How to Overcome the Apathy of Big Data
summary
Claim: start with a goal and cut noise to the handful of variables that actually move outcomes. Demonstrations include a tweet classifier built in Ruby using the t gem and a nimbus gem random forest that runs, takes minutes, and returns a classification tree with variable importance showing a few specific accounts dominating retweet probability. Visualization tricks include Chernoff faces via a Shoes app. Cited result: a study compresses 310 clinical variables down to 10, turning big data into actionable metrics.
AI-written summary · about
chapters (11)
- 0:00Big data apathy intro
- 2:50Big data distraction industry
- 4:1180 20 data principle
- 7:00Goal-focused decisions
- 8:08AARRR metrics for clarity
- 11:15Data beyond 3D visualization
- 16:55Chernoff faces for sentiment
- 19:23Filtering data variables
- 20:06Classification trees for selection
- 26:55Retweet prediction variables
- 29:41Actionable data insights
related talks
15mUnderstanding Decisions Driven by Big Data
35mIntelligent Ruby: Getting Started with Machine Learning
33mData Science in Healthcare: Beyond the Hype
01h 26mData Driven Action: a primer on Data Science
38mDo's and don'ts of big data analytics
35mMental Models to Use and Avoid as a Data Scientist
36mWhen the Only Way to Scale up, is to Scale Down
34mData Science beasts (failures) and where to find them