 "Microsoft Bing")
Microsoft has announced a new experimental Deep Search feature for Microsoft Bing. In a blog post on Bing Blogs, Microsoft said that Deep Search provides more relevant and comprehensive answers to the most complex search queries.
Microsoft said that the Deep Search feature will not replace Bing's existing web search, but offers an enhancement that offers the option for a deeper and richer exploration of the web.
Deep Search feature uses the GPT-4 language model to expand a search query into a more comprehensive description. The expanded description might capture users' intent and expectations more accurately and clearly than a few keywords. It also helps Bing understand what kind of information the user is looking for.
In case the input query is ambiguous, Deep Search figures out all the possible intents and computes a comprehensive description for each of them. The feature then prompts the user to select the right intent of their query and the corresponding comprehensive description that should be used.
For example, a web search query for "how do points systems work in Japan" could refer to rewards points, immigration policy or something else. In that case, Deep Search generates an expanded description for each possible intent and asks the user to select one.
Deep Search then uses the web to fetch results that might match the expanded query that it has generated on the users behalf. Once it has gathered a collection of web pages, it reviews and ranks them according to how well they match the comprehensive description.
Deep Search uses a variety of signals to determine the relevance and quality of each result, considering factors like how well the topic matches the query, its level of detail and the source's credibility.
Microsoft said Deep Search is not meant for every query but is designed for those who have complex questions that require more than a simple question.
