Paper Info
Title
Combining crowd-generated media and personal data: semi-supervised learning for context recognition
Abstract
The growing ubiquity of sensors in mobile phones has opened many opportunities for personal daily activity sensing. Most context recognition systems require a cumbersome preparation by collecting and manually annotating training examples. Recently, mining online crowd-generated repositories for free annotated training data has been proposed to build context models. A crowd-generated dataset can capture a large variety both in terms of class number and in intra-class diversity, but may not cover all user-specific contexts. Thus, performance is often significantly worse than that of user-centric training. In this work, we exploit for the first time the combination of both crowd-generated audio dataset available in the web and unlabeled audio data obtained from users' mobile phones. We use a semi-supervised Gaussian mixture model to combine labeled data from the crowd-generated database and unlabeled personal recording data. Hereby we refine generic knowledge with data from the user to train a personalized model. This technique has been tested on 7 users on mobile phones with a total data of 14 days and up to 9 context classes. Preliminary results show that a semi-supervised model can improve the recognition accuracy up to 21%.
Year
DOI
Venue
2013
10.1145/2509352.2509396
PDM@ACM Multimedia
Keywords
Field
DocType
context class,unlabeled personal recording data,personal data,free annotated training data,annotating training example,total data,semi-supervised learning,context recognition,crowd-generated dataset,online crowd-generated repository,unlabeled audio data,crowd-generated media,mobile phone,crowd-generated database,activities of daily living,semi supervised learning
Training set,Semi-supervised learning,Class number,Computer science,Exploit,Artificial intelligence,Mobile phone,Labeled data,Multimedia,Mixture model,Machine learning
Conference
Citations 
PageRank 
References 
6
0.45
6
Authors
4
Name
Order
Citations
PageRank
Long-Van Nguyen-Dinh1955.78
Mirco Rossi224014.02
Ulf Blanke369936.03
Gerhard Tröster42493250.70