DataRobot University provides self-paced content and labs to help you learn how to use DataRobot to solve data science business problems. Lessons related to the labs walk you through the problem-solving workflow in DataRobot.
The following describes each lab and the related lessons.
Fuel Efficiency Lab—In this lab, you will predict the fuel efficiency of a new car that has not yet been designed. This is a regression problem, so you will learn how to frame, set up, evaluate, and interpret predictions for a continuous target.
Lessons: Regression, What-if App, Model Evaluation, Insights
Salary Prediction Lab—In this lab, you'll use machine learning to determine how different survey responses predict developer salaries. Think of this in the context of a Human Resources department determining the salary of an individual based on the experience needed for the position.
Lessons:Regression, Messy Data (Lots of Categoricals and Text), Optimization Metrics
COVID Lab—In this lab, you'll use look-ahead modeling to classify US counties that are at high risk for developing COVID-19.
Medical Fraud Lab—In this lab, you will predict fraud using a medical claims dataset. (You can find a detailed description of this use case in DataRobot Pathfinder.) To achieve this, you will approach this as a binary classification problem and optimize the profitability by adjusting the prediction threshold.
Lead Scoring Lab—In this lab, you'll use lead scoring to predict the probability that a prospect will become a customer. To achieve this, we are going to use binary classification.
Lessons:Classification, Lead Scoring, Learning Curves and Dataset Size, Prediction Explanation Excel Analysis
HR Bias Lab—In this lab, you will use DataRobot to learn how ethical issues can impact your machine learning projects. We are going to do that by building an ML model that will eventually be free of gender bias.
Lessons:AI Ethics, Classification, HR Use cases
Churn I Lab—In this lab, you'll learn a basic approach to churn modeling using many of the tools available in DataRobot.
Lessons:Classification, Churn, Unit of Analysis, Aggregating Data, Confusion Matrix, Thresholding, Profit Curve
Churn II Lab—In this lab, you'll learn additional techniques to aggregate and join data for your churn model.
Lessons:Classification, Churn, Unit of Analysis, Aggregating Data, Feature Engineering in Excel
Credit Card Fraud Lab—In this lab, you'll learn to use Frequency-Severity modeling to classify fraudulent credit card transactions. To achieve this, we are going to use a regression target.
Beyond Autopilot Lab—In this lab, you will learn about the different modeling families available in DataRobot, and determine the best model to use and improve (if necessary) before you deploy it.
Time Series Lab—On this mission, you add more and more sophisticated modeling such as Forecast Distance, Feature Derivation Windows, dealing with regime changes, and using calendar files to build increasingly sophisticated models.
Lessons: Forecasting Unemployment, Dealing with Regime Changes, Forecasting Sales Across Multiple Stores
Improving Time Series Anomaly Detection Models Lab—In this lab, you learn and apply strategies for improving the accuracy of your anomaly detection models, including Feature Derivation Windows (FDWs), repository blueprints, additional feature lists, advanced tuning, and blenders.