An analysis of machine- and human-analytics in classification
- Submitting institution
-
University of Oxford
- Unit of assessment
- 12 - Engineering
- Output identifier
- 9448
- Type
- D - Journal article
- DOI
-
10.1109/TVCG.2016.2598829
- Title of journal
- IEEE Transactions on Visualization and Computer Graphics
- Article number
- -
- First page
- 71
- Volume
- 23
- Issue
- 1
- ISSN
- 1077-2626
- Open access status
- Compliant
- Month of publication
- August
- Year of publication
- 2016
- URL
-
-
- Supplementary information
-
https://ieeexplore.ieee.org/ielx7/2945/7747554/7539314/tvcg-tam-2598829-mm.zip?tp=&arnumber=7539314
- 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
-
2
- Research group(s)
-
-
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper compares an automatic and a semi-automatic approach to machine learning, and for the first time, uses information-theoretic measures to estimate quantitatively the human knowledge available to the semi-automatic approach. The work builds on two real-world case studies (i.e., facial expression recognition and image categorization), and apply information-theoretic measures consistently across the machine learning processes, including data for training and testing, metrics for decision tree learning, and human knowledge. It received the best paper award in IEEE VAST2016, and helped stimulate a surge of VIS-4-ML research activities (from 4 papers in VAST2016 to 14 in VAST2017).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -