Yahoo published on its Search Blog a discussion with Dr. Ben Shahshahani, head of Search Sciences at Yahoo Labs. Here are the main outlines of the conversation.
Yahoo's Approach To Search
Echoing the words of the company's Senior VP of Search, Shashi Seth, and definitely putting an end to speculation that Yahoo is abandoning search outright through its Search Alliance with Microsoft's Bing, Dr. Shahshahani explained Yahoo's approach to search as an ongoing process towards more and more sophistication. "We like to consider search not as a stateless, information-extraction, but sort of an ongoing dialogue between the user and the system. User intent even in the context of search can be beyond the query that the user submitted; it could be in the context of the entire session," he said.
He went into more details on search shifting towards sorting through layered content. The content comprises all sorts of media such as video, photos, local listings, real-time news/results or social data. The challenge for the Labs team is to surface the results to users, regardless of the difficulty, insofar as it's no longer just text on a page but rich content.
"We have a pretty active technology and science effort in trying to understand the main object, attributes, and relationships - not just the text on a web page," he said. Yahoo Labs "focuses on the scientific areas of information retrieval, machine learning, data and text mining, and natural language processing," Yahoo said in the blog post.
Search suggestions, which appear in your search box as easy scroll-down options are in fact the result of predictive analysis that enable the engine to surface the most relevant search terms for a specific user. Mix that with the real-time factor and you get suggestions related to the trending topics at the moment you start typing in your query. Dr Shahshahani cited the example of the World Cup: "you start typing in "w-o-r," you're going to see queries that are related to World Cup instead of a less timely suggestion like 'world map'." "... we are really looking towards trying to guess the intent of the user even before they start submitting a query, to give them the information they need before they even think to search," he added.
Context and Social
He went on to put forth the two big trends of search - namely contextualized search and social search. About Social search, he warned that although Yahoo has pushed its integration of social platforms, the company's search feature won't necessarily serve the user a thread of results from social platforms if it deems it not relevant to the person.
While speaking yesterday with Seth, we asked him whether Yahoo was storing personal data in order to be able to predict user-targeted content. His answer was "everyone does that." Indeed, country-specific legislation though regulates the length of time during which the data is being kept. Typically, it is around two years.
Other pointers for user-targeted content are the social "currencies" or "rewards" that take the form of "likes," "thumbs ups/down," or "check-ins."
You can checkout the video for yourself below: