Data Science Project for Beginners
summary
Consulting-focused practical report on how projects actually run: Python, pandas and scikit-learn drive quick exploration, Jupyter hosts demos, models must match client goals. A booth prototype asks 22 taste questions from 113 responses, trains XGBoost for feature importance then collapses to a simple decision tree that dynamically asks fewer questions and surprisingly predicts hazelnut from a dislike of black; emphasizes assembling the right stakeholders, iterative workshops, and that simple models often win.
AI-written summary · about
chapters (17)
- 0:00Intro
- 1:19Where to start
- 2:32Getting started
- 5:08About the project
- 5:56How to start a workshop
- 7:41Dont expect your client to speak with one voice
- 9:15Machine learning and chocolate
- 10:25Take them seriously
- 12:05Data analysis
- 15:22Workflow
- 16:45Predictive Chocolate
- 17:58Client Goals
- 19:19Decision Tree
- 21:35Visualization
- 23:23Takehome message
- 24:42Interview
- 28:50Conclusion
related talks
28mUsing Python in a Data Hackathon
27mTips and tricks for data science projects with Python
04h 02mHands-on Session on Data Science for Beginners
35mMental Models to Use and Avoid as a Data Scientist
48mBecoming a Data Head, a reflection on the trade offs and decision
40mEveryone can Data! - A Practical Data Science Intro
01h 17mOne workshop that data scientists don't want you to attend...
43mEveryone can do Data Science in Python