top of page
  • Writer's pictureJohn Ennis

Eye on AI - January 24th, 2020

Welcome to Aigora's "Eye on AI" series, where we round up exciting news at the intersection of consumer science and artificial intelligence!


This week, we focus on the use of Natural Language Processing (NLP), the subfield of AI dedicated to understanding human language, and its use in consumer-facing technology.

New NLP Adoption in Retail and e-Commerce

As previously mentioned, NLP is the area of AI dedicated to understanding the human language. It uses computational linguistics that provide parsing and semantic interpretation of text to allow systems to learn, analyze and understand human language, and is currently being used in many of our most commonly used AI devices (think Alexa, Siri, or any voice-led AI technology).

Retail and e-Commerse were among the first industries to adopt NLP on the enterprise level, according to the Emerj Artificial Intelligence Research article titled “Natural Language Processing in Retail – Current Applications”. While NLP adoption is already prevalent in those industry markets, NLP systems are now evolving to further improve business processes.

Take the chatbot, for instance, perhaps the most common form of NLP. Most people think of chatbots as being used mostly on the phone, online, or through other voice-activated devices, such as Volvo’s newly released voice-activated A/C. But there are other, less obvious ways retailers may soon begin to utilize chatbots to their advantage, such as through robots adoption. Niccolo Mejia, Emerj reporter, explains:

“Retailers could apply chatbots to in-store touchscreen interfaces and robots to provide an interactive customer service experience. This automates some of the work human employees would have to do on top of their normal responsibilities.” (video example of Pepper, the robot)

In-store virtual assistants are also high on Mejia’s list of soon-to-evolve NLP examples. By combining voice assist technology with a computer-generated avatar, for instance, companies could create a more personable, emotionally-driven shopping assistant than your typical chatbot.

“This [NLP example] could improve the customer’s relationship to a brand,” continues Mejia. “A virtual assistant of this kind would use NLP for the same functions as other chatbots, such as Apple’s Siri. However, it would also need to signal when and how to move the virtual avatar based on the customer’s responses.” (video example of Millie, the retail assistant)

Lastly, according to Mejia, NLP can be better utilized for product recommendations. Currently, companies can predict customer preferences based on purchase history through predictive analytics. When NLP is coupled with predictive analytics, relevant recommendations can be given to in-store customers as they engage with robots and virtual assistants.

IBM’s Debate AI Could Drastically Transform Voice Assistants

We continue with an article out of MIT Technology Review, titled “IBM’s debating AI just got a lot closer to being a useful tool”, which discusses a lesser known, more complex form of NLP called argument mining, which may soon supercharge voice assistants.

“A core technique used to help machines reason, known as argument mining, involves building software to analyze written documents and extract key sentences that provide evidence for or against a given claim,” writes MIT Tech Review reporter Douglas Heaven. “These can then be assembled into an argument. As well as helping us make better decisions, such tools could be used to catch fake news—undermining dodgy claims and backing up factual ones—or to filter online search results, returning relevant statements rather than whole documents.”

The ultimate goal is to build a system that can trawl through as many sources of information as possible and build an argument using every bit of evidence it can find. IBM’s Project Debater is perhaps closer than any other argument mining system in reaching the ultimate goal of this technology, which is to build a system that can trawl through infinite information sources to build an argument using all the evidence it can find. Drawing on 400 million documents taken from the LexisNexis database of newspaper and journal articles, Project Debater reviewed some 10 billion sentences, a natural-language repository 50 times larger than Wikipedia.

“The resulting neural network can handle queries on a wide variety of topics, returning sentences that are more relevant than those identified by previous systems,” continues Heaven. “…. when tested Project Debater achieved 95% accuracy for the top 50 sentences across 100 different topics... Other systems have coped with only a few dozen topics. It is also a big improvement over the live debate system Slonim showed off last year.”

IBM hopes to soon offer Project Debater as an AI cloud service, giving other companies the ability to use it to improve voice assistants, chatbots, rate debate performances, or snuff out fake news stories before they gain traction, among other things.

“Argument plays an important part in how people communicate: we list reasons for our choices, we ask for advice, we persuade and cajole,” writes Heaven. “Talking to virtual assistants that could converse at that level would feel far more natural. ‘What we are doing touches on something fundamental to our lives,’ [IBM Spokesman Christopher Sciacca] says. ‘We’re trying to tie language-understanding technologies together to help people make better decisions.’”

Other News


That's it for now. If you'd like to receive email updates from Aigora, including weekly video recaps of our blog activity, click on the button below to join our email list. Thanks for stopping by!

bottom of page