Search Engine that finds sense in a heap of Web pages
SenseBot (www.sensebot.com) represents a new type of Search Engine that delivers a text summary in response to your search query instead of a collection of links to Web pages. SenseBot parses top results from the Web and prepares a summary of them. The summary serves as a digest on the topic of your query, blending together the most significant and relevant aspects of the search results. SenseBot's approach is unique and protected by patents.
SenseBot is a semantic search engine, meaning that it attempts to understand what the result pages are about. It uses text mining to parse Web pages and identify their key concepts. It then performs multidocument summarization of content to produce a coherent summary. You can read more about the approach in our interview to Search Engine Journal.
Imagine you are not familiar with the concept of the dot-com bubble, and want to understand it. You type “dot-com bubble” into your favorite search engine and get back ten links with some context around them. However, glancing over the context does not really give you a coherent idea of what the dot-com bubble was. So you start clicking on the links, and soon realize that some of the pages presume that you are already familiar with the subject; others are too shallow, or biased, or just aggregate links to other resources instead of answering your question. By the time you finally understand what the dot-com bubble was, you’ve clicked on too many links and sifted through too much redundant information. Wouldn’t you rather click on one link and receive a concise summary, a digest on the phenomenon of the dot-com bubble, compiled from the top most relevant pages on the subject? Enter SenseBot, which does exactly that – generates a summary of the top pages returned by the search engine. After looking at the summary, you should be able to get a good idea of what the dot-com bubble was, understand the dot-com model and its key stages, and also see which of the sources about it are worthy should you decide to dig deeper. Note that SenseBot will skip the sources that are shallow on the content relevant to your topic – no matter how highly they were ranked by major search engines. You can see examples of such queries at our Samples page.
Here are more specific examples. If you are a student writing a midterm paper the night before it’s due, SenseBot could save you time by providing an overview of the topic, and pointing you to the right sources. If you have an interview tomorrow morning, and want to get broadly familiar with a particular area that your prospective employer wants you to know, SenseBot may help you. Better yet, the summary may reveal to you which of the sources are worth drilling into if you need to explore the topic further. Aside from the education and learning, many other types of queries would benefit from SenseBot: medical queries; business intelligence; news analysis; etc. Check out our applications page to see where our technology can be applied.
Generally, any type of search where the user is trying to understand a concept or a particular area of knowledge; or get a digest on a person, an event, or an activity; or quickly identify the best source of information on the topic, will benefit greatly from SenseBot. The more diverse are views on the topic, the higher would be the benefit of seeing their summary rather than studying each view separately. But if you are just looking for a particular fact (e.g., “When was John Lennon’s birthday?” or “What is the lowest price of gas in my neighborhood?”), you really don’t need a summary.
We have seen very encouraging results from SenseBot searches, some of which are posted on the Samples page. We employ sophisticated methods of text mining and Natural Language Processing, proven by many years of commercial application. Still, it would be unrealistic to expect SenseBot to match a summary prepared by a human expert. It is unlikely that you will be able to take a SenseBot summary and hand it to your boss, presenting it as a result of your hard work (but if you succeed, please let us know). However, we believe that even less than perfect summaries may be of great help for Internet users, and be preferable over numerous links to Web pages.
We are continuously working on improvement of the quality of the summaries. We have a number of applications based on the technologies used in SenseBot.