Back to Search
Start Over
Can People Really Do Nothing? Handling Annotation Gaps in ADL Sensor Data
- Source :
- Algorithms, Vol 12, Iss 10, p 217 (2019), Algorithms, Volume 12, Issue 10
- Publication Year :
- 2019
- Publisher :
- MDPI AG, 2019.
-
Abstract
- In supervised Activities of Daily Living (ADL) recognition systems, annotating collected sensor readings is an essential, yet exhaustive, task. Readings are collected from activity-monitoring sensors in a 24/7 manner. The size of the produced dataset is so huge that it is almost impossible for a human annotator to give a certain label to every single instance in the dataset. This results in annotation gaps in the input data to the adopting learning system. The performance of the recognition system is negatively affected by these gaps. In this work, we propose and investigate three different paradigms to handle these gaps. In the first paradigm, the gaps are taken out by dropping all unlabeled readings. A single &ldquo<br />Unknown&rdquo<br />or &ldquo<br />Do-Nothing&rdquo<br />label is given to the unlabeled readings within the operation of the second paradigm. The last paradigm handles these gaps by giving every set of them a unique label identifying the encapsulating certain labels. Also, we propose a semantic preprocessing method of annotation gaps by constructing a hybrid combination of some of these paradigms for further performance improvement. The performance of the proposed three paradigms and their hybrid combination is evaluated using an ADL benchmark dataset containing more than 2.5 &times<br />10 6 sensor readings that had been collected over more than nine months. The evaluation results emphasize the performance contrast under the operation of each paradigm and support a specific gap handling approach for better performance.
- Subjects :
- 0209 industrial biotechnology
lcsh:T55.4-60.8
Computer science
02 engineering and technology
Machine learning
computer.software_genre
lcsh:QA75.5-76.95
Theoretical Computer Science
Task (project management)
Set (abstract data type)
Activity recognition
Annotation
020901 industrial engineering & automation
0202 electrical engineering, electronic engineering, information engineering
Preprocessor
lcsh:Industrial engineering. Management engineering
activity recognition
Hidden Markov model
Numerical Analysis
business.industry
Computational Mathematics
annotated and unannotated data
Computational Theory and Mathematics
smart environments
Benchmark (computing)
020201 artificial intelligence & image processing
Artificial intelligence
hmm
lcsh:Electronic computers. Computer science
Performance improvement
business
computer
Subjects
Details
- Language :
- English
- ISSN :
- 19994893
- Volume :
- 12
- Issue :
- 10
- Database :
- OpenAIRE
- Journal :
- Algorithms
- Accession number :
- edsair.doi.dedup.....4cfefe4d8103ff74b06311b5fdfe36a8