Rishabh Arora Email & Phone Number

New Brunswick, NJ, US

Driving product recommendation and media optimization using data and ML; develop statistical & ML models to take data-driven decisions via media measurement and personalized marketing.- Next Best Action Recommender Engine integration with software platforms (Azure SQL DB & AWS EC2)- In-market A/B experiment - Diff-in-Diff, Bonferroni Correction (Python).

Raleigh, North Carolina, US

• Integrated end-to-end Machine Learning workflow - build, train & deploy in cloud platform AWS using SageMaker and EC2 servers; Leveraged in-built SageMaker ML libraries, hyper parameter tuning jobs for creating models.
• Derived significant features leading to COVID diagnosis using machine learning models: Logistic Regression, Random Forest & XGBoost. Leveraged k-prototype clustering for unstructured segmentation to define different.
• Published an article that demonstrated the bi-directional relationship between COVID-19 and Substance Use Disorder using machine learningPublication: https://www.intechopen.com/chapters/78693

Raleigh, North Carolina, US

• Spearheaded a project to find the most relevant physicians via analyzing claims data which resulted in $250K+ savings; Extracted huge dataset from 10M+ records and evaluated key metrics using descriptive statistics
• Built supervised machine learning models (Random Forest, XGBoost, Adaptive Boosting, and SVM) to predict cancer patients and derive top features for creating a diagnostic test. Created data structure, executed data.
• Implemented Upsampling, Downsampling & SMOTE to handle imbalanced dataset, and feature selection using Boruta, LightGBM, and Recursive Feature Elimination; Created Ensemble models which boosted model efficiency by 5~6%
• Developed a metric system to predict early diagnosis via analyzing real-life claims data with 92M+ records; Derived top variables contributing to diagnosis by performing Two-Sample t-Test on test & control groups.
• Automated data retrieval processes & deployed code which proved to be effective in reducing the operational turnaround time by ~5 hours for each data pull request; Executed 1K+ complex ad-hoc SQL queries to support.

Raleigh, North Carolina, US

• Profited pharmaceutical clients with $2 million via predictive analytics measures to increase drug market share
• Revamped machine learning models in Python to predict early diagnosis and target the physicians; Extracted data from 50M+ records in SQL; Implemented stratified sampling and propensity scoring matching for robust.
• Developed a tool in R Studio for feature engineering and used Lasso, Ridge regression and Elastic Net for tuning the machine learning models; Reduced variable set from 5K+ to less than 300 significant variables

Philadelphia, PA, US

• Performed predictive modeling using linear regression, k-means clustering in R Studio and time series forecast on telematics data to find insights.
• Suggested measures, how telematics can help to reduce idle time and optimize routes. The insights will help the company to save $5M if data is leveraged strategically.

Hauz Khas, Delhi, IN

• Saved state’s million-dollar from being invested in trivial activities by leveraging data driven insights from a research; using advanced analytics, identified potential candidates of a political party who are likely.
• Monitored timely survey-data collection & designed analysis plan for data structure; Extracted real-time data in SQL from database (million records) and used multivariate linear regression to identify the significant.

• Established new business opportunities by analyzing industry trends and market share to maintain and develop new client base; Tracked & recorded metrics, data, statistics on company’s performance, client retention and.

New York, NY, US

• Member of the global team, Center of Excellence (COE), expert in analyzing Coca-Cola survey responses. Produced reports to answer business questions: demographic section, category incidence, consumption pattern and.
• Extracted, with Q software, consumer behavior data relevant to the client’s resulting concerns. The study helped to discover consumer’s choice and consumption pattern, and thus, contributed in scaling up their customer.

London, GB

• Responsible for project execution and other essential activities including data analysis, data extraction and reporting. Researched on analyzing market trends that gave a brief idea on how other competitors are.
• Collaborated with reporting team to analyze quantitative data. Used data insights to measure advertising effectiveness. The end-result identified the potential advertisement contributing most towards brand recall.

Paris, FR

• Implemented A/B testing for advertisement launch to understand better performing ad, potentially capture higher audience; Made analysis plan for data structure and created PowerPoint presentations to answer business.
• Analyzed quantitative and qualitative data to find insights. It helped in decision making for effectively making the ad for their product

Master Of Science, Business Statistics, Data Science

Bachelor Of Technology, Information Technology