• Choice modeling• Trade-off modeling • Python and R programming

• Made use of Dataiku ETL platform and Python to make a recommender to improve the marketing of electronics components.• Clustering of customers into distinct groups• Using ETL on large datasets and making Machine Learning projects on Snowflake and Azure on the Dataiku platform.• Time series in Python to make queue projections on the Dataiku platform.• Use of SQL language.

• Time Series modeling using sklearn that predicts Train Ridership at better than 81 percent accuracy during the volatile COVID period• Use of Python sklearn to predict the probability of a particular train to be on-time• Implemented sentiment analysis of text using NLP in Python and Keras with 98% accuracy• Comment categorization using NLP in Python and Keras with 98% accuracy• Use of non-parametric classification algorithms (Decision Trees, Association Rules, and Naïve Bayes, etc.) and parametric classification algorithms (Logistic Regression, Support Vector Machine, etc.)• Investigation of Train Lateness using Association Rules that show patterns of late trains in the system• Oracle SQL • Automation of content creation of websites using Python• Programming in R and Python

Development of an User Interface using RShiny.

• Statistical modeling predicting Repeat Donors for Non-Profit• Extensive SQL queries on Cloud-based Data Warehouse• Worked with Salesforce Databases• Programming in R and Python

• Teamed with DJS Consulting for modeling of survey data for several Healthcare clients• Produced excel-based classifiers that identified favorable market segments• Data wrangling in R and Python to prepare for Data analysis• Produced statistical models for brand and drug affinity• Learned Tableau Visualization Tool

• Executed analytics for over 100 studies in survey research, about 50% healthcare clients.• Machine Learning over 100 customer segmentation studies (unsupervised learning). These projects usually included an Excel-based classifier tool that indicated the predicted type of client. These segmentations covered many different types of businesses: Technology firms, automotive companies, healthcare, customer goods, B2B and many others.• Placed special emphasis on analyses in Segmentation and in Customer Loyalty/Retention and Reputation. Achieved a high rate of return business.• Used association rules for a Grocery store study that analyzed demographics and customer visit patterns. The study was lauded for its analysis.• Mentored the newest member of our 5-member Advanced Analytics team.• Knowledgeable about survey research and applying weighting to analysis to make it representative to a population• Performed a large study for a major car manufacturer with the intention to increase market share of all their brands.• Logistic modeling, Linear modeling, Random Forest, SVM, Neural Networks, Dimensionality Reduction, Clustering, Probabilistic modeling, Bayesian modeling using MCMC techniques.• Expert programmer in SAS, and R. Knowledge of Python programming, SPSS, and excel VBA.

• Executed analytics for over 100 studies in survey research, about 50% healthcare clients.• Machine Learning over 100 customer segmentation studies (unsupervised learning). These projects usually included an Excel-based classifier tool that indicated the predicted type of client. These segmentations covered many different types of businesses: Technology firms, automotive companies, healthcare, customer goods, B2B and many others.• Placed special emphasis on analyses in Segmentation and in Customer Loyalty/Retention and Reputation. Achieved a high rate of return business.• Used association rules for a Grocery store study that analyzed demographics and customer visit patterns. The study was lauded for its analysis.• Mentored the newest member of our 5-member Advanced Analytics team.• Knowledgeable about survey research and applying weighting to analysis to make it representative to a population• Performed a large study for a major car manufacturer with the intention to increase market share of all their brands.• Logistic modeling, Linear modeling, Random Forest, SVM, Neural Networks, Dimensionality Reduction, Clustering, Probabilistic modeling, Bayesian modeling using MCMC techniques.• Expert programmer in SAS, and R. Knowledge of Python programming, SPSS, and excel VBA.

Lecturer in Business, Law, and Computer Sciences Faculties

Research into the understanding of student loan markets. Areas of investigation include modeling probability of borrower default and market analysis of student loans across the US. This required SAS programming, understanding of statistics, and manipulation of multiple datasets between PC servers and mainframe.