Food is a fundamental concept in our daily lives and is one of the most important factors that shape how healthy we are or how good we feel. Although research on the users’ food preferences has been a well-established research area over the last decades, only very little research was devoted yet to understand, how the World Wide Web influences the way we consume or produce food offline.
In this talk, I will therefore highlight recent research in online food communities and interesting findings in terms of online food recipe consumption and production patterns. I will show, how these studies might be useful when drawing conclusions about health related issues, such as obesity or diabetes of a large population, and how these insights might be used to tune current recommender approaches in this domain.
Last but not least, I will discuss the limitations of these studies and will highlight the need for a joined European taskforce, that studies food, health related issues and recommender systems on a much larger and more useful way, as proposed by current research in this area.
Studying Online Food Consumption and Production Patterns: Recent Trends and Challenges
1. 1
. Christoph Trattner 23.10.2015 – Bolzano
Studying Online Food Consumption and Production
Patterns: Recent Trends and Challenges
Christoph Trattner
Know-Center
@Know-Center, TUG, Austria
2. 2
. Christoph Trattner 23.10.2015 – Bolzano
Outline
Short background info
Why is the topic important
Research on Consumption Patterns
Research on Production Patterns
Cultural Differences
Challenges
Funding opportunities – H2020
5. 5
. Christoph Trattner 23.10.2015 – Bolzano
Academic Background & Working Exp.
Started in 2004 with my studies at TUG - finished in 2008
In 2009 I started with my PhD at TUG - finished in Oct.
2012
After that I worked at the Know-Center (Research Center
for Big Data Analytics ) until Sept. 2014
From Oct 2014 until Sept. 2015 I worked as Marie Curie
Alain Bensoussan Fellow at NTNU
Research visits to Yahoo! Labs & CWI
Since 1st of Oct. 2015 back to Know-Center
7. 7
. Christoph Trattner 23.10.2015 – Bolzano
Importance (1)
Food is one the main concepts that shapes how
good we feel and how healthy we are
According to the WHO, if common lifestyle risk
factors, among others diet-related ones, were
eliminated, around 80% of cases of heart disease,
strokes and type 2 diabetes, and 40% of cancers,
could be avoided (European Comission
Recommendation C(2010) 2587 final, 2010).
8. 8
. Christoph Trattner 23.10.2015 – Bolzano
Importance (2)
According to the WHO, within the last three decades
overweight and obesity in the EU population rised
dramatically > 30% (especially for the younger
generation)
Resulting in a cost of approx. € 81 billion a year to
help people with chronic diseases
9. 9
. Christoph Trattner 23.10.2015 – Bolzano
Studies on Food Consumption Patterns
on the Web
10. 10
. Christoph Trattner 23.10.2015 – Bolzano
West, R., White, R. W., & Horvitz, E. (2013, May). From
cookies to cooks: Insights on dietary patterns via
analysis of web usage logs. In Proceedings of the
22nd international conference on World Wide Web
(pp. 1399-1410). International World Wide Web
Conferences Steering Committee.
14. 14
. Christoph Trattner 23.10.2015 – Bolzano
Abbar, S., Mejova, Y., & Weber, I. (2015). You tweet
what you eat: Studying food consumption through
twitter. ACM CHI 2015.
15. 15
. Christoph Trattner 23.10.2015 – Bolzano
Correlation between food mentions on
Twitter & Obese
p=.772
s=.784
Abbar, S., Mejova, Y., & Weber, I. (2015). You tweet what you eat: Studying food consumption through twitter. ACM CHI 2015.
http://www.caloriecount.com/
50 million tweets
Food related keywords
19. 19
. Christoph Trattner 23.10.2015 – Bolzano
T. Kusmierczyk, C. Trattner, K. Nørvåg: Temporal
Patterns in Online Food Innovation. WWW
Companion 2015: 1345-1350.
T. Kusmierczyk, C. Trattner, K. Nørvåg: Temporality in
Online Food Recipe Consumption and Production.
WWW Companion 2015: 55-56.
T. Kusmierczyk, C. Trattner, K. Nørvåg: Understanding
and Predicting Recipe Uploads in online food
communities. under review.
21. 21
. Christoph Trattner 23.10.2015 – Bolzano
constant entropy of
ingredients
continuous growth of
ingredients combinations
complexity
consequence:
H(combination | ingredients)
grows
21
T. Kusmierczyk, C. Trattner, K. Nørvåg: Temporal Patterns in Online Food Innovation. WWW Companion 2015: 1345-1350
Community Evolution
22. 22
. Christoph Trattner 23.10.2015 – Bolzano
Innovation (1)
Two phases:
1.strong decline
1.slow but steady
increase
22
2010
T. Kusmierczyk, C. Trattner, K. Nørvåg: Temporal Patterns in Online Food Innovation. WWW Companion 2015: 1345-1350
recipe r similarity to other all
recipes r’
23. 23
. Christoph Trattner 23.10.2015 – Bolzano
Innovation (2)
interesting outliers
23
T. Kusmierczyk, C. Trattner, K. Nørvåg: Temporal Patterns in Online Food Innovation. WWW Companion 2015: 1345-1350
25. 25
. Christoph Trattner 23.10.2015 – Bolzano
Temporal Patterns
25
T. Kusmierczyk, C. Trattner, K. Nørvåg: Temporality in Online Food Recipe Consumption and Production. WWW Companion 2015: 55-56
26. 26
. Christoph Trattner 23.10.2015 – Bolzano
Lifetime of a Recipe
26
T. Kusmierczyk, C. Trattner, K. Nørvåg: Temporality in Online Food Recipe Consumption and Production. WWW Companion 2015: 55-56
35. 35
. Christoph Trattner 23.10.2015 – Bolzano
Ahn, Y. Y., Ahnert, S. E., Bagrow, J. P., & Barabási, A.
L. (2011). Flavor network and the principles of food
pairing. Scientific reports, 1.
Laufer, P., Wagner, C., Flöck, F., & Strohmaier, M.
(2015). Mining cross-cultural relations from
Wikipedia-A study of 31 European food cultures.
ACM WebSci.
39. 39
. Christoph Trattner 23.10.2015 – Bolzano
Cuisines as perceived by countries in Wikipedia
40. 40
. Christoph Trattner 23.10.2015 – Bolzano
And how is the progress in recommender
research?
41. 41
. Christoph Trattner 23.10.2015 – Bolzano
Teng, C. Y., Lin, Y. R., & Adamic, L. A. (2012, June).
Recipe recommendation using ingredient networks.
In Proceedings of the 4th Annual ACM Web Science
Conference (pp. 298-307). ACM.
43. 43
. Christoph Trattner 23.10.2015 – Bolzano
Elsweiler, D., & Harvey, M. (2015, September).
Towards Automatic Meal Plan Recommendations for
Balanced Nutrition. In Proceedings of the 9th ACM
Conference on Recommender Systems (pp. 313-
316). ACM
44. 44
. Christoph Trattner 23.10.2015 – Bolzano
Small user study (100 users over 3 years)
Goal: predict rating of users according to eating
guidelines
Personas: age, gender, hight, goal,...
Findings: In general possible but also not so easy
task
Hard Profiles: some users tend to only rate highly calorific and fatty
recipes
very few breakfasts rates
recipes with a lower diversity of ingredients
number of recipes they have rated is low
46. 46
. Christoph Trattner 23.10.2015 – Bolzano
WebScience Research
Recommender Research
Nutrition (Food) Research
Survey based (small scale & expensive) – 100s of papers dealing with
issues related to food & health related issues
Data driven, large scale offline studies, cheap, fast –
hardly any research yet – no evidence of correlation & causation
Mostly data driven, small to large scale offline „studies“, cheap,
Fast – not much evidence yet for usefulness
47. 47
. Christoph Trattner 23.10.2015 – Bolzano
Is there also funding for this kind of
research?
48. 48
. Christoph Trattner 23.10.2015 – Bolzano
H2020:
Food Scanner Challenge
Web:
http://ec.europa.eu/research/horizonprize/index.c
m?prize=food-scanner
Social Media:
https://twitter.com/EU_eHealth
Video:
https://www.youtube.com/watch?v=v0uggsj4Ars
49. 49
. Christoph Trattner 23.10.2015 – Bolzano
H2020 - WPs
Health, demographic change and well-being
SC1-PM-15-2017: Personalised coaching for well-being and care of
people as they age
SC1-PM-17–2017: Personalised computer models and in-silico
systems for well-being
SC1-PM-05–2016: The European Human Biomonitoring Initiative
Information and Communication Technologies
ICT-11-2017: Collective Awareness Platforms for Sustainability and
Social Innovation
ICT-19-2017: Media and content convergence
Food security, sustainable agriculture and forestry,
marine and maritime and inland water research and
the bioeconomy
50. 50
. Christoph Trattner 23.10.2015 – Bolzano
The Sustainable Food Security call will address resilience
and resource efficiency in the primary sectors (agriculture, forestry,
fisheries and aquaculture) and in the related up- and downstream
industries to ensure the food and nutritional security of EU citizens.
Investments in innovation will support stability and competitiveness of
the agri-food chains, such as the food industry, the largest EU
manufacturing industry. This call will also help to safeguard and make
efficient use of the natural capital as the basis of primary sectors, while
factoring in climate and environmental challenges. Finally, the
call will explore innovative approaches in the
food value chain to empower citizens to change
towards sustainable and healthy food
consumption patterns and lifestyles.
51. 51
. Christoph Trattner 23.10.2015 – Bolzano
H2020 - WPs
FET Open supports the early-stages of the science
and technology research
FET Proactive addresses promising directions for
research to build up a European critical mass of
knowledge and excellence around them.
FET Flagships are science-driven, large-scale,
multidisciplinary research initiatives
Opening: 08 Dec. 2015
Budget: 84.00 (2016)
Deadline: 11 May 2016
http://ec.europa.eu/research/participants/data/ref/h2020/wp/2016_2017/main/h2020-wp1617-fet_en.pdf
52. 52
. Christoph Trattner 23.10.2015 – Bolzano
People/Institutions interested
L3S Research Center, Germany
PUC, Chile
NTNU, Norway
CWI, The Netherlands
University of Tallinn, Estonia
GESIS, Germany
Yahoo Labs!, UK
University of Bolzano, Italy
Graz University of Technology, Austria
MedUni Graz, Austria
University of Regensburg, Germany
Qatar University, Qatar
53. 53
. Christoph Trattner 23.10.2015 – Bolzano
Thank you!
Christoph Trattner
Email: trattner.christoph@gmail.com
Web: christophtrattner.info
Twitter: @ctrattner
54. 54
. Christoph Trattner 23.10.2015 – Bolzano
References
Kusmierczyk, T., Trattner, C., & Nørvåg, K. (2015, May). Temporality in online food recipe consumption and production. In
Proceedings of the 24th International Conference on World Wide Web Companion (pp. 55-56). International World Wide Web
Conferences Steering Committee.
Kusmierczyk, T., Trattner, C., & Nørvåg, K. (2015, May). Temporal Patterns in Online Food Innovation. In Proceedings of the 24th
International Conference on World Wide Web Companion (pp. 1345-1350). International World Wide Web Conferences Steering
Committee.
Wagner, C., Singer, P., & Strohmaier, M. (2014). The nature and evolution of online food preferences. EPJ Data Science, 3(1), 1-
22.
Laufer, P., Wagner, C., Flöck, F., & Strohmaier, M. (2015). Mining cross-cultural relations from Wikipedia-A study of 31 European
food cultures. ACM WebSci.
Rokicki, M., Herder, E., & Demidova, E. (2015). What’s On My Plate: Towards Recommending Recipe Variations for Diabetes
Patients. Extended proc. user modeling, adaptation and personalizationumap 2015.
Elsweiler, D., & Harvey, M. (2015, September). Towards Automatic Meal Plan Recommendations for Balanced Nutrition. In
Proceedings of the 9th ACM Conference on Recommender Systems (pp. 313-316). ACM.
Said, A., & Bellogín, A. (2014). You are what you eat! tracking health through recipe interactions. Proc. of RSWeb, 14.
Abbar, S., Mejova, Y., & Weber, I. (2014). You tweet what you eat: Studying food consumption through twitter. arXiv preprint
arXiv:1412.4361.
Mejova, Y., Haddadi, H., Noulas, A., & Weber, I. (2015, May). # FoodPorn: Obesity Patterns in Culinary Interactions. In
Proceedings of the 5th International Conference on Digital Health 2015 (pp. 51-58). ACM.
Teng, C. Y., Lin, Y. R., & Adamic, L. A. (2012, June). Recipe recommendation using ingredient networks. In Proceedings of the
4th Annual ACM Web Science Conference (pp. 298-307). ACM.
Ge, M., Ricci, F., & Massimo, D. (2015, September). Health-aware Food Recommender System. In Proceedings of the 9th ACM
Conference on Recommender Systems (pp. 333-334). ACM.
Ahn, Y. Y., Ahnert, S. E., Bagrow, J. P., & Barabási, A. L. (2011). Flavor network and the principles of food pairing. Scientific
reports, 1.
Freyne, J., & Berkovsky, S. (2010, February). Intelligent food planning: personalized recipe recommendation. In Proceedings of
the 15th international conference on Intelligent user interfaces (pp. 321-324). ACM.
Elahi, M., Ge, M., Ricci, F., Berkovsky, S., & David, M. (2015) Interaction Design in a Mobile Food Recommender System.