Conceptualizing, developing, and evaluating machine learning models based on DayTwo’s proprietary gut microbiome database and Electronic Medical Records (EMR), to model relationships between the microbiome and medical conditions, response to drugs, as well as bloodwork parameters.

Developed machine learning models by employing a dataset of hundreds of millions of medical claims records. Sample projects included recommendation of medical providers based on the patient’s medical history, prediction of hospital length of stay, and prediction of medical visits due to extreme weather or pollution.

Developed machine learning models to predict clinical deterioration in hospitals and other healthcare settings. Employed data from a continuous-monitoring sensor, as well as EMRs from hospitals and long-term hospitalization facilities, to engineer features and develop an ensemble of machine learning models to predict ICU transfers and death.

Developed innovative analytical solutions through the use of machine learning, natural language processing, and big data tools applied to structured, semi-structured, and unstructured data.- Used machine learning and NLP to classify billions of tweets and meta-data of over a million YouTube videos.- Used Apache Spark and ElasticSearch to infer locations and themes in tens of billions of tweets.- Applied clustering techniques to hundreds of megabytes of semi-structured log data with dozens of features.

Identified patterns and formed predictions about online user engagement on the largest citizen science platform in the world, through the use of machine learning and statistical analysis.- Used clustering methods to group user session log data, in order to derive different types of user behavior modes.- Used machine learning and NLP to classify forum messages in order to predict message helpfulness to others.

– Taught a Master's level online course in applied data science.

Performed a series of research support tasks including automation of text pre-processing, evaluation of NLP packages for entity extraction, and assessment of NLP tools for unobtrusive monitoring of spaceflight team functioning (NASA funded). Coached Executive Doctorate students in research methods and statistics.

Developed an ontological content-based personalization algorithm (conceptualization, implementation, and evaluation) as part of a team working on an electronic newspaper project.

Led sections in the following courses:- Introduction to Information Systems- Data Mining and Data Warehouse

Developed decision support tools by utilizing business intelligence, operations research, and data analysis. Performed information systems analysis & design to support intelligence activities as well as process improvement.