Yanping grew up in China, and moved to Canada to complete her PhD at the University of Sherbrooke in Machine Learning and Data Mining. While completing her studies, she had strong a strong interest in further researching Machine Learning, which led her to pursue becoming a professor in Natural Language Understanding (NLU) at Xiamen University. In 2015, she decided to move to the AI hub, also known as Montreal, to dive into the industry dimension of her specialization. With NLU being one of the key technologies of our platform, Kepler, we got down to the nitty gritty with Yanping, who gave us a behind-the-scenes look at what her life as an NLU research scientist is actually like. — Can you explain what Natural Language Understanding is exactly? NLU allows computers to understand the structure and meaning of human language. The human interprets and communicates the intent of a sentence for the computer to read, decipher and understand in a valuable manner. When the text has been provided, AI uses algorithms to depict different classifications, such as locations, timing, and sentiments to collect all essential data. For example, a sentence structure discussing the road trip to Maine for the July 4th fireworks would be interpreted as: “The road trip [intent] to Maine [location] for the July 4th [date] fireworks was a great family activity [sentiment].” What’s a “research scientist”, anyway? How do you spend your time? As a research scientist, my days are spent researching and learning the latest published NLU papers and codes to better understand how others implement their ideas. My work is based on the different clients we have in our roster, so the techniques may differ depending on the project at that time. The cycle of work is typically: research, implement, experiment. Can you give us an example of NLU in one of your recent projects? A great example of NLU is topic classification where you are given a piece of text and you have to classify it depending on what topic it falls into. Depending on the given project, text NLU is not only limited to English, on one of my most recent projects, I worked on auto-tagging online reviews in Portuguese (no, I am not fluent in Portuguese) but it does keep things exciting. “The most interesting part of NLU is trying different ideas based on the latest machine learning techniques (not limited to NLU) to reduce human efforts on annotations and to improve performances of multiple NLU tasks.” What are your go-to research resources? Two of my favourite sources that I look at daily are GitHub and Google open source. GitHub is used specifically for open-source coding; it’s a great resource for individuals who want to learn how to run and build ideas from other open source projects. Google is more of an inspiration point, I like to see what new ideas they propose. What do you love most about Stradigi AI? There are so many little things that are put in place that make the culture so unique here. From having a kids holiday party to receiving your own cake on your birthday, small gestures really do go a long way. I would also have to say my favourite part of my job are my leaders, Jaime and Carolina. They drive us to be successful with their passion, and by instilling an open and supportive work environment. Want to join our diverse team of AI experts? Apply here today.