The Laws of Classical Physics Also Apply to Big Data!

on YouTube ↗37m

" summary

Argues that analogies from classical physics can guide Big Data practice and proposes five concrete concepts: set a project start time and spatial/temporal referentials; maintain a central metadata store for provenance and field semantics; measure dataset mass and density and track percentage of tables enriched; compute a valence score per table/field to predict joinability. Recommends keeping sensitive attributes in a protected “noyau” and producing anonymized intermediary datasets, and illustrates the ideas with an A400 meal-planning example while citing Equifax as a failure case; slides are provided for details.

AI-written summary · about

" chapters (27)

1. 0:00Start
2. 0:30The 5 elements of the presentation
3. 0:56Who is Bill Metangmo?
4. 1:32What is physics?
5. 1:43What is classical physics?
6. 2:01The glass of water placed on a table
7. 3:07Big Data: giving you a framework
8. 4:37What is the set formed by the object under study and its environment?
9. 5:40What is the typical Big Data workflow?
10. 6:19What is the reference frame?
11. 7:21What is time?
12. 7:56The largest plane that was never built
13. 11:57Laplace's formula
14. 14:36The open system
15. 15:22Top 4: In a lake
16. 16:07Top 3: Mass
17. 16:48Top 2: Density
18. 17:47Interpretation of density
19. 20:27Microscopic interpretation
20. 21:11Why do we need two hydrogen atoms for one oxygen atom?
21. 21:50Why do molecules need atoms?
22. 24:59Why is a good understanding of this data necessary?
23. 27:34Periodic table of elements
24. 28:05Predicting the behavior of future atoms
25. 29:18Energy is the sum of two parameters
26. 31:11Recap
27. 33:30Man and nature

" related talks

• 38mRole of Statistics in the era of Big DataRajeeva Karandikar · SkillUp 2021
• 15mMapping the Future of Data & AIFrank Theisen · Big-Data.AI Summit 2022
• 01h 31mBig Data: The Computation/Statistics InterfaceMichael Jordan · Big Data Boot Camp 2013
• 19mA world of dataRaghu Ramakrishnan · Collision 2017
• 01h 04mStatistical Physics and Computation in High DimensionFlorent Krzakala, Lenka Zdeborova · Probability, Geometry, and Computation in High Dimensions Boot Camp (Simons Institute, Fall 2020)
• 39mDoing Data: The Critical Process of Data Preparation (...)Ellen Friedman · Berlin Buzzwords 2019
• 36mBig Data, Disruption And The 800 Pound Gorilla In The CornerMichael Stonebraker · Big Data LDN 2018
• 20mCloud Native Patterns for Geospatial Analytics and AILinda Sheard · Spatial Data Science Conference 2022 London