Image-based dietary information mining for community creation in a social network

Authors:
Gamhewage C. de Silva;Kiyoharu Aizawa
Affiliations:
The University of Tokyo, Bunkyo-ku, Japan;The University of Tokyo, Tokyo, Japan
Venue:
Proceedings of second ACM SIGMM workshop on Social media
Year:
2010

Citing 3
Cited 1

Food log by analyzing food images

MM '08 Proceedings of the 16th ACM international conference on Multimedia
FoodLog: capture, analysis and retrieval of personal food images via web

CEA '09 Proceedings of the ACM multimedia 2009 workshop on Multimedia for cooking and eating activities
Motivating contributors in social media networks

WSM '09 Proceedings of the first SIGMM workshop on Social media

Support of self-management for chronic kidney failure patients

Proceedings of the 2011 Visual Information Communication - International Symposium

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present the initial results of an investigation on automated community formation in a social networking site, named FoodLog, that manages dietary information of its users. The site analyzes photos submitted by users, to estimate the dietary composition of their daily meals. In order to automatically identify natural groupings among meals consumed by different users, we apply simple expectation maximization algorithm on a selected data set from this site. Visual observation of images from the resulting clusters proves that the groupings actually correspond to different categories of meals. We demonstrate how communities of users can be formed by assigning them to clusters. We also discuss the steps that should follow this initial study, and possible future directions.