Microsoft Bing is getting a new “Deep Search” feature powered by OpenAI's GPT-4. The feature is designed to provide users with more relevant and comprehensive answers to complex search queries. Microsoft notes that Deep Search does not replace Bing's existing web search, but rather is an enhancement that allows for deeper exploration of the web.
in a blog entryMicrosoft explains that the new feature builds on Bing's current web index and ranking system and enhances it with GPT-4, which takes the search query and turns it into a more complete description of what the results should include.
For example, let's say a user searches for loyalty programs in different countries and enters the query: “How do points systems work in Japan?” Deep Search would take the query and expand it to the following:
“Please provide an explanation of how the various loyalty card programs in Japan work, including the benefits, requirements, and limitations of each. Include examples of popular loyalty cards from different categories, such as convenience stores, supermarkets, and restaurants. Show a comparison of the advantages and disadvantages of using loyalty cards versus other payment methods in Japan, including current rewards and benefits. Highlight the most popular services and participating merchants.”
With this expanded description, you will be able to explain your intent better than just a few words.
In cases where your search query is more ambiguous, Deep Search will find all possible intents and create a complete description for each of them. Deep Search then shows you these intents, allowing you to select the right one.
Image credits: bing
Once the extended description has been created, Bing will return relevant results that often don't appear in typical search results, Microsoft says. Deep search finds pages that might match the expanded query, rewrites the query, and then searches for those variations as well.
Continuing with the same loyalty points query example above, Deep Search can also search for things like “loyalty card programs in Japan”, “best loyalty cards for travelers in Japan”, “loyalty program comparison by category in Japan”, “loyalty card redemption in Japan”. Japan”” and “manage loyalty points with phone applications”.
“By doing this, Deep Search can find results that cover different aspects of my query, even if they don't explicitly include the original keywords,” Microsoft wrote in the blog post. “Regular searches on Bing already consider millions of web pages for each search, and Deep Search does ten times more to find results that are more informative and specific than those that rank higher in regular search.”
Once Deep Search has collected web pages, it ranks them based on how well they match the extended description. When ranking, Deep Search considers how well the topic matches the query, whether it has an appropriate level of detail, how trustworthy the page is, and how new and popular it is.
Microsoft notes that deep search is optional and can take up to thirty seconds to complete, so it is not designed for all queries or users. For users who don't want more complete answers, they can get regular search results on Bing instantly.
