By Ultan O'Broin-Oracle on Jun 22, 2016
Yes, the whole Boaty McBoatface thing has now entered the language space.
Boaty McBoatface: Your future in translation may lie in machine learning and related technology.
Parsey McParseface, part of Google's SyntaxNet, an open-source neural network framework implemented in TensorFlow that provides a foundation for Natural Language Understanding (NLU) systems is out there. Google tell us:
"Parsey McParseface is built on powerful machine learning algorithms that learn to analyze the linguistic structure of language, and that can explain the functional role of each word in a given sentence. Because Parsey McParseface is the most accurate such model in the world, we hope that it will be useful to developers and researchers interested in automatic extraction of information, translation, and other core applications of NLU."
I wonder could Parsey McParseface have a role in determining if a translation was correct or not, given the context (i.e., as the UK's Daily Telegraph newspaper would so earthily have it, act as a "bolloxometer")? Whither the QA or real-time interpretation possibilities. In fact, the Globalization, Internationalization, Localization, and Translation (or GILT) industry offers a fertile ground for innovation and exploring possibilities: from pop-up restaurant ventures to pondering the age-old man versus machine-type questions.
This is all fascinating stuff on one level. But it is a serious business on another. Definitely, machine learning is a driver of smart user experiences, along with other areas.The Oracle Applications User Experience team (OAUX) is, naturally, exploring all these areas and what they offer for the smart user experience of tomorrow's world of work. Check out Smart User Experiences: Machine Learning and the Future of Enterprise Applications on the Voice of User Experience (VoX) blog to get a great primer on what technology can do.
Oracle partners, and customers too, need to be on board with these emerging technologies and explore their possible application for user experiences. Often, for partners in particularly, emerging technology and research and development is a "chicken-and-egg" situation: they cannot sell something unless they have it; yet they won't have it unless someone asks for it! That said, we (OAUX) are here to help partners build solutions they can show and sell.
It's the kind of thing I had intended to talk about at Localization World 31 in Dublin, with a language angle naturally (yes, I even included Parsey McParseface). Alas, personal circumstances intervened and I did not speak. Some other time perhaps.
In the meantime, I am sharing the slides I had intended as a backdrop to the discussion. Perhaps they will help you orient yourself to the differences between machine learning, artificial intelligence, NLP, Big Data, robots, and more. They may even help you figure out if you whether you might end up owning a robot or working for one, and what your future working life might look like.