The founders of pioneering shopping search engine MySimon have teamed up again to produce Become.com, a new service that provides information and options for all aspects of the shopping experience.
Become.com is starting with an impressive index of more than 3.2 billion pages of shopping related information from more than 40 million web sites. Unlike many other shopping search services that rely on feeds and sponsored listings, Become.com is uses crawler-based technology designed to focus exclusively on shopping related information.
Currently, Become.com crawls U.S. sites that provide product and buying information. The site is free to use; no registration is required.
The crawler automatically identifies and categorizes web content, indexing only shopping related information and discarding other types of content. Become.com also uses human editors to further refine its index. “Before we push our latest index into production, our researchers identify authoritative and highly rated pages, and these are given preference in ranking and crawling,” said Yeogirl Yun, Become.com’s founder, chairman and CTO.
The result is a shopping search engine that provides lots of reviews, product details and other helpful information while you’re in the research stage, as well as comparison shopping tools and direct links to merchants when you’re ready to buy.
Become.com has a number of behind-the-scenes technologies that are helpful, such as a shopping-specific spell check feature that uses a dictionary including product and brand names, model numbers, SKUs and other information missing from standard search engine spell check programs.
The “search suggestions” at the bottom of result pages are generated by analyzing large numbers of pages related to your query and identifying key concepts, which are then used to help you further refine your query.
The company has developed a ranking algorithm called AIR (Affinity Index Ranking) that performs “contextual connectivity analysis” on the link structure of the web. This algorithm differs from other link analysis techniques such as Google’s PageRank largely because it looks at both incoming and outgoing links for a page.
And unlike PageRank, which essentially pays no attention to the content of a web page, relying entirely on link analysis to compute a ranking, AIR analyzes the content of a page and only indexes it if it can determine that the page contains shopping-related content. It also only crawls on-topic links, discarding non-shopping or spam links.
Yun says that since Become.com focuses exclusively on shopping related content, it’s comparatively easy for the AIR algorithm to recognize spam.
Currently, the company is relying entirely on crawling to build its index, though there’s no mechanism for submitting URLs for consideration. A short FAQ for site owners offers more information about the BecomeBot crawler and how to control its behavior.
In time, the company plans to allow merchants to submit feeds, but like Google’s Froogle, will not charge for this service. Instead, Become.com plans to implement a cost-per-click model.
Michael Yang, Become.com’s founder, president and CEO says the company plans to grow initially by word of mouth, building organic traffic to get word out. Later, the company plans to do syndication deals with other sites, and may do paid keyword advertising to get traffic.
Become.com is a site worth keeping an eye on. After selling MySimon to C|Net in January 2000 for $700 million, Yun created the Wisenut search engine and sold it to Looksmart in April 2002 in an all stock deal valued at just under $10 million at the time.
Want to discuss or comment on this story? Join the Become.com Crawler Out on the Prowl discussion in the Search Engine Watch forums.
NOTE: Article links often change. In case of a bad link, use the publication’s search facility, which most have, and search for the headline.