Business impact Business Insights Workbench (BIW), an innovative set of techniques developed by IBM Research, enabled this pharmaceutical company to mine both structured and unstructured data from public domain information, then to analyze the results in ways that offered better understanding and enhanced competitive strength.
Issue Intellectual property and patents are central to the value of products developed by pharmaceutical and biotechnology companies. To protect their properties while staying on the cutting edge of innovation with unique product development, these companies must keep the details of their own patents readily accessible. In addition, fierce competition and the stipulations of patent laws require that they continually monitor patents owned by their competitors.
The ability to explore both structured data, such as numerical values, and unstructured data, such as notations created in patent files, offers significant value in internal and external analyses and strategic planning. In-depth data-collection capabilities also can be useful beyond the pharmaceutical development and biotechnology fields, into such areas as quality and customer relationship management programs (examining repair reports, for example), purchase order and compliance, and insurance and healthcare record analysis.
Executive summary This international healthcare products company sought to gather data about patented pharmaceuticals to analyze details of existing patents, gain insights into industry trends and needs, and enhance its total product strategies.
The challenge was daunting: In the public domain alone, the U.S. Patent and Trademark Office database contains more than 7.9 million full text and drawings of U.S. patents since 1976 and full-page images since 1790. In addition, Medline, an indexing service for research in medicine and related fields provided by the U.S. National Library of Medicine, contains nearly 15 million citations and abstracts in the life sciences and biomedical fields.
What IBM did Using a new class of technology that allowed the company to mine and explore both structured and unstructured information, IBM Research worked with the company to search through vast databases and extract millions of unique chemical name allocations, identifying various compounds recorded in the patent and Medline documents. The same technology can be expanded to other areas of interest, such as diseases and genetic data.
Capabilities applied The solution was developed using the IBM Business Insights Workbench (BIW), which uses WebSphereŽ Information Integrator OmniFind technology for full-text search and content extraction. BIW’s capabilities explore vast stores of both structured and unstructured data.
To search and analyze unstructured information, BIW creates natural classifications (taxonomies) of the data objects, using expert human intervention at every stage of taxonomy generation and modeling. Once a taxonomy is complete, it is treated as another dimension in the database and analyzed against other taxonomies or in conjunction with structured information to find trends, co-occurrences and other insights.
BIW enables real-time data manipulation to offer an extensive range of capabilities, including the ability to retrieve minute levels of detail, analyze specific parts of a document, visualize data categories over time, compare and merge analysis from different sources or previous analysis, detect patterns using multi-dimensional affinity analysis, automatically identify non-obvious relationships with scatter charts, and automatically detect knowledge gaps in a set of text and suggest areas for further analysis, as well as to prioritize solution suggestions according to return on investment and strategic value.
| |
