The Bing team announced new Intelligent Search Features for Bing, powered by Artificial Intelligence, to provide answers faster, give more comprehensive and complete information and allow users to interact more naturally with the search engine.
They announced several new intelligent answers that leverage the latest state of the art in machine reading comprehension, powered by project brainwave running on Intel’s Field-programmable gate array (FPGAs), to read and analyse billions of documents to understand the internet and assist users faster and more confidently get the answers, users need.
Bing now uses deep neural networks to cross check answers by aggregating across multiple reputable sources, rather than just one, this allows users to be rest assured that the answers are verified.
Not every question has a definitive answer and sometimes users may be looking for expert opinions, different perspectives or collective knowledge. In such cases, if there are various authoritative perspectives on a subject, such as pros and cons, Bing will aggregate the two viewpoints from reputable sources and intelligently bring them to the top of the page to save you looking through pages.
If there are multiple ways to answer a question, you’ll get a carousel of intelligent answers, saving you time searching from one blue link to another.
Bing is also expanding it’s comparison answers beyond just products, so you can get a snapshot of the key differences between two items or topics in a table, so you don’t have to sift through numerous dense documents to find what you are looking for. Bing’s Comparison answers understand entities, their aspects, and using machine reading comprehension reads the internet and summarises the differences of the two items to save time.
Bing uses technology built in Microsoft’s Research Labs to help make sense of numbers, users increasingly encounter in the digital world. Bing translates the data into simple concepts so it’s easier to understand what a data like the population of another country means.
These answers will be rolling out to users over the next month in the US with expansion to other markets over time.
More conversational search:
The Bing team often hears that search would be easier if only Bing could complete sentences. It takes effort to know the right words to query. The expertise in web scale mining of billions of documents and with conversational AI with chat bots like Xiaoice and Zo, has created a new way to search that is interactive and can build on users previous searches to get the best answer. Now if the user needs help figuring out the right question to ask, Bing will help, with clarifying questions based on your query to better refine the search and get the user the best answer the first time. The Bing team will be adding more topic areas over time to this experience, and because they’ve built it with large-scale machine learning, the experience will get better over time as more users engage with it. This experience will be rolled out over the next month.
The Bing team also shared more detail on Bing’s advanced image search features. Bing Image Search leverages computer vision and object recognition to give users more ways to find what they are looking for. Search any image or within images to shop for fashion or home furniture. Bing will detect and highlight different products within images or you can click the magnifying glass icon on the top right of any image to search within an image and find related images or products. The team also previewed their landmark recognition technology, which can name a landmark in any picture and share pertinent information about the landmark such as the origins and other relevant trivia.
Robotics News, December 14