Artificial intelligence (AI) vs. natural language processing (NLP): What are the differences?
How often do you now casually converse with a black (or white) box of some sort? Natural language processing (NLP) has become an integral part of our daily lives: Whether we’re asking our smartphone for directions or engaging with Alexa or Google, NLP and its sub-categories are hard at work behind the scenes, translating our voice or text input and (hopefully) providing an appropriate voice or text output.
But what is NLP, and how does it relate to artificial intelligence (AI) in general? The distinctions are important, as NLP has just as much, if not more, value in the enterprise as it has in our personal lives.
Let’s start with AI, the broader category under which NLP and a number of other flavors of machine-based intelligence reside. “AI is the use of intricate logic or advanced analytical methods to perform simple tasks at greater scale in ways that mean we can do more at large scale with the workers we have, allowing them to focus on what humans are best at, like handling complex exceptions or demonstrating sympathy,” says Whit Andrews, vice president and distinguished analyst with Gartner.
AI is essentially some computerized simulation of human intelligence, says Zachary Jarvinen, head of technology strategy, AI and analytics at OpenText, that can be programmed to make decisions, carry out specific tasks, and learn from the results.
With AI, computers can learn to accomplish a task without ever being explicitly programmed to do so, says Timothy Havens, the William and Gloria Jackson Associate Professor of Computer Systems in the College of Computing at Michigan Technological University and director of the Institute of Computing and Cybersystems.
For those who prefer analogies, Havens likens the way AI works to learning to ride a bike: “You don’t tell a child to move their left foot in a circle on the left pedal in the forward direction while moving your right foot in a circle… You give them a push and tell them to keep the bike upright and pointed forward: the overall objective. They fall a few times, honing their skills each time they fail. That’s AI in a nutshell.”
AI vs. NLP, explained
When you take AI and focus it on human linguistics, you get NLP.
Like machine learning or deep learning, NLP is a subset of AI. But when exactly does AI become NLP? SAS offers a clear and basic explanation of the term: “Natural language processing makes it possible for humans to talk to machines.” It’s the branch of AI that enables computers to understand, interpret, and manipulate human language.
NLP itself has a number of subsets, including natural language understanding (NLU), which refers to machine reading comprehension, and natural language generation (NLG), which can transform data into human words. But, says Wayne Butterfield, director of cognitive automation and innovation at ISG, “the premise is the same: Understand language and sew something on the back of that understanding.”
Natural language processing makes it possible for computers to extract keywords and phrases, understand the intent of language, translate that to another language, or generate a response.
NLP has its roots in linguistics, where it emerged to enable computers to literally process natural language, explains Anil Vijayan, vice president at Everest Group. “Over the course of time, it evolved from rule-based to machine-learning infused approaches, thus overlapping with AI.”
In addition to techniques derived from the field of computational linguistics, NLP (which may also be referred to as speech recognition in some contexts) might employ both machine learning and deep learning methodologies in order to effectively ingest and process unstructured speech and text datasets, says JP Baritugo, director at business transformation and outsourcing consultancy Pace Harmon.
What does NLP look like in action? Let’s look at some of the problems it can solve:
NLP use cases
While NLP may get the most attention in consumer applications today, it has significant implications for organizational IT. “Understanding language and communication in general is huge for the enterprise as we spend most of our day communicating in one form or another,” Butterfield says.
Any area of the business where natural language is involved may be fodder for the deployment of NLP capabilities, says Vijayan. Think chatbots, social media feeds, emails, or complex documentation like contracts or claims forms.
“NLP is typically deployed to categorize content, extract content, analyze sentiment, summarize documents, translate languages, deploy voice-driven or chat-driven interfaces, etc.,” Baritugo adds.
The discipline of NLP may be broken down into any number of more discrete NLP tasks, based on the enterprise’s need. Some may involve serious AI, while others are more rules-based.
Those tasks then combine to create NLP capabilities like content categorization, contextual extraction, sentiment analysis, document summarization, or speech-to-text and text-to-speech conversion, as examples.
The challenges of enterprise NLP
NLP applications come with the same risks of failure as any other AI deployments, says Vijayan: Most notably, they can suffer from inflated expectations, unclear business cases, and lack of training data. Additionally, NLP opportunities may require entirely different training sets depending on the language being processed and the context, Vijayan says. You may need one set of training data when creating an NLP-enabled solution for processing contracts and entirely different data when coming up with a solution to answer payroll queries, for example.
Butterfield points out some of the hurdles NLP must overcome: Interpreting the actual meaning of voice or text correctly, dealing with sarcasm, understanding local dialects, parsing multiple potential intents, and generating bespoke responses, to name just a few.
Can’t picture where NLP will fit into your organization’s work? Consider this: “NLP is everywhere, and [is] much farther-reaching than the more recently developed smart assistants,” Keiland Cooper, director at ContinualAI and a neuroscience researcher at the University of California, Irvine, recently told us. “Everything from search, email spam filtering, online translation, grammar- and spell-checking, and many more applications [use NLP]. Any machine learning that is done involving natural language will involve some form of NLP.”