Relevancy of results has been the dominant theme in my semantic search posts so far. And in this piece I will be discussing about SenseBot, a search tool that has the added feature of summarizing results. Very often, the intent of a user is to obtain brief information on a concept or a keyword. In such scenarios it is more relevant to the user if the results describe the phenomenon or event more and don’t just pluck in documents that contain the keywords.
SenseBot, an engine that summarizes the top results for a query (from Google and Yahoo) presents brief summaries on the keywords that the user selects. Certain examples with results can be viewed here. Most search engines either present summaries from a particular link or just present the whole bunch of links to the user. SenseBot’s approach, though far from perfect, is a peek into what the search engines of the future could be capable of doing very well – Present the whole swath of data in a lucid style with user selected detail.
The engine uses Natural Language processing combined with Multi Document summarization techniques to present a summary that becomes the main link for the result. In this Q/A session featured on AltSearchEngines, Dmitri Soubbotin of SenseBot details on several of SenseBot’s features.
Prospects
The engine is still in beta so it’s too early to expect expert grade summaries on topics (Read a review at the Unofficial SEO Blog) . Also, with a plethora of online encyclopedias, SenseBot’s technology would be taking Search engines from Link providers to information providers. As mentioned by Dmitri, the engine can return more relevant summaries if associated with vertical segment (medical, legal) or as an added feature that feeds major search engines. Do try SenseBot and share your comments on it.
