Bing's new Deep Search feature
Microsoft's Bing browser is getting a new "Deep Search" feature powered by OpenAI's GPT-4. This feature is designed to give users more relevant and comprehensive answers to complex search queries. Microsoft notes that Deep Search is not a replacement for Bing's existing web search, but is instead an enhancement that allows for deeper web exploration.
In a blog post, Microsoft explained that the new feature builds on Bing's existing web index and classification system and enhances them with GPT-4, which takes a search query and turns it into a more comprehensive description of what the results should include.
For example, let's say a user is searching loyalty programs in different countries and enters the query: "How do points systems work in Japan." The deep search will take the query and expand it to the following:
Provide an explanation of how different loyalty card programs work in Japan, including their benefits, requirements, and limitations. Include examples of popular loyalty cards from different categories, such as convenience stores, supermarkets, and restaurants. View a comparison of the pros and cons of using loyalty cards versus other payment methods in Japan, including current rewards and benefits. Highlight the most popular services and participating merchants.
By expanding this description, you can explain your intent better than you could with just a few words.
In cases where your search query is more vague, Deep Search will find all possible intents and create a comprehensive description for each. Deep Search shows you these intents, allowing you to choose the right one.
Image credits: Bing
Once the expanded description is created, Bing will pull relevant results that don't often appear in typical search results, Microsoft says. Deep search finds pages that might match the expanded query, rewrites the query and then looks for those differences as well.
Sticking with the same loyalty points query example above, a deep search might also search for things like “Loyalty card programs in Japan,” “Best loyalty cards for travelers in Japan,” “Compare loyalty programs by category in Japan,” and “Redeem cards Loyalty. In Japan” and “Managing loyalty points through mobile 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 per search, and deep search does 10 times that to find results that are more useful and specific than those that rank higher in a regular search.”
Once Deep Search collects web pages, it ranks them based on how well they match the expanded description. When ranking, Deep Search takes into account how well the topic matches the query, whether it has an appropriate level of detail, how authoritative the page is and how recent and popular it is.
Microsoft notes that Deep Search is optional and can take up to 30 seconds to complete, which is why it's not tailored to every query or user. For users who don't want more comprehensive answers, they can get regular Bing search results instantly. Microsoft Bing is getting a new "Deep Search" feature powered by OpenAI's GPT-4. This feature is designed to give users more relevant and comprehensive answers to complex search queries. Microsoft notes that Deep Search is not a replacement for Bing's existing web search, but is instead an enhancement that allows for deeper web exploration.