Linguamatics and InforSense this week rolled out major new upgrades to their respective text-mining and -analysis software products in response to what they describe as increasing demand for such tools from the life science market.
Linguamatics, which specializes in natural language processing technology for the life sciences, released version 3.0 of its flagship I2E software, which includes a number of new features for improved querying and ease of use.
InforSense, meantime, released a new product called Document Explorer as part of a three-pronged product launch that also includes a new visualization module for its workflow platform and a central repository that allows organizations to store and share workflows internally.
Document Explorer is designed to augment the text-mining capabilities of InforSense’s TextSense module, and offers a browser interface for interactively browsing, searching, and viewing unstructured data.
Dave Menninger, vice president of InforSense, told BioInform that the new product serves as a “simple, easy-to-use front end” for users who want to deliver the results of TextSense searches to their colleagues.
Menninger cited an increasing awareness of textual information in the life science research community as a key trend that the InforSense sales team is focusing on in the year ahead. “There’s a lot more activity there than we have seen in the past,” he said.
Linguamatics is also bullish about the text-mining market. “There’s a groundswell of recognition that this technology can help people,” said Phil Hastings, director of business development. He added that the company is seeing increasing interest from customers who realize that they “need to be more efficient in mining textual information.”
Nevertheless, he acknowledged that “text-mining is still a new technology for most people, and one of our goals is to make it more accessible.” Therefore, a number of improvements in I2E 3.0 are related to ease of use, he said.
There is “still quite an educational challenge” involved in marketing text-mining software, Hastings said. “People are not always aware of what text-mining means, how it differs from text search, and what it can do for them. … It’s still an educational challenge to bring awareness to everyone that might be interested.”
The new version of Linguamatics’ I2E includes a feature called “smart queries,” which “hides” the “complexity” of natural language processing from the end user, Hastings said. For example, users could simply fill out a template to indicate what gene class or disease area they are interested in, and the software automatically extracts a "range of different relationships" between those entities of interest.
I2E 3.0 also includes an automated method for disambiguating gene names and improved reporting capabilities.
Hastings said that the privately held Linguamatics “had a great year last year” financially and is currently profitable. He said the Cambridge, UK-based firm currently employs 15 staffers and has plans to expand over the next year, though he did not provide specific details.
Text, Visualization, and Workflows
InforSense is likely to benefit from the improvements in I2E as well. The companies signed a partnership agreement in 2006 under which InforSense offers access to I2E through its TextSense module.
The new Document Explorer module was designed as “a browser-based interface that delivers text-based information in a domain-specific fashion,” Menninger said. “It’s a way to use domain-specific categorization or ontologies and navigate to documents most relevant for your task.”
“There’s a groundswell of recognition that this technology can help people.”
In addition to the new text analytics tool, InforSense also threw its hat in the data-visualization ring this week — a move that brings it into the realm of its long-time partner Spotfire.
Menninger said that the company’s new VisualSense system offers many interactive visualization capabilities that are available in Spotfire’s DecisionSite and other visual analytics tools, but in a browser-based application rather than a fat client.
Menninger said that InforSense’s partnership with Spotfire has not changed, but the company decided to build its own visual analytics offering in response to customer demand. Following Spotfire’s acquisition by Tibco last year, he said, some customers were “less happy with Spotfire post-acquisition than they were prior to the acquisition” — feedback that he attributed to the slower responsiveness of the resulting larger organization.
However, he noted that this “created an opportunity” for InforSense to develop its own visualization tools in order to “offer a range of capabilities within one environment” for those customers who prefer to deal with a single, smaller software provider.
InforSense also released a third new product this week — a central repository called the InforSense Workflow Library that is targeted toward developers within large organizations who want to replicate analytical processes across the enterprise.
The system stores internally developed workflows in a secure environment so that they can be disseminated across an organization.
The model is similar to the company’s Customer Hub, which is a repository of workflows that is open to the entire InforSense customer community, but Menninger said that the Workflow Library is designed for corporate customers who have information that they don’t want to share externally.