with a quick definition. NLP is often divided into two components that go hand-in-hand:

  1. Natural Language Understanding (NLU)– A machine’s attempt to decode the written and verbal messages we give to it.
  2. Natural Language Generation (NLG)– A machine’s response in a “human,” non-programming language, to the message, question, or direction it was given.

How does Natural Language Processing work?

When you write a message to the program, it pulls out the nouns, verbs, adjectives, and the other parts of speech to hypothesize what it is you want. From there, it searches through a database of its own capabilities, and responds in a way that is most likely to achieve what it perceives the desired outcome you requested.

A program will go through a similar, albeit more complicated, process with verbal requests. While written messages often provide straightforward and complete context, verbal messages may become convoluted due to accents, interruptions, pauses, tone of voice, and background noise. These can all create ambiguity for the machine as it references a programmed database of sounds with corresponding words before it can translate what was heard into its native programming language.

Once the program has received the message and completed its programming for how it can achieve the instruction, it will respond back to you in a non-verbal (written) or verbal manner. Modern examples, such as those listed below, use NLP machine learning meaning the programs improve and become more accurate with each use.

Major uses of NLP Tools 

  • Transcription – One major use of NLP is the ability to transcribe spoken word into text. We’ve rolled this technology out in our Smart Meeting Assistant, so users can get a transcribed copy of all their recorded meetings. This becomes incredibly helpful when someone wants to refer back to a specific part of a meeting or presentation without needing to watch the whole recording. It also reduces the amount of note-taking within a meeting, allowing for more active participation.
  • Chatbots  This is probably the most widely used application of NLP today. (Did you know that Facebook Messenger surpassed over 300,000 active chatbots this year alone?) While often used for simple tasks such as texting your bank to find out your account balance or asking your favorite news organization on Facebook for today’s top headlines, chatbots today can perform more advanced functions from scheduling physician appointments to booking flights. A 2018 Gartner study predicts that 15% of all customer service interactions by 2021 will be completely handled by AI; a 400% increase from 2017.
  • Virtual Assistants – Whether you are on team Alexa, team Google, or team Siri, there is no arguing that virtual assistants have been on a dramatic uprise in the last couple of years. Although technically chatbots, these assistants do not function as experts of one specific domain. Instead, they have the capacity to complete a variety of tasks to a certain degree. When the assistant gets stumped, it does what any human would do and performs a Google search. Pro Tip for GoToMeeting users: Download the GoToMeeting Alexa Skill for a hands-free way to schedule and review your GoToMeetings.
  • Summarization One of NLP’s greatest assets is its ability to read through a large amount of text and generate a concise and comprehensible summary. While using a program to summarize a novel you forgot to read for book club may provoke some frowns, this function is incredibly beneficial in medical and other scientific fields where there are constant discoveries and endless journal articles that no human could ever read in a lifetime. In a perhaps more fascinating example, The Washington Post, The Associated Press, and Yahoo! have all used AI and NLP applications to write articles.
  • Translation Your high school Spanish teacher may have told you that she could easily spot an essay that you put through an online translator. Even though most students still cannot get away with having Google Translate write their essays, NLU advancements now exhibit more contextual understanding to make translations more comparable to native speakers. Pretty soon Spanish teachers will know their students used a translator bot because the essay is too good.

As we move into the future of work and the collaboration market, we expect to see new uses for NLP appear across a number of industries and roles. New technologies and AI will continue to fulfill the simple and mundane tasks for people, giving them the time back in their day to do what they do best – brainstorming, collaborating, and innovating.

For our GoToMeeting users who want to enable transcription through Smart Meeting Assistant, read this blog post to get started! 


Subscribe to our blog

Submit your email address to get updates right to your inbox!