Title :
Deeper into innovation forecasting
Author :
Scott W. Cunningham
Author_Institution :
Faculty of Technology, Policy and Management, Delft University of Technology, Netherlands
Abstract :
Recent work questions whether publication and patenting time series actually follow the familiar S-shaped growth curve. Evidence suggests that most fields of scientific activity undergo dramatic bursts of growth, growing by two orders of magnitude in a matter of a year. Scientific activity in research fields may be measured by appropriately selected keyword phrases. The dynamics of publication suggest temporary, higher order positive feedback loops. These may involve the intellectual migration of scientists to nearby fields of interest, or it may involve other community-related benefits created by having a suitable group of likeminded researchers at hand. Unfortunately, given standard publication by year counts we cannot be certain what the governing dynamics of the system actually entail. In this paper we supplement standard innovation forecasting measures with additional richer details including numbers of unique authors over time, use of novel vocabulary, and citation patterns. This is used to prove, or disprove, a number of competing hypotheses about the emergence of new scientific fields. Recommendations are provided for using these extended indicator systems for innovation forecasting.
Keywords :
"Technological innovation","Maximum likelihood estimation","Indexes"
Conference_Titel :
Management of Engineering and Technology (PICMET), 2015 Portland International Conference on
DOI :
10.1109/PICMET.2015.7273181