Learning Bayesian networks from big data with greedy search: computational complexity and efficient implementation
- Submitting institution
-
Brunel University London
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 069-202552-5379
- Type
- D - Journal article
- DOI
-
10.1007/s11222-019-09857-1
- Title of journal
- Statistics And Computing
- Article number
- -
- First page
- 1095
- Volume
- 29
- Issue
- 5
- ISSN
- 0960-3174
- Open access status
- Compliant
- Month of publication
- February
- Year of publication
- 2019
- URL
-
https://link.springer.com/content/pdf/10.1007%2Fs11222-019-09857-1.pdf
- 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
-
2
- Research group(s)
-
1 - Artificial Intelligence (AI)
- Citation count
- 10
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper developed a novel approach to scalable learning of Bayesian networks and was tested on numerous real-world datasets. The journal is ranked among top 10% in Statistics and Probability. The implementation is part of the bnlearn R package which has received 113028 downloads by Nov 2020 and has been used by many institutions including the Medicine and Health Regulatory Authority who have used it to generate synthetic data (https://www.gov.uk/government/news/new-synthetic-datasets-to-assist-covid-19-and-cardiovascular-research). The approach has been tested on both environmental air quality data and multiple clinical datasets and is discussed on the main bnlearn website: https://www.bnlearn.com/research/statco19/
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
