Yet another "Microsoft acquires Powerset" blog

PowerSet and MSFT finally announced this rumored acquisition. MSFT has acquired Powerset for close to $100 million. Powerset is a semantic search engine that extends the XEROX-PARC’s licensed linguistic technology. I have tried Powerset when it was a beta and when it got (almost abruptly) released (with search limited to wikipedia and freebase).

The use of semantic technologies for horizontol search engines look highly improbable. The nature of the web is too diverse to be nailed by a single hammer. A related article on making search relevance more meaningful is in this earlier post.

MSFT plans to integrate the Powerset team with its search relevance team and explore advanced search capabilities while taking the Powerset technology beyond wikipedia. MSFT has been creating noise around enterprise and local search, both of which are vertical in nature. This acquisition can add value in making vertical search offerings from smarter and probably scaring the Mountain View behemoth. Whether MSFT takes Powerset’s saplings and nurtures it into its Redmond forests – this only time will tell.

On a related note, I find results from Hakia more useful. Try a sample search for “is EPS the right measure for stock performance” on both powerset and hakia and you will see the difference. Hakia gives results much more relevant to EPS – at least it displays on the very first page results about the use of EPS measure for measuring stock performance. Powerset returned results that had nothing to do with “EPS ~ Earnings per share”. The results from Powerset were purely keyword matches – it mostly matched Extended Play (EP) vinyl records.

In general, most semantic searches seem to be working better on Hakia than on Powerset. I am surprised why Hakia was not approached by suitors if the motivation has been to ramp up semantic abilities for search engines. Or may be it was!

Personalized Search and Disambiguation – The answers to search engine relevance

There is a flurry of announcements around horizontal and vertical search engines offering ‘better’ and ‘more relevant’ content retrieval. has an interesting article on the ability (or inability) of search engines to return relevant results. In this entire discussion of retrieving more relevant there are two seemingly related problems – ‘personalized search’ and ‘disambiguation (of search terms)’.

Personalized search is the ability to convolute the search engine results with the user’s profile in order to return the most relevant search results. The user profile can be created by using the following information:

  • User browsing (search, click-through, time-visited on site, etc) history
  • Desktop data (files, content, most-accessed files)
  • User feedback (asking users to rank order the results based on their preferences)

Continue reading