Elsevier’s new Web-based research tool integrates natural language search with billions of search data points.

Elsevier, a publisher of scientific, technical, and medical information products and services with U.S. headquarters in New York, N.Y., recently announced their launch of illumin8 (www.illumin8.com). This is their first Web-based research tool that integrates natural language search technology with content from Elsevier’s full-text scientific articles, millions of scientific abstracts, patents, and billions of web sources to give users actionable solutions for research initiatives.

To build illumin8, Elsevier partnered NetBase Solutions, Inc. ( www.netbase.com), a leading provider of research software tools for enabling knowledge workers to mine the Internet and other online data to accelerate their research. Many Fortune 500 companies in a variety of vertical industries already use NetBase technology to perform technology intelligence, technology scouting, competitive intelligence, and market discovery. NetBase was developed initially at the Massachusetts Institute of Technology’s Media Labs in 2003 and has evolved to become the world’s largest natural language index covering more than 1.1 billion solutions.

“We found illumin8 to be a unique and effective way to mine the Internet and premium content for solutions to technical problems and questions,” says Mary Poul, marketing manager at 3M Medical Systems, Westhaven, Conn., an illumin8 beta tester. “The value for us is the unique search capability with the deep and broad content set. Further, doing the research without illumin8 would have taken weeks, and I would have spent countless hours looking at data that was not relevant to my project.”

How it works

Illumin8 combines search and semantic indexing technologies to distill deep meaning, purpose, and insight from the vast amount of Elsevier’s full-text content, scientific abstracts from 4,000 publishers, patents, and billions of web pages. The tool extracts and analyzes solutions, which are then categorized under organizations, products, technologies, approaches, and experts. Illumin8 is designed to go beyond simple keyword searches, quickly finding and extracting crisp summarized answers and interrelationships that are semantically related to the context of the search query.

In addition to finding solutions from more than 5 billion web pages and millions of patents, illumin8 users will be able to easily access the full text of Elsevier journals if they have an online subscription to the journal through ScienceDirect—the world’s largest online platform of scientific, technical, and medical content.

“R&D knowledge workers no longer need to spend hours, days, or weeks sifting through thousands of search results or full-text documents, and then manually synthesizing results to find actionable answers,” says Rafael Sidi, VP for product development at Elsevier’s engineering and technology division.

Elsevier developed illumin8 because they saw a need to reduce the R&D risk and fully exploit the investments that companies are making in their R&D operations. They saw a need for decision support systems that could help these companies identify both opportunities and threats. And they didn’t see any competitive products that were able to satisfy this need.

Their initial research (from Outsell Inc.) found that researchers spend about 5.5 hr/wk gathering, looking for, and collecting information. They then spend about 4.7 hr/wk analyzing and applying that collected information. Researchers’ current research methods include a variety of methods including GYM (Google, Yahoo, and Microsoft search tools), consultants, networking, tradeshows, expert marketplaces, tech scouts, journals, and peers. The problems encountered with these methodologies is that the researchers had difficulty in identifying relevant information, it took weeks for them to complete their searches, it was very resource intensive, it was expensive, and often the results were based on unfocused efforts and provided unpredictable results.

This last item was often most important—conventional searches yielded a lack of trusted answers to their questions—the conventional search tools were very inefficient, and while there is a very large amount of information available, that data lacks insights into problems.

Specific searches

Illumin8 starts its search by crawling through:

• 5 billion web pages, blogs, and forums;
• 3 million full-text scientific and technical articles from 1,800 Elsevier journals;
• 33 million scientific records from 15,000 peer-reviewed journals (>4,000 publishers); and
• 21 million patents from five worldwide patent offices.

It performs these searches by topic, problem, or benefit. It then extracts and synthesizes summarized solutions and related entities and puts them into a database in seconds. Finally, it filters, visualizes, and analyzes the solutions to provide which products, solutions, organizations, technologies, and/or experts provide the best solution.

Illumin8 performs these intelligent searches by extracting specific concepts, ideas, and entities from a single sentence in the searched items. Users input into illumin8:1) the specific terms describing a benefit, capability, or problem they’re looking for; 2) the technology, material, ingredient, chemical, process, or technique involved; and 3) the company or organization name that’s most appropriate. In return, they obtain from illumin8:1) the companies or organizations that are affiliated with solutions to the search input; 2) products that offer solutions related to the search input; and 3) approaches, processes, and techniques related to the search input.

Later this year, Elsevier will introduce enhancements to illumin8 which include preferences, filters, annotable solutions, and shared solutions. There is an ambitious customer-driven roadmap for illumin8 for the rest of the year that will make the product even more powerful for researchers.

—Tim Studt