The media update
team answers three quick questions about NLU and why it’s such a vital element of AI technologies in use today.
You’ve heard all about natural language processing
(NLP): a form of AI that helps computers identify the context and grammatical use of words in sentences. Through NLP, machines can accurately process natural language to translate languages, find important information in documents, analyse text and answer human questions.
But which part of an NLP system allows machines to grasp the meaning of a question or ‘understand’ what a sentence means? That would be NLU.
What is natural language understanding?
For a machine to find meaning in text, it needs a subset of NLP called NLU.
NLP can break up a sentence into various word classes and recognise the various parts of speech that make up a sentence. But without NLU, it would not be able to extract meaning from the text or infer what the author’s intent is.
NLU also allows computer systems to determine how one sentence relates to other sentences in a paragraph or document, and the emotion they carry.
Wait, isn’t it the same as natural language processing?
Other components of NLP can identify parts of speech, categorise text, translate certain languages and recognise the entities – people or places for instance – within documents.
But, NLU gives machines the ability to analyse these texts for meaning and emotion. It breaks down the complex human language in documents into commands or queries that a system can act on.
The technology helps computers process language with grammatical or spelling errors, mispronunciations and colloquial wording into a structure it can use.
What is NLU used for?
It’s a vital element of AI systems, especially those that rely on human speech and casual language to work. Chatbots and voice assistants, for instance, have to know what your question means in order to act on it.
NLU is also being used to perform sophisticated analysis of text. In the media intelligence industry, companies rely on NLP systems to instantly determine the sentiment of news articles and social media posts. With this information, brands and organisations can, for example, identify unhappy social media users or negative coverage as soon as the content is made public.
NLU helps these companies address problems before they get widespread media attention. But it’s also helping everyday people. Because of NLU, Google Assistant, Siri and Alexa mostly have the right answers to our questions. It will also be the reason why chatbots and robots will be able to perform the tasks we want them to – no matter how we phrase our questions or commands.
Want to stay up to date with the latest media news? Subscribe to our newsletter
More and more artificially intelligent technologies, like NLP systems, are being developed to make our lives – and our work – easier. Read more in our article, How AI will save you time in your workday.
Image designed by Freepik