Breaking the 80/20 Dilemma in Data Science Projects
summary
The talk analyzes the ‘80/20 dilemma’ in data science where most project time is spent on data discovery, collection, cleaning, and encoding business rules rather than analytics. It presents concrete cases from Fligoo and the technologies they used to address these problems, and outlines current trends, tips and methodologies to optimize effort for data scientists, data engineers and project managers.
AI-written summary · about
chapters (10)
- 0:0080/20 dilemma in data science
- 1:5580/20 problem explained
- 3:19Triple constraint in projects
- 6:09Process optimization
- 10:03Real cases: Champions & Netflix
- 17:25Proof of concept in 3 months
- 24:51Fast production development
- 28:45Recommendation optimization with linear programming
- 31:19Future trends in DataOps governance
- 35:13Breaking the 80/20 dilemma
related talks
32mWhy Most Data Projects Fail and How to Avoid It
31mWhy Most Data Projects Fail and How to Avoid It
24mA Practical Approach To Unlock Value From Data and Analytics
26mData For The 99%: Unlocking Data Curiosity Across Your Organisation
31mWhy Most Data Projects Fail and How to Avoid It
22mBridging Business and Tech for Data Science Success
30mDevOps For Data Teams
53mThe State of Data Engineering