Skip to content
Open Fraud LabsAcademy
Open for enrolment

Data Science from Scratch

Learn how data becomes decisions: describing and cleaning data, the statistics behind it, and building and explaining your first machine learning models. Examples come from lending, payments and fraud.

Taught by Ayodele Odugbile, data and analytics professional and founder of Open Fraud Labs.

Course outline

Lessons unlock in order. Complete each one by watching the video, reading the notes and passing the quiz.

Data Science from Scratch5 of 30 lessons released
Module 1: Foundations
  1. 01What is data science, really?Released
  2. 02Types of data, and why they matterReleased
  3. 03Your first dataset: rows, columns & featuresReleased
Module 2: Statistics and exploring data
  1. 04Mean, median & mode: describing data with one numberReleased
  2. 05Spread: range, variance & standard deviationReleased
  3. 06Distributions and the normal curveComing soon
  4. 07Outliers: errors, or the most interesting rows?Coming soon
  5. 08Missing data and what to do about itComing soon
  6. 09Data cleaning basicsComing soon
  7. 10Exploratory data analysis (EDA)Coming soon
  8. 11Choosing the right chartComing soon
  9. 12Correlation is not causationComing soon
  10. 13Sampling and sampling biasComing soon
  11. 14Probability basics for data scienceComing soon
Module 3: Tools of the trade
  1. 15Why Python for data scienceComing soon
  2. 16Meet the pandas DataFrameComing soon
  3. 17SQL basics: SELECT, WHERE, GROUP BYComing soon
Module 4: Machine learning essentials
  1. 18Train/test split: why we hide data from the modelComing soon
  2. 19Linear regression intuitionComing soon
  3. 20Classification and logistic regressionComing soon
  4. 21Overfitting vs underfittingComing soon
  5. 22Why accuracy can lieComing soon
  6. 23The confusion matrix, precision & recallComing soon
  7. 24Decision treesComing soon
  8. 25Feature engineeringComing soon
  9. 26Cross-validationComing soon
  10. 27Imbalanced data: the fraud detection problemComing soon
Module 5: Responsible data science and next steps
  1. 28Explainable AI: why did the model decide that?Coming soon
  2. 29Data ethics and bias in modelsComing soon
  3. 30Your data science roadmapComing soon

What you'll be able to do

  • Explain what data science is and how projects run
  • Identify data types and read a dataset's structure
  • Summarise data with averages, spread and distributions
  • Spot outliers, handle missing data and clean datasets
  • Choose the right chart and avoid common traps
  • Train, test and evaluate simple models, including for fraud
  • Explain model decisions and recognise bias

Earn your certificate

Pass all 30 lesson quizzes and get your capstone project approved. Your certificate shows your registered name and an ID anyone can verify.

Certificate of CompletionYour name hereData Science from Scratch

Read the capstone brief