Abstract—This paper considers the task of semantic innovation (new idea or trend) detection possibility in corporate information field applying the analysis of dynamic changes in enterprise information flows. Here we consider such hypothesis that the sign of innovation appearance is the dramatic increase of new terminology employment (that has occurred rarely or has not been used at all), or the appearance of previously not occurred combinations of old “terms” (words) and “N-grams” (sequences of words). Presented results of this experiment are based on ENRON, TREC2005 databases, email exchange among employees of Russian company. We have overseen conditions under which the innovation detection is statistically significant and made the conclusion about the future of above-listed hypothesis of innovation detection.
Index Terms—Innovation detection; innovation sources; terms; innovation; managing technological change; computeraided text analysis; knowledge management; enterprise content; unstructured information
The authors are with S&E Center of Information Control Technologies National State University Higher School of Economics (e-mail: firstname.lastname@example.org, email@example.com, firstname.lastname@example.org)
Cite: Dmitry Romanov, Mikhail Ponfilenok, and Nikolay Kazantsev, "Potential Innovations (New Ideas/Trends) Detection in Information Network," International Journal of Future Computer and Communication vol. 2, no. 1 pp. 63-66, 2013.