Michael Sun Email & Phone Number

Detroit, Michigan, Us

Santa Clara, California, Us

Newark, New Jersey, Us

Provide actionable insights for the business through data science to improve customer experiences, increase revenue and reduce operational costs. The team under my leadership has contributed about $100M to Prudential’s business. • Lead a team of data scientists to support Prudential’s global business, including Group Insurance, Retirement, and Long Term Care. • Created a new cloud based AI underwriting process for life insurance applicants, which reduces processing time from 30 to 45 days to under 20 seconds and lower the cost of underwriting by 90%. •. Created a comprehensive claims fraud detection program, based on supervised and unsupervised learning, graph technology and social media data scrapping. Increased effective detection and termination of fraudulent claims by 400%. • Using behavior predictive analytics, the team was able to pinpoint drivers for individual consumer purchasing decision, and effectively guide their actions through this insight. It is projected to increase sales by 22% to 33% in the first year alone for a multi-billion dollar product line.

Build up data science practices to improve services, predict consumer behavior and provide timely insights to senior business, technology and operations stakeholders. Initiatives and PoCs include: NLP analysis of unstructured data, topic modeling and sentiment analysis; supervised classification and unsupervised clustering; time series trending, forecasting and anomaly detection in financial data. ML algorithms used are: NLP, K means clustering, SVR/Decision tree for classification, linear and logistical regression, CNN for text classification, and LSTM for time series prediction.

My role and responsibility are to utilize the Big Data technology and concept to provide a competitive edge for my employer in the Human Capital Management space. Specific interests are: Hadoop and MapReduce, functional language (Scala and JavaScript), Node.js, D3.js, graph database, graph processing engine, graph label propagation, crowd sourcing with built-in bias detection, binary ranking, classification and cluster analysis.

New York, Ny, Us

Los Angeles, Us

Zurich, Ch

Phd, Neuroscience

Bs, Biology