Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Spatio and Temporal Dietary Patterns
Claudia Wagner
GESIS – Leibniz Institut für die Sozialwissenschaften
Motivation
• Who eats what, when and how?
• Health
– Relation between deseases and nutrition

• Sociology
– Social Groups ...
Motivation
• Diet monitoring is expensive
– Consumer panels
– NEMONIT
• One monitoring per year
– NVS
• Laste one: Novembe...
Idea

4
Data
• Most popular online recipe
platform in Austria
• Recipes provided on the
platform
• Server logs (Region – click
– t...
Are the food preferences of
geographically close regions
more similar than those of distant
regions?

8
0.15

0.2

0.20

0.25

0.3

0.30

0.35

0.4

0.40

0.5

0.45

0.50

0.6

Geographic proximity
and food preferences

distan...
To what extent does the weekday
/ season impact users' diet?

11
Temporal patterns

12
Normalized Access
Volume per Weekday

Meat
Carbohydrates

Fish

Vegetable
Alcohol
13
Change Rate
per Weekday
Meat
Carbohydrates

Fish

Vegetable
Alcohol
14
Seasonal Nutrients

15
Seasonal Nutrients

16
Seasonal Nutrients

17
Next Steps
• More data: chefkoch.de and maybe kochbar.de
• Make data available via GESIS’ secure data center?
• Algorithm ...
Upcoming SlideShare
Loading in …5
×

Spatio and Temporal Dietary Patterns

605 views

Published on

Computational Social Science Workshop at GESIS in Cologne, December 2013

Published in: Technology, Health & Medicine
  • Be the first to comment

  • Be the first to like this

Spatio and Temporal Dietary Patterns

  1. 1. Spatio and Temporal Dietary Patterns Claudia Wagner GESIS – Leibniz Institut für die Sozialwissenschaften
  2. 2. Motivation • Who eats what, when and how? • Health – Relation between deseases and nutrition • Sociology – Social Groups differentiate from each other by • What they eat • How they prepare the food • When they eat • Which things they eat together Eva Barlösius, Soziologie des Essens, Kaptitel 6, S. 146, Juventa Verlag 1999 2
  3. 3. Motivation • Diet monitoring is expensive – Consumer panels – NEMONIT • One monitoring per year – NVS • Laste one: November 2005 - Januar 2007 • 20k people • 24h Recall • Diet History Interviews • Weight protocolls 3
  4. 4. Idea 4
  5. 5. Data • Most popular online recipe platform in Austria • Recipes provided on the platform • Server logs (Region – click – timestamp - page) • ~180.000 Rezepte, ~1.700 Regionen • August 2012 – November 2013 5
  6. 6. Are the food preferences of geographically close regions more similar than those of distant regions? 8
  7. 7. 0.15 0.2 0.20 0.25 0.3 0.30 0.35 0.4 0.40 0.5 0.45 0.50 0.6 Geographic proximity and food preferences distant closed Ingredients distant closed Recipes 10
  8. 8. To what extent does the weekday / season impact users' diet? 11
  9. 9. Temporal patterns 12
  10. 10. Normalized Access Volume per Weekday Meat Carbohydrates Fish Vegetable Alcohol 13
  11. 11. Change Rate per Weekday Meat Carbohydrates Fish Vegetable Alcohol 14
  12. 12. Seasonal Nutrients 15
  13. 13. Seasonal Nutrients 16
  14. 14. Seasonal Nutrients 17
  15. 15. Next Steps • More data: chefkoch.de and maybe kochbar.de • Make data available via GESIS’ secure data center? • Algorithm for estimating nutritional values of online recipes • What drives the evolution of online food preferences? – Can we simulate the popularity dynamics? – How does the popularity of ingredients spread across different regions? – Weekday-specific or season-specific spreading? – How do social factors of regions impact the spreading? 18

×