Microsoft’s Bing search engine has deployed deep learning techniques for both text & image search.
In a detailed blog post, Bing revealed how the “smart’ search works.
We would like to share how our image search is evolving further towards a more intelligent and more precise search engine through multi-granularity matches, improved understanding of user queries, images and webpages, as well as the relationships between them.
– Bing
According to the post, Bing image search has, some time ago, employed many deep learning techniques to map both query & document into semantic space greatly, in an effort to improve search quality. There were still instances when users looking for certain objects were not satisfied by the results. This prompted Bing to develop further enhancements, & the new vector-match, attribute-match & Best Representative Query match techniques helped address even these problems.
To read up the rest of the post, click here.
