Looking for Great Ideas: Analyzing the Innovation Jam

In 2006, IBM hosted the Innovation Jam with the objective of identifying innovative and promising “Big Ideas” through a moderated on-line discussion among IBM worldwide employees and external contributors. We describe the data available and investigate several analytical approaches to address the challenge of understanding “how innovation happens”. Specifically, we examine whether it is possible to identify characteristics of such discussions that are more likely to lead to innovative ideas as identified by the Jam organizers. We demonstrate the social network structure of data and its time dependence, and discuss the results of both supervised and unsupervised learning applied to this data.