Adaptive mobile activity recognition system with evolving data streams
- Submitting institution
-
Birmingham City University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 11Z_OP_D0002
- Type
- D - Journal article
- DOI
-
10.1016/j.neucom.2014.09.074
- Title of journal
- Neurocomputing
- Article number
- -
- First page
- 304
- Volume
- 150
- Issue
- -
- ISSN
- 0925-2312
- Open access status
- Out of scope for open access requirements
- Month of publication
- -
- Year of publication
- 2014
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
-
- Research group(s)
-
-
- Citation count
- 51
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The paper introduced STAR, a phone-based dynamic framework for activity recognition with evolving data streams. STAR incorporates incremental and active learning for real-time recognition and adaptation in streaming settings. A significant feature of STAR is its ability to refine, enhance and personalise in order to accommodate the natural drift in data streams. STAR influenced many researchers to redirect their effort towards using adaptive models instead of the traditional static models to build more realistic yet accurate models. The experiments demonstrated the superiority of STAR adaptability especially for personalisation. STAR’s low computational cost and real time recognition is another important result.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -