Data mining has the potential to reveal tremendous amounts of previously hidden information and insight. In countless areas, organizations are leveraging data mining solutions to revolutionize their operations.
One particularly interesting and potentially valuable ongoing effort in this area can be found at Oxford University. There, Hyejin Youn and several colleagues are using data mining tools to examine hundreds of years of patent office records to better understand the nature of innovation and invention, the MIT Technology Review reported.
Old records, new insights
As the news source pointed out, many technologists see the nature of invention as one of ongoing combinations. An innovator does not come up with a new idea with no precedent, but rather finds new combinations of existing notions to create something that did not exist previously.
Through their data mining project, Youn and his colleagues aim to explore the truth and limits of this hypothesis, the MIT Technology Review explained. The researchers are conducting data mining and analytics upon information collected by the U.S. Patent Office, whose records date back to 1790. In the Patent Office's system, every new invention receives a code to identify which pre-existing technologies it relies upon. A device that depends on a single technology receives a single code while a gadget that uses multiple pre-existing technologies receives a combination code. According to the researchers, this system created the possibility to effectively explore the deeper nature of how inventions relate to one another and how innovation evolves over time.
Youn's analysis has revealed that approximately 40 percent of all new inventions registered in the U.S. Patent Office rely on previously existing combinations of technology. Sixty percent represented a new, unseen combination of technologies, the news source reported.
As the MIT Technology Review noted, these findings have significant implications.
"The huge gap between the possible and the actual number of combinations indicates that only a small subset of combinations become inventions," Youn and his colleagues explained in their study, the source reported.
Perhaps even more importantly, these findings may hold significance for biological evolution. For both new technologies and organisms, existing combinations inevitably play a major role in determining the future path of development.
"Studying patent, comparative and systemic records of inventions will open a way to make quantitative assessments for a counterpart of these features of biological evolution in technological evolution," the report explained.
Data mining impact
This study, along with countless others, demonstrates the simple fact that for all of the ways that data mining has already been implemented, there are still many new applications to be discovered, and these efforts can yield extremely valuable insight.
For data mining efforts to prove useful, though, organizations must consistently apply best practices to their data mining projects. Choosing effective commercially available mathematical and statistical functions is key to speeding up development and reducing risk. Leveraging such resources can have a major impact on the ultimate potential of any given data mining project.