Future of Work Insights

Data insights: Conversing
with machines

5 minute read time

Glenn Exton, head of data and analytics at RBS International, discusses the rapid advances in natural language processing (NLP) technology and the benefits this could bring to businesses.

It’s a common scenario: you overhear a familiar song, but you’re frustrated when the title and artist’s name escape you. The good news is that a plethora of digital assistance – from Alexa, to Siri, to Shazam – exists to provide you with the information you’re seeking, helping you to find the song and save it in your music library for the future.

You might not be aware of it but, in this scenario, you have unconsciously benefited from natural language processing (NLP).

What is NPL?

NLP is a branch of artificial intelligence (AI) that helps computers understand, interpret and manipulate human language. NLP draws from many disciplines, including computer science and computational linguistics, in its pursuit to fill the gap between human communication and computer understanding.

NLP isn’t new. However, the technology is rapidly advancing thanks to an increased interest in human-to-machine communications, plus an availability of big data, powerful computing and enhanced algorithms. This means it is now possible for computers to read text, hear speech, interpret it, measure sentiment and determine which parts are important. In general, NLP tasks break down language into shorter, elemental pieces to try to understand relationships between the pieces and explore how they work together to create meaning. But how is this technology helping us?

Faster, accurate and effective typing

Any time you type while composing a message or query, NLP helps you to type faster. In a mobile digital world where we are often communicating on the move, many people enjoy the automatic benefits of autocomplete and predictive typing. NLP identifies the closest possible term to your misspelling and automatically changes it to an accurate one by analysing the troves of data available and continually learning with your every usage to increase predictive accuracy.

Customer behaviours, attitudes and motivations continue to change rapidly so, as the volume of unstructured information continues to grow exponentially, we will benefit from computers’ tireless ability to help us make sense of it all

Better search results

Simply put, NLP ensures that, however convoluted the phrasing of your search is, it will return results. Search engines such as Google and Bing use a variety of algorithms and NLP embedded within the search engine to help you find things faster.

If you misspelled something or used a less common name, autocorrect finds the right search keywords for you. Duplicate detection collates and filters content republished on multiple sites to display a variety of search results. Spam detection removes pages that match search keywords but do not actually provide the search answers – all of which allows you to minimise time wasted and helps to provide intuitive interactions.

These are some of the basic uses for NLP; now let’s see how the technology can help with ways of working.

Using NLP to increase efficiency

NLP can be used to support any number of business compliance processes. Tagging unstructured data facilitates search across thousands of digital documents and allows compliance officers to, for example, swiftly determine whether regulations have been followed. Or, it could help them to gather customer data so they can prove it has been deleted when a customer asks for it to be purged in line with the General Data Protection Regulation (GDPR). In these ways, it helps to minimise potential future risks.

NLP can also help customers check their account balance through chatbots and voice assistants instead of logging into individual accounts, letting the technology do the work for them.

Organisations can analyse data from emails, surveys, social media and call centre conversations to identify the root cause of customer dissatisfaction. For example, at NatWest Group we use text analytics, an NLP technique, to extract important trends from customer feedback. NLP is also a core component of our chatbot Cora, and allows us to heavily automate customer service in a highly scalable way. Cora is now handling over 400,000 conversations a month; NLP has given customers a personal banking assistant who is always at the ready. In our ever-changing environment, and particularly amid the disruption arising from the Covid-19 pandemic, this has proven especially valuable.

Advances in NLP have enabled the extraction of semantics, which seek to identify context in natural language. Each type of future communication – whether it’s a status, review or tweet – contains potentially relevant and contextual information, helping us to identify and extract subjective meaning from the vast amount of information in front of us. Machines learn by firstly studying the meaning of individual words and then the words in combination, clustering them together to help automatically capture the true meaning of unstructured text and speech.

Supporting purpose

NLP is used in and out of our working lives, helping organisations automate and streamline processes, improve customer service and reduce operational costs.

Customer behaviours, attitudes and motivations continue to change rapidly so, as the volume of unstructured information continues to grow exponentially, we will benefit from computers’ tireless ability to help us make sense of it all. Each interaction with NLP-enabled processes helps us gain insight on where and how to improve ways of working with the customers we serve. And that can only be a good thing.

By Glenn Exton