Wednesday 29 April 2009

New suite of tools to enhance search capabilities

I thought this announcement from Califronian based 'DeepDyve' would be of particular interest given the growing use of semantic web techniques.


DeepDyve Unveils Suite of Tools for Publishers

New Tools Enable Advanced Search and Discovery

APRIL 28, 2009 – Sunnyvale, CALIF — DeepDyve, the research engine for the Deep Web, today unveiled a suite of tools for publishers and scientific societies of all sizes that want to enhance the search capabilities on their websites. The suite of tools leverages DeepDyve’s KeyPhraseTM algorithm, which allows users to input whole sentences, paragraphs or entire articles as their query to find related results.
"Our vision is that search is becoming more sophisticated and more decentralized. Increasingly, users are initiating their research online and they want to have search integrated seamlessly with their reading and browsing behavior — in other words, they want their content to be their query for finding comprehensive answers to difficult questions. These tools give our partners the ability to make their content more findable and to demonstrate the breadth and depth of their collection," said William Park, CEO of DeepDyve. "We're making available some of our most frequently used search capabilities to publishers that want to give their visitors a more compelling search experience."

The products being announced today are designed to first make the publisher's content more discoverable in search engines which is where, according to a report from Outsell, more than 70% of users begin their research. From there, other tools are available to increase the engagement at the publisher's site by allowing users to quickly find related articles based on what they are viewing.

Publisher Landing Pages
Publisher Landing Pages use DeepDyve's technology to enhance the findability of publisher content by Google and other search engines. These co-branded or private label pages present the Web searcher with not just one article at a time, but with a whole page full of closely related articles from the same publisher, with links pointing them to the publisher site. One Landing Page can be set up for each article, yet they are easy to deploy, totally automated and hosted by DeepDyve.

Custom Search API
DeepDyve's next-generation search technology is available to publishers via a web services API (Application Programming Interface). The API can be set up to search only a publisher's own content, or to help users discover other highly relevant documents in the DeepDyve index. Searches can be launched with a few keywords, or by allowing users to use a paragraph or an entire document as a query to find articles that match the concepts described. Results are returned via an XML feed or as a hosted, co-branded web page.

More Like This Document API

The DeepDyve More Like This Document API enables websites to directly interface with the DeepDyve database to search for articles that are similar to a designated document within the DeepDyve index. It is designed for use by publishers whose content has been indexed by DeepDyve and who would like to include a 'related articles' functionality on their site without the painful implementation. The user may select any document to use as a query, and the title and body are compared to other documents in the DeepDyve index. The resulting documents can be limited to a publisher's own content, or may include other content in the DeepDyve index. Results are returned via an XML feed or as a hosted, co-branded web page.

Content Highlight Widget
The Highlight Widget enables users at the publisher site to simply highlight any block of text up to 5,000 characters, then run that selection as a query. DeepDyve returns only the Publisher's articles in the search results via an XML feed or as a hosted, co-branded page of results.
DeepDyve is a search engine that was developed to scour the depths of the so-called Deep Web, the vast collection of information-rich content that is largely overlooked by today's traditional search engines. Since the company's launch in September 2008, DeepDyve has worked closely with major publishers, building an index with hundreds of millions of pages that showcases content from the industry's most respected research organizations, academic institutions and professional associations. The API tools that are being announced today are the next step in DeepDyve's vision for enabling publishers to better utilize the Internet to reach as large an audience as possible.

Early Customer Adoption
The Public Library of Science (PLoS) is one of the first organizations to implement DeepDyve's search technology.
"We are looking forward to DeepDyve powering the search on," said Peter Jerram, CEO of The Public Library of Science. "DeepDyve's cutting-edge technology and ability to use entire sentences as a query will make it much easier for our users to find and discover new original research in science and medicine."

Pricing and Availability
Each of these tools is available for free with advertising revenue sharing, or for a fee which varies depending on volume.

About DeepDyve
DeepDyve, formerly known as Infovell, is the research engine for the Deep Web. DeepDyve unlocks the vast and rich collection of information that is out on the web, but is hidden from today's search engines. Using DeepDyve, people find the in-depth, high-quality information they need to answer tough research questions. The company is headquartered in Sunnyvale, CA, with offices in Shanghai, China. To learn more about DeepDyve, go to or call 1-408-773-0110.

No comments:

Post a Comment