How can machines learn to understand human language? Natural language processing (NLP) expert Maria Niessen from the Volkswagen Data:Lab can explain. Part 3 of a series about IT jobs at Volkswagen.
“Hello. If you have a question about apples, say ‘apples’,” the taped voice on the telephone hotline explains. You dutifully answer “apples”. The machine replies, cheerfully: “You said ‘pears’,” and redirects you to the nightmare of being on perma hold. This kind of thing used to happen, but is now history. That’s because nowadays machines are getting better and better at “understanding” what people want from them. Whether it be Apple’s Siri, Amazon’s Echo or Google Home, voice control operation of computers, programs, or machines is on the increase. Intelligent systems that understand human language are increasingly important for innovative operating concepts in Volkswagen’s vehicles and company processes.
Translating between humans and computers
“This is very difficult for machines to do,” says Maria Niessen from the Volkswagen Data:Lab in Munich. The Data:Lab is the AI competence center in the Volkswagen Group. Niessen works there as a natural language processing (NLP) expert and researches how computer programs process human language. “I translate between humans and computers,” says the 37-year-old and smiles. She is used to expressing complex issues in a nutshell. Niessen originally comes from Groningen in the Netherlands, where she studied cognitive science and completed a PhD on the subject of noise recognition. For two years now, she has been working at the Volkswagen Data:Lab in a team whose focus is text comprehension in programming.
Easy for humans, difficult for machines
But, as is so often the case, things that people can do very easily and intuitively – at least in their native language – are difficult for machines. And vice versa: computers solve differential equations in an instant, while humans find them much harder. Why is natural language a problem for machines, apart from grammar and vocabulary?
Volkswagen at CEBIT 2018
Volkswagen is a digital company that drives modern information technology forward. In the run-up to CEBIT in Hanover, we are presenting a series of portraits of people in the Group with exciting IT jobs. At CEBIT (June 11-15), the Volkswagen Group will be in the Future Mobility Hall (Hall 25) offering a forum for interested parties and experts alike – with stimulating presentations and first-class exhibits as well as interesting panel discussions and talks. The range of topics and highlights is considerable and includes not only new forms of digital automotive design, quantum computing and test projects with Blockchain, but also applied artificial intelligence in the company and data-supported traffic optimization in European metropolises. There will also be a world premiere at the exhibition stand.
“The magic word is context”
“We, the human users, want the machine to solve a problem or fulfill a wish,” Niessen explains. “The magic word is context. Within it, the program tries to extrapolate the meaning of the spoken or written word.” She is currently working predominantly with texts – they are easier to analyze than the spoken word, and are not complicated by the idiosyncrasies of the speaker, for example, their dialect or speed of speech. Nonetheless, the programs have quite a lot to learn.
To this end, Niessen and her team feed programs with data, which is the basis from which they learn, and correct them where necessary. This method is known as “machine learning”. The team researches systems which are not only able to recognize and classify language but also to identify and carry out the writer’s or speaker’s wishes. “This is of interest for many areas. For example, for customer dialog systems, or for voice control systems in the vehicle, when a driver or passenger requests information or multimedia entertainment,” explains Niessen.
State of the Art
Requests for this service arrive from the entire Volkswagen Group, worldwide. “That’s what makes it so exciting,” says Niessen. “We can work with actual problems, and solve them too.” Whether it be the chatbot for internal orders, text analysis tools for IT, telephone services or user experience in the vehicle, Niessen and her team help people by helping machines to understand human language better. And this under optimal conditions: “We make state-of-the-art solutions here,” says Niessen, proudly. New departments are constantly approaching the NLP team. “All the colleagues are extremely competent and pleasant,” says Niessen. A scientific approach combined with the possibilities of a large corporation – it’s a researcher’s dream.