Thinking Like a Data Scientist
summary
Explains why simple exploration and visualization solve problems better than blind ML: John Snow maps cholera cases, ties deaths to one contaminated well and removes the pump handle; brewery workers survive because boiled beer water. Describes building a train-sensor pipeline that uses GPS + timestamp -> derived acceleration and adds weather as a feature, uses PCA to drop useless dimensions, and argues data scientists must ship APIs, infra as code, and negotiate organizational and ethical constraints.
AI-written summary · about
chapters (11)
- 0:00Start
- 0:14Thinking Like A Data Scientist
- 2:01What Does a Data Scientist Do
- 4:26Controversial opinions about data scientists
- 5:42So how do data scientists actually think?
- 5:56Cholera Facts (yay!)
- 9:43The Broad Street Cholera Outbreak of 1854
- 10:10Formally write your hypothesis
- 18:18So what are the lessons?
- 19:04Data Exploration: Refining your mental model
- 19:34Let's talk about models
related talks
32mWhy every developer should be a data scientist
47mA Practical(ish) Introduction to Data Science
51mFrom Data Confusion to Data Intelligence
20mDon't Look. See! Are We Blinded by Data (Visualization)?
44mBringing Data to L I F E and Life to Data
40mVisual Analysis for Data Science
38mThe Real World of Data Analytics
33mBuilding Data Science Tools for Sustainable Transformation