'Bing Search Quality Insights' Giving Deeper Insight into the Algorithms, Trends and People Behind Bing

Dr. Harry Shum, Corporate Vice President, Bing R&D introduced a new series of blog posts, focusing on "Search Quality Insights" with a note, "There's more web data created in a single day than in the entirety of 1999. Moreover, the types of data crawled by search engines are evolving in astonishing ways. What started as […]

Dr. Harry Shum, Corporate Vice President, Bing R&D introduced a new series of blog posts, focusing on "Search Quality Insights" with a note, "There's more web data created in a single day than in the entirety of 1999. Moreover, the types of data crawled by search engines are evolving in astonishing ways. What started as indexing simple web documents has blossomed into a dizzying array of data types - from rich multimedia content to real-time streams to social conversations, just to name a few."

The series kicks off with a post by Dr. Jan Pederson, Chief Scientist for Core Search at Bing, on "Bing Search Quality Insights: Whole Page Relevance". Jan delves into how Bing go beyond the traditional concept of page rank to deliver rich "answers" like video, images and maps that are relevant and help you get more done. Describing how Bing has evolved Jan stated:

"If you compared the results you see with Bing today with those of our search product five years ago, at first glance they might seem similar - ten blue links presented with text summaries. However, if you look closer you would see some important differences. The first thing you would notice is that many of the queries you perform today produce excellent results that would have simply not been possible five years ago. This is a tribute to the continued growth and evolution of the Web coupled with our unceasing focus on core search quality." And, second, "Bing's search results today include videos, images, maps, news items and other media objects, which we call answers."

He also talks about the Bing technology for blending together blocks of content, Web pages and answers, into a single result set, called "Answer Ranking."

Adding further he writes, "we add in three kinds of additional inputs: confidence scores from the answer provider, query characterizations, and features extracted from other answers and web pages that will be shown on the page."

Alson, that a new blending function has been generated by our offline learning method, it is deployed to a serving component internally called APlus that puts all that data into action and runs after all candidate content blocks that have been generated, where it can be tested via online experimentation and finally placed into production.

In other words, "we can deliver a rich set of results that are statistically more likely to get you what you're looking for- whether it's a link, a video, a news item, a map or snippet of information," he said.

"Most recently we've focused on new inputs that can improve our ability to place temporally relevant answers," says Jan. "For example, the news answer is designed for speed and timeliness. Our aim is for the answer to update as quickly as possible when a news story breaks then decline in prominence as the news cycle for the story winds down."