Ever wonder how a chatbot learns to communicate with humans? For us humans, what may be a simple case of reading the words in a piece of text and understanding their context to form a response is not such a straightforward task for a computer. Machines need training to make sense of human language and, even then, they have their limits.
Why? As Morgan Zimmermann, CEO of Dassault Systèmes NETVIBES, explained in his Compass column, many nuances of language are easily lost in translation. “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,” he wrote.
When you consider that 80% of overall information is unstructured and locked in text, it’s easy to see how businesses have previously struggled to tap into valuable data. Many companies are developing semantic processing software to help. But Dassault Systèmes offers a solution that is even smarter. Our natural language processing (NLP) technology understands human language at a deeper level to unlock the reams of data previously hidden in unstructured text.
It all came about through the acquisition of Proxem, a France-based specialist in AI-powered semantic processing software. Proxem’s solution, Proxem Studio, is now integrated in the 3DEXPERIENCE platform and works alongside NETVIBES EXALEAD information intelligence applications. Together, they deliver a combination of rule-based natural language understanding, natural language processing, and machine learning technologies to see and understand the bigger picture.
,” industry experts share how this technology is making a difference to the way companies understand their customers’ needs and identify feedback that may previously have gone unnoticed. This is helping them to make meaningful changes that improve customer service. Energy provider Engie is one company taking advantage of the power of NETVIBES EXALEAD and Proxem to implement actions that improve customer satisfaction and reduce attrition. It uses the software to analyze written customer feedback and speech-to-text data from telephone conversations to pick up on major and minor customer irritants.
Another energy company, France-based TotalEnergies, deployed NETVIBES EXALEAD to improve the reliability of its equipment and improve search results on its website. The NLP software enables TotalEnergies to analyze hundreds of thousands of reviews linked to breakdowns or maintenance reports so it can determine common equipment faults and develop strategies to prevent them from reoccurring.
What’s next for NLP?
And this is just the beginning. François-Régis Chaumartin, founder and vice president of data science at Proxem, has a vision where companies use NLP to shape future product development. “You could express thousands of requirements in language that is technical but also natural,” Chaumartin told Compass.
Soon, NLP could be used in all manner of exciting ways to help companies perform better. Imagine being able to optimize your manufacturing operations, using NLP software to comb through thousands of daily reports and spot patterns such as bottlenecks and recurring accidents. Imagine being able to combat subconscious bias in the recruiting process by eliminating gendered, age or race-related language from job descriptions. And imagine the hours that could be saved in the legal system if computers could really comprehend legalese.
Read the full article here for more insights on how companies like Engie and TotalEnergies use NLP to truly understand and respond more meaningfully to what their customers are saying. You can also hear from the CEO of Dassault Systèmes NETVIBES on the true potential of NLP.