Quantifying the informativeness of similarity measurements
- Submitting institution
-
The University of Liverpool
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 12095
- Type
- D - Journal article
- DOI
-
-
- Title of journal
- Journal of Machine Learning Research
- Article number
- 76
- First page
- 1
- Volume
- 18
- Issue
- -
- ISSN
- 1532-4435
- Open access status
- Compliant
- Month of publication
- July
- Year of publication
- 2017
- 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
-
3
- Research group(s)
-
-
- Citation count
- 2
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This article introduces a formal notion of informativeness of similarity measures and applies it to several machine learning tasks. The contribution and its potential applications have been investigated and positively evaluated in the TOPEX project of the US Airforce on topological data analysis (https://apps.dtic.mil/sti/pdfs/AD1100477.pdf). The new informativeness measure has been applied, for example, to graph data in "Scalable analysis of open data graphs" (IEEE IRI 2019) and to real-life mixed data in "Mix2Vec: Unsupervised Mixed Data Representation" (DSAA 2020).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -