With the help of algorithms scientists scan the rapidly growing amount of data in order to prevent diseases, improve life and to maximize business profits. In Germany, however, one may encounter obstacles in the field of data mining.
For almost 30 years FIZ Karlsruhe has been dealing with data mining. Via the online service "STN International" the Leibniz Institute for Information Infrastructure provides global access to over 150 databases containing about 1.5 billion documents with professional research and patent information. At FIZ Karlsruhe, the employees are convinced that these documents have more potential. With the help of the STN platform, an important data technology tool, the research institute has successfully been able to deal with data and text mining for many years.
The scientists of FIZ Karlsruhe and the University of Hildesheim are now taking this a step further. From the databases of STN they are now trying to identify developments and trends as soon as possible. This is called "Trend Mining". In this context significant terms are identified for certain patents in order to keep track of their occurrence in the databases. "If one term appears more often than others during a certain period of time, this could indicate a developing trend", explains Michael Schwantner, who is in charge of the project. The information experts analyze these hints in detail to check whether these trends are exploitable. For example, in 2012, a search in an STN patent database suddenly showed a meaningful increase of patent publications for so-called flow batteries. In 2013, the number remained significantly higher than in the years before 2012. This is a clear indication that the research for such liquid batteries with memory technology is booming. These types of trends are especially interesting for companies that are highly involved in research and development. In the pharmaceutical industry research for the enforcement of a patent often costs several million euros. The development of drugs, which often takes many years, may even cost billions. Therefore, identifying what the competitors are focusing on in their research at a very early stage could contribute positively to the business success.
Extract from the Leibniz-Journal 1/2014: Der vernetzte Mensch (translation by FIZ Karlsruhe).
Complete download: Leibniz-Journal 1/2014 (PDF-File, 8.5 MB).