Natural language processing helps computers understand people

22 July 2021

2 min read

In the real world, business operations generate data continuously and exponentially. In the virtual world, artificial intelligence translates real-world data into actionable evidence, but nuances of language are easily lost in translation. If we can teach computers to understand syntax, can we improve information intelligence?

A colleague of mine likes to say that understanding language is easier said than done. It’s an idea I’d find easy to dismiss, if not for the fact that his catch phrase itself is evidence that the spoken and written word often is more than meets the eye – or the AI.

“We have forgotten this as adults, but we spent several years at school learning how to read and write,” François-Régis Chaumartin, founder of natural language processing (NLP) software company Proxem told me when I first met him in 2020. “Why is it so difficult? Because ambiguities are omnipresent in human language. A single word can describe many concepts, and a given concept can be described by many words.”

Human language is beautiful, but its intricacies—the slang, clichés, turns of phrase, ironies and homonyms, just to name a few—have proven difficult for computers to comprehend. Around the world, each language and dialect also has its own nuances, leaving room for error with every translation and interpretation. That’s why NLP has been an elusive achievement in data science and AI for many years.

As much as 80% of company’s knowledge is implicit and hidden behind text documents, including regulations, requirements, contracts, emails and social media. Without proper semantic analysis, weak signals are easily overlooked. Misspellings, industry-speak and internet shorthand lead to miscategorization or, worse, disregard of valuable data. Being able to turn mountains of textual data into accurate sets of related concepts and actionable insights at lightning speed is a key business advantage.

Strong semantic intelligence that combines natural language and machine learning technologies enables knowledge interpretation to be automated, transforming it from implicit to explicit. When explicit knowledge is available to everyone, closer connections between the business and its consumers, patients, partners or employees are generated, capturing and contextualizing insights from their experiences and expectations. Because NLP translates huge information flows into clear, sharp insights and trends, individual experiences become collective, reusable knowledge. These insights and trends can be used not only to improve operational excellence, but also to help identify opportunities for innovation. Innovation, in turn, increases customer satisfaction and loyalty and, thus, improves business performance.

Historically, NLP has been used for analyzing reports, responding to customer feedback and improving search results. But now we’re seeing advanced applications that enable predictive maintenance of equipment, improve operating efficiencies. For example, the ability to identify automatically similar customer claims enables early detection of quality defects in any industry.  Thanks to NLP, we can layer unlimited knowledge and know-how as part of virtual twin experiences – the accurate 3D experiences that allow designers and engineers to create and test new products and processes without breaking them – leading to endless possibilities.

We’re in the age of Big Data. But all the information we need is not quite at our fingertips. As we inch closer to teaching AI to truly understand human language, we also begin to unlock the potential for businesses to be even more intelligent, innovative and customer-centric.

Morgan ZimmermannChief Executive OfficerDassault Systèmes NETVIBES

Learn how companies are using NLP to improve business intelligence and customer experience

Being human

Natural language processing helps organizations to take a human-centric view

Elly Yates-Roberts

4 min read

Modern organizations generate vast amounts of data, in multiple forms, from human beings – customers, employees, suppliers and more. Getting a balanced view of what it all means requires more processing power than humans can manage, but involves nuances that have been beyond the understanding of computers. At last, natural language processing promises to bridge the gap.

The biggest fans and fiercest critics of an organization usually have the greatest impact on decision-making. Why? With too much feedback to process, humans tend to listen more to those who talk the most or the loudest.

Even relatively small businesses can have vast amounts of human input in multiple forms, including telephone conversations, social media, survey results, incident logs, product specifications and RFPs. Analyzing that data should help deliver a balanced view of all stakeholder feedback, allowing businesses to understand customer needs and industry trends faster, guiding new product development, improving operations, minimizing risk and accelerating research and development. But humans don’t speak binary, and computers haven’t been adept at the nuances of human speech.

Natural language processing (NLP), a branch of artificial intelligence (AI) dedicated to converting human language to binary code for search engines, translation services and voice assistants, promises to solve the dilemma. After decades of development, and with billions of words written by humans to learn from, NLP has gradually advanced to the point where it is ready for a particularly difficult task: business communications.

“When you bring together mathematics, computing and linguistics, it enables you to teach computers how to understand human languages,” said François-Régis Chaumartin, founder and vice president of data science at NLP software company Proxem, now a part of Dassault Systèmes. “Using NLP, we can create a multi-dimensional world in which each word is associated with a vector. We can then do mathematics on these vectors. For instance, if you take the vector for ‘king,’ remove ‘man’ and add ‘woman,’ you produce something that is very close to the vector for ‘queen.’ NLP makes the language computable so we can handle words as mathematical concepts.”


Fortune Business Insights reports that the global NLP market stood at US$16.53 billion (€14.04 billion) in 2021 and is expected to reach US$127.26 billion (€108.09 billion) by 2028.

As these technologies evolve, they are delivering effective human-to-machine communication and providing a greater ability to center decision-making on the needs and desires of humans.

Multinational energy provider ENGIE has more than 24 million customers worldwide. Each customer interaction ends with a questionnaire to generate feedback.

“When you bring together mathematics, computing and linguistics, it enables you to teach computers how to understand human languages.”

François-Régis Chaumartin
Founder and vice president of data science, Proxem

“Being able to analyze their comments is essential to allow us to understand the reasons why a customer is satisfied or dissatisfied and act to re-engage the customer,” said Florence Bigeard marketing project manager in customer experience at ENGIE.

By deploying a semantic analysis tool that uses artificial intelligence and NLP, ENGIE has been able to identify strong signals – a topic that everyone is talking about- but also “weak signals” that might otherwise have gone unnoticed. Weak signals seem irrelevant on the surface, but can add up to a significant pattern. “This can help feed the design loop for new products and prevent further problems,” Bigeard said.

For instance, NLP helped ENGIE quickly detect customer irritation over deployment of the company’s smart “Linky” meter, which transmits usage data and receives orders electronically. Customers did not understand why they had to pay a commissioning fee when the work is done remotely, without a technician visit to the home.

By deploying a semantic analysis tool that uses artificial intelligence and natural language processing, companies can identify strong signals – a topic that everyone is talking about – as well as ‘weak signals’ that might otherwise have gone unnoticed. (image © NETVIBES)

“It is also a matter of dealing with the dissatisfaction before it becomes too great by explaining these costs are charged by the [utility] distributor,” Bigeard said. “There are few customers who complain about paying these fees, but they are the ultra-detractors.”

Following that success, ENGIE is planning to implement new sources of feedback and analyze speech-to-text data from telephone conversations, in addition to written responses. “We also want to collect customer opinions on social networks through daily monitoring, and hope to better detect customer emotions following our interactions with them,” Bigeard said.


Another broad energy company, France-based TotalEnergies, is using NLP in multiple ways, including improvements in the relevance of search results on its website and safety analysis of equipment breakdowns at industrial sites.

“This analysis was previously done manually on a very small sample of data,” said Pierre Jallais, who is responsible for innovation at TotalEnergies. “We wanted to implement NLP to analyze all these unstructured mini reports and check if the equipment is working properly.”

NLP has enabled TotalEnergies to analyze nearly 400,000 reviews linked to breakdowns or maintenance reports. As a result, the business was able to calculate the failure rates of its equipment and develop strategies to improve them.

TotalEnergies intends to expand the use of its NLP solution by collecting all the information on breakdowns for improved visibility on maintenance costs, create action plans for quality checks, compare performance at different sites to identify best practices, and review the equipment purchasing strategy.

“Evaluating the quality and safety of our equipment on a large scale is the goal,” Jallais said.


NLP is demonstrating its value at improving customer experiences and ensuring high standards in operational performance and safety. But its potential – especially when paired with technologies like virtual twins that show data in the context of scientifically accurate 3D equipment models – could be significantly greater.

Natural language processing can create a multi-dimensional world in which each word is associated with a vector. We can then do mathematics on these vectors. (image © NETVIBES)

A big step in the industrial field would be using NLP to communicate the design requirements of physical products. An airplane’s requirements, for example, might include that it must take off within 300 meters, or that ice accumulations on a plane wing must not exceed 12 kilograms.

“You could express thousands of requirements in language that is technical but also natural,” Chaumartin said. “By filtering these requirements through NLP, the technology can start to produce mathematical formulas that are equivalent to what is expressed in natural language. In other words, NLP can help with the virtual development of physical products.”

Chaumartin said he has been "amazed" by the language capabilities of computers equipped with advanced NLP, and has even found it difficult to distinguish some NLP-generated texts from those produced by humans. But we should not to mistake this for intelligence, he said.

“It is incredible what we can do with NLP, but the machine is not intelligent,” Chaumartin said. “It has no deep understanding of what a sentence means. It is a very good imposter, but it is just pretending to understand. Machines are not yet intelligent enough to understand language like humans do. That is still something for the science fiction movies.” 

Learn more about NLP

Read one industry executive's view on using NLP to improve information intelligence.

Engineering exciting marketing experiences with interactive 3D

7 July 2021

2 min read

Something big is happening in retail, but it’s not what you may think.

Yes, we are making more purchases online. The COVID-19 pandemic accelerated e-commerce to about 20% of retail sales, but even bigger shifts are coming. Experts predict that online purchases will account for 95% of all retail by 2040.

In the digital age, however, we no longer need to own something to benefit from it. Uber has taught us that owning a car to travel reliably and affordably isn't a requirement any longer.  We "order" a car when we want to go somewhere. If we can order a car, why not order different vehicles for different purposes: a minivan or SUV for a family vacation, a sports car for a weekend drive, a compact car for a quick errand?  Voila! We can have whatever we want, whenever we want it.

For marketers, this has profound implications for selling users on personalized-mobility-experience value. In ride services like Uber, a specialized app could offer passengers control of stereo settings and temperature levels. It could offer customized routes, beverages, additional stops and other amenities during the trip. You see where this is going.

Selling experiences presents a different challenge than selling a car. Consumer interactions will skyrocket, and the content itself will change. Product sales are driven by features and functions, and consumers live with their choices for months or years. An experience sale is driven by convenience, personalization and immediacy. Content must become more rewarding, always on and cloud-based.

The people driving these sales will be service packagers; they will acquire content from the product makers and build experiences for different types of consumers in varying situations.

If coders – not marketers – are creating content, it should be easier to develop and publish, but it needs to be more compelling. As with online commerce, technology is driving the change, but this time through powerful, interactive, mobile-accessible 3D content.

The emerging GL Transmission Format (glTF™) has laid the groundwork for a 3D marketing revolution. Much like the jpeg revolutionized the way we view images online, glTF is making it possible to view rich 3D material quickly and consistently across many different applications. This is poised to revolutionize how marketers introduce new products and experiences. Thanks to glTF, it’s possible to create 3D digital marketing assets directly from the design and engineering source data. This means new products, services and experiences will move from concept to marketplace faster than ever before.

glTF makes online demos both possible and powerful, using light 3D assets that look and act like the real thing. Unlike physical items, users can turn and even “explode” them on the screen to see the inner workings. This content is employable online months before manufacturing begins. The value is immediately evident. Soon, interactive 3D models will be on every ecommerce website and viewable on every mobile phone, making life better for consumers.

Public interactions with these models also provide product-makers and experience packagers instant insights to improve their offers. With a photograph or video, marketers can’t “see” which features draw consumer’s attention – or send them to a competitor’s offer. They can’t know which demographics respond to which offers. glTF 3D models give marketers precise “digital exhaust” to hone in on exactly what consumers want.

That’s key, and it opens the door to an understanding of how a product creates value for consumers. Which in turn is a boost for sustainability. If distributors, retailers and consumers don’t like a concept, it won’t be made. Pointless products will disappear. We’ll create satisfaction with fewer resources. And that’s a good thing.

Tom Acland, Chief Executive Officer, Dassault Systèmes 3DEXCITE

Read more about how companies are already starting to create realistic 3D marketing assets with glTF

Marketing becomes interactive

glTF transforms engineering data into maneuverable, online 3D marketing assets

Lindsay James

6 min read

Firms no longer need to manufacture their product before they create assets to promote it. New, digital technologies are poised to transform engineering and design data into easy-to-share visualizations for powerful online product experiences. The result? Realistic 3D marketing assets that, thanks to a fast-developing open standard asset format called glTF, can be viewed and manipulated on any device.

When the COVID-19 pandemic forced the world’s biggest trade shows to cancel their physical events, many exhibitors reacted by putting their marketing efforts on hold.

“The industry really scrambled to figure out how to stay relevant,” said Darrin Hill, director of marketing for R&D and intelligence at Canadian automotive technology firm Magna International. “We noticed that many of our peers were not going all-in on digital at the time.”

In fact, the Center for Exhibition Industry Research reported in late March 2020 that 50% to 80% of the 2,500 business-to-business events scheduled for March 1-May 15, 2020 were canceled – a situation that has continued for 18 months. The loss of just the first 2.5 months of events cost the exhibits industry as much as $22 billion, CEIR estimated. By the third quarter of 2020, 97% of events were canceled.   

Magna, however, was determined to turn the challenge into an opportunity.

“We were still doing research and development activities and had things we really wanted to shout about,” Hill said. “For us, the Consumer Electronics Show (CES) is our most important and probably our largest public event that we do every year – and it’s where we gain a tremendous amount of ground with our target audience. When the physical show got canceled and replaced with a digital version, we decided to really push forward and find a different, better, more impactful way to communicate what we offer.”

But how could Magna communicate effectively to virtual attendees in an online environment? By leveraging advances in 3D digital technology, using data from its engineering teams to create an extensive portfolio of digital marketing assets in just eight weeks.

The results far exceeded the company’s expectations.

“We developed a 3D photo-realistic generic Magna vehicle on which we were able to showcase our full product offering,” Hill said. “We were really looking for a platform that could offer a similar impact to a physical environment and help increase our reach digitally. Along the way we realized that this was the first time we’ve been able to demonstrate the complete power of Magna in an intuitive and manageable way.”


Interactive 3D digital assets provide a differentiated experience for customers. Quite apart from traditional video assets – where the marketer dictates the story that is told – 3D digital assets offer a highly personal approach. Customers can interact with them at their own discretion, zoom in on the key product features that interest them most, experience product operation in a hands-on way and even explore the finer details via 360-degree rotation.

These benefits appealed to Hill and his team. Using the 3D digital model, they showcased Magna’s entire portfolio, from its autonomous driving offering to its electrified power train, doors that open and close by themselves, innovative lighting solutions and more.

mousse over
In Magna’s glTF presentation of its concept truck at CES 2021, visitors could explore each Magna system in detail. Here, viewers can choose each of the five sensor systems to interactively explore details of their benefits and operation. (Image courtesy of Magna)

“We were able to incorporate many more technologies than we had originally thought possible,” Hill said. “Working with a completely digital canvas allowed us to do things that would be impossible with a physical property. It allowed us to do more, to show more, and to be more of a storyteller. We were able to showcase all of our innovations in context, and in coordination with one another.”

The approach proved so beneficial that Hill doesn’t expect a complete return to the traditional way of exhibiting. “I expect we will take a hybrid approach in the future, using 3D to add a new dimension to physical events,” Hill said.


Although Magna is a trailblazer in the online 3D marketing space, it isn’t alone.

Tobias Brode, head of business unit medical engineering and biotechnology at Fraunhofer IPA, a research development organization based in Germany, reports similar benefits.

Brode and his team have created a mobile lab robot called KEVIN®, designed to automate repetitive, manual laboratory work and assist humans with work that must be done overnight or on weekends. Creating digital marketing assets from design data allowed Brode to expedite KEVIN's development and garner interest from the market – before the design prototype robot has been built.

Creating digital marketing assets from design data allows Fraunhofer IPA to search for investors for its mobile lab robot KEVIN before the prototype is built.

“We are able to create a virtual showroom for KEVIN, where we can demonstrate to key stakeholders his key functions and features, showcase potential use cases and enable 360-degree interaction – all in an incredibly realistic way,” Brode said.

The approach has saved Brode and his team a significant amount of time and money.

“Traditionally, we would have had to either ship KEVIN across the world so that customers can see him in the flesh, or bring them to our lab,” Brode said. “We don’t have to do that anymore – it’s a far more sustainable approach.”


For both Brode and Hill, the forced plunge into digital 3D marketing coincided with growing adoption of an open standard for 3D assets called the GL Transmission Format (glTF™). glTF converts heavy digital engineering files – sometimes measured in terabytes – to a light format that even mobile phones can run. The breakthrough gives marketers’ access to a much larger audience – virtually the entire world, in fact – while giving that audience the freedom to examine any aspect of the model from any angle, a significant improvement over demonstration videos.

“glTF has been designed by Khronos, an open, non-profit, member-driven consortium of over 150 industry-leading companies, to enable 3D models and scenes to be efficiently transmitted and loaded at runtime into diverse 3D engines, viewers and applications,” said Neil Trevett,

Khronos Group’s president. “glTF’s superpower is that it is carefully designed to be usable everywhere on the web and on mobile devices.”

Although most online sites cannot host glTF models yet, major online retailers, search engines and social media sites are working with Khronos to enable the capabilities on their platforms. By enabling their users to interact with virtual products in virtual scenes, their efforts promise to transform the online experience.

“The 3D Commerce working group at Khronos is working very closely with glTF to enable 3D in e-commerce at industrial scale,” Trevett said. “They have realized that in parallel to glTF’s technical advancements, they need to solve the current process friction of hundreds of companies trying to cooperate over the design, manufacture and presentation of thousands of products across multiple platforms.”

The 3D Commerce Working Group has been instrumental in inspiring a series of glTF extensions for more flexible model deployment, including metadata for integrating 3D models into enterprise-level asset management systems.

“3D Commerce has also worked on best practice asset creation guidelines to help 3D artists and tool vendors understand how to create 3D assets that can be reliably and easily deployed across multiple companies and platforms,” Trevett said. “And finally, Khronos is launching a 3D Commerce Viewer Certification Program to ensure that assets created using best practices will be correctly displayed by diverse platforms and devices.”


Thomas Rilke, managing director at the Deutsche Messe Technology Academy in Germany, believes that the glTF standard – along with advances in technologies that help companies create 3D digital assets quickly and effectively – will lay the groundwork for the future of industry trade shows.

To this end, Rilke has created a new business it calls Media Factory, which enhances the value of the company’s trade shows, including the world-famous Hannover Messe, by offering Deutsche Messe customers access to the tools they need to create effective digital marketing assets.

“Add the glTF standard to the mix, and . . . digital assets can be accessed from even the most basic mobile phone, instead of a high-performance computer. This is truly game-changing.”

Thomas Rilke, Managing Director, Deutsche Messe Technology Academy

“Our future is undoubtedly hybrid events, where we fuse the physical with the digital,” Rilke said. “Through our Media Factory, we will be able to ensure that more of our customers are equipped to succeed in this new reality.”

The benefits are compelling.

“It’s important to recognize that 90% of our events are for really technical industries,” Rilke said. “Our typical exhibitors are machinery companies, or manufacturers of complicated pieces for machinery. They seldom build single products – they may have 30 or 40 different models, and many of these models can be customized, resulting in an almost infinite number of variations. So, you see, it’s impossible to showcase their entire offering at a physical event, which may mean they miss out on potential business.”

By having engineering-quality, 3D digital assets and simulations to exhibit at physical events, Rilke’s customers can present their entire product offering in a realistic way.

“Add the glTF standard to the mix, and the benefits are even greater; it means these digital assets can be accessed from even the most basic mobile phone, instead of a high-performance computer,” he said. “This is truly game-changing.”

Magna’s Hill shares Rilke’s excitement about what can be achieved, both now and in the future. “Because we touch so many parts of a vehicle, it is very difficult to quickly create a unique experience for our diverse audiences,” Hill said.

That challenge is now just a memory.

“With our new digital marketing assets, everything fits together properly,” Hill said. “It belongs. There’s a logic flow which we can supplement with stories to explain our technologies in a way we’ve never been able to do before. It’s compelling, it’s impactful, and it’s really changing the game for us.”

Experience glTF-enabled interactivity at the “damaged helmet” demo on Khronos’ glTG page

Learn more about how to create gITF assets

Read one industry executive's view on the potential impact of gIFT on the future of marketing

Editor’s Note: The Fraunhofer IPA spin-off project, KUTOA, is supported by the 3DEXPERIENCE Lab through its incubator partner Industrial Future Hub, operated by Deutsche Messe Technology Academy.

Stay up to date

Receive monthly updates on content you won’t want to miss


Register here to receive a monthly update on our newest content.