The traditional underwriting process used in life insurance is based on manually examining an applicant’s health, behavioral, and financial profile. The existence of large historical data sets provides an unprecedented opportunity for artificial intelligence and machine learning to transform underwriting in the life insurance industry. We present an overview of how a rich application data set and survival modeling were combined to develop a state of the art life score, known as lifeScore360, that is inpretable by consumers and carriers alike. Through a novel evaluation framework, we show that the score outperforms traditional underwriting by 6% on the basis of claims.
Transforming Underwriting in the Life Insurance Industry
AAAI Innovative Applications of Artificial Intelligence, Jan. 2019.
M. Maier, H. Carlotto, F. Sanchez, S. Balogun and S. Merritt
The Verge (February, 2019)
Using data from nearly 1,000,000 publicly available resumes, I am constructing a network of occupations. Using the ordered sequences of occupations contained in the resumes, I plan to construct a probabilistic model of career path dynamics that connects the network structure with workers' career path decisions. The goal of this research is to uncover the relationship between the structure of occupation networks and the dynamics of career paths.
Occupation networks: career path dynamics and poverty traps, slides
Using data from the last decade, I conducted a statistical analysis on millions of scoring events from four different professional sports: basketball, hockey, soccer, and American football. I studied inter-arrival times and correlations between events as well as distributions of streaks and final scores, among other statistical properties. This research developed tools for the analysis and modeling of scoring dynamics and shed new light on the existence of universal properties of scoring dynamics in sports.
Scoring dynamics across professional team sports: tempo, balance and predictability
EPJ Data Science, Feb. 2014.
S. Merritt, A. Clauset
Scoring dynamics in professional sports: tempo, balance and predictability, slides
Competition is ubiquitous in complex social systems, from informal online environments to professional sports, to economic interactions between firms. Traditional studies of competition use theory and small-scale controlled experiments to study the outcomes of competition, not the dynamics that occur within them. Here we have analyzed the dynamics of competition using a rich and vast data set from the video game Halo: Reach, using novel quantitative methods and big data processing systems (e.g. Hadoop, Hbase). There are two goals of this research. The first is to produce data-driven mathematical models that contribute to a general understanding of the interactions between participants and the environment. The second is designing new algorithms and tools that have the ability to predict, influence, and control the dynamics in these systems. In doing this, we will gain the ability to systematically design competitive social systems to the preferences of participants.
Social Network Dynamics in a Massive Online Game: Network Turnover, Non-densification, and Team Engagement in Halo Reach
S. Merritt, A. Clauset
Knowledge Discovering and Data Mining (KDD) Workshop on Mining and Learning with Graphs (MLG), Aug. 2013.
Environmental structure and competitive scoring advantages in team competitions
S. Merritt, A. Clauset
Nature Scientific Reports, Oct. 2013.
Detecting friendship within dynamic online interaction networks
S. Merritt, A. Jacobs, W. Mason, A. Clauset
International Conference on Weblogs and Social Media (ICWSM), Jul. 2013.
Inferring massive dynamic social networks from ad hoc human competitions
S. Merritt, A. Jacobs, A. Clauset
NetSci, Jun. 2012. (abstract and poster)
Social Network Dynamics in a Massive Online Game: Network Turnover, Non-densification, and Team Engagement in Halo Reach, slides
Scoring dynamics in team competitions, slides
Detecting friendship within dynamic online interaction networks, slides
The Daily Camera (March, 2013)
How do competing TCP connections interact with one another when transiting a common bottleneck asymmetric link? This is an old problem that still remains unsolved because the fundamental interactions between link speed, buffer sizing, and TCP's congestion control algorithm are not precisely understood. This research describes the interactions in a new way and turns this new understanding into a practical solution.
Two-way TCP connections: old problem, new insight
M. Heusse, S. Merritt, T. Brown, and A. Duda
ACM SIGCOMM Computer Communication Review, Vol. 41, No. 2, Apr. 2011.
Understanding Asymmetric Links, Buffers and Bi-Directional TCP Performance
M.S. Thesis, University of Colorado, Aug. 2010.