Leveraging Active Learning and Conditional Mutual Information to Minimize Data Annotation in Human Activity Recognition

2019 ◽  
Vol 3 (3) ◽  
pp. 1-23 ◽  
Author(s):  
Rebecca Adaimi ◽  
Edison Thomaz
Keyword(s):  
Active Learning ◽  
Mutual Information ◽  
Activity Recognition ◽  
Human Activity ◽  
Human Activity Recognition ◽  
Conditional Mutual Information ◽  
Data Annotation

Human Activity Recognition Based on Dynamic Active Learning

2020 ◽  
pp. 1-1
Author(s):  
Haixia Bi ◽  
Miquel Perello-Nieto ◽  
Raul Santos-Rodriguez ◽  
Peter Flach
Keyword(s):  
Active Learning ◽  
Activity Recognition ◽  
Human Activity ◽  
Human Activity Recognition

scholarly journals A Semi-Automatic Annotation Approach for Human Activity Recognition

Sensors ◽  
2019 ◽  
Vol 19 (3) ◽  
pp. 501 ◽  
Author(s):  
Patrícia Bota ◽  
Joana Silva ◽  
Duarte Folgado ◽  
Hugo Gamboa
Keyword(s):  
Activity Recognition ◽  
Human Activity ◽  
High Performance ◽  
Model Performance ◽  
Human Activity Recognition ◽  
Embedded Sensors ◽  
Data Annotation ◽  
Indoor Location ◽  
Annotation Process ◽  
Comprehensive Study

Modern smartphones and wearables often contain multiple embedded sensors which generate significant amounts of data. This information can be used for body monitoring-based areas such as healthcare, indoor location, user-adaptive recommendations and transportation. The development of Human Activity Recognition (HAR) algorithms involves the collection of a large amount of labelled data which should be annotated by an expert. However, the data annotation process on large datasets is expensive, time consuming and difficult to obtain. The development of a HAR approach which requires low annotation effort and still maintains adequate performance is a relevant challenge. We introduce a Semi-Supervised Active Learning (SSAL) based on Self-Training (ST) approach for Human Activity Recognition to partially automate the annotation process, reducing the annotation effort and the required volume of annotated data to obtain a high performance classifier. Our approach uses a criterion to select the most relevant samples for annotation by the expert and propagate their label to the most confident samples. We present a comprehensive study comparing supervised and unsupervised methods with our approach on two datasets composed of daily living activities. The results showed that it is possible to reduce the required annotated data by more than 89% while still maintaining an accurate model performance.

scholarly journals Online active learning for human activity recognition from sensory data streams

2020 ◽  
Vol 390 ◽  
pp. 341-358 ◽  
Author(s):  
Saad Mohamad ◽  
Moamar Sayed-Mouchaweh ◽  
Abdelhamid Bouchachia
Keyword(s):  
Active Learning ◽  
Activity Recognition ◽  
Human Activity ◽  
Data Streams ◽  
Human Activity Recognition ◽  
Sensory Data
Sign in / Sign up

Export Citation Format

Share Document