MLDS 2019: Pradeep Gulipalli of Tiger Analytics at Machine Learning Developers Summit 2019
summary
States that 20% causes drive ~80% effects and gives concrete counts: today 1% own 50% of wealth, 10% own ~85%; examples: wine retailer sees 20% SKUs → 60% revenue, railway 30% routes → 80% trips, beverage 10% flavors → 95% sales, and a credit-card ad campaign where <1% clickers convert. Proposes data-science solutions: predict future “macro Pareto” winners (open-source projects, new product flavors), prioritize via a value vs complexity matrix, triage data as must-have/good-to-have/optional, reuse standard analysis modules and stop when the initial ~20% of modeling yields most value.
AI-written summary · about
chapters (6)
- 0:00Start
- 7:22The Pareto Principle
- 7:51Pareto Principle
- 19:41Project Selection
- 26:21Analysis and Modeling
- 27:57Model Building Phase
related talks
32mMaking Models Work: How an optimized ML lifecycle can drive better ROI
07mPredictive Modeling and Machine Learning
27mUnlocking Product Growth With Big Data, Data Science And Machine Learning
24mPost-Digital Shift: Exponential Business Models for Precision Decisions at Scale
36mML: Predictive analytics in Ad Engineering
45mData Science 2.0: From Outputs to Outcomes
23mThe unreasonable effectiveness of data for better monetisation
23mValue in Data Science Beyond Models in Production