Social Media Analytics
In our recent work on user profiling in social media we have looked at the role of conversational structures plays in identifying personality (Appling, Briscoe, Hayes, & Mappus, 2013). Current work applying this knowledge has shown a mean precision improvement of 20% on predicting dimensions of personality when conversational structure is combined with content analysis.
See: Appling, D. S., Briscoe, E. J., Hayes, H., & Mappus, R. L. (2013). Towards Automated Personality Identification using Speech Acts. Workshop on Computational Personality Recognition. International AAAI Conference On Weblogs And Social Media (ICWSM), 2013.
PCA Analysis of Credibility Factor Dimensions
See: Briscoe, E., Appling, S., & Hayes, H. (2014). Cues to Deception in Social Media Communications. Proceedings of the 47th Hawaii International Conference on System Sciences. Hawaii, Hawaii.
Additionally, we are conducting research on the differences in automated measurement methods of sentiment (opinions) and how they compare to traditional polling methods, especially during times of civilian uprest.
See: Weiss, L., Briscoe, E. Hayes, H., Kemenova, O., Harbert, S., Li, L., Lebanon, G., Stewart, C., Miller, D. Foy, D. A Comparative Study of Social Media and Traditional Polling in the Egyptian Uprising of 2011. Proceedings of the 2013 International Conference on Social Computing, Behavioral-Cultural Modeling, and Prediction. (2013).
Price Targeting for Optimal Adoption
Also we have conducted research into using social network information to target individuals in terms of optimizing profit through pricing strategies .
See: Galloway, E., Mappus, C., & Briscoe, E. (2013). Social Price Targeting: Use of Network Structure in Firm Decision Making. 9th Conference of the European Social Simulation Association Warsaw School of Economics, Warsaw, Poland, September 16-20, 2013.
Comparison of Pricing Strategies in a Social Network
Insider Trading Social Network Analysis
Chip Mappus has been working on determining insider-trading strategies through the use of network analysis of companies’ insider rosters.
Heat-map depiction of anomalous computer activity
Insider Threat Detection
We have also modeled behavior to understand and detect when people are acting maliciously.
See: Mappus, C. & Briscoe, E. (2013). Layered behavioral trace modeling for threat detection. Proceedings of the 2013 IEEE Intelligence and Security Informatics Conference. Seattle, Washington.