Sökning: onr:"swepub:oai:lup.lub.lu.se:a3527bd8-8ffc-4c59-a221-9c90a1227ac1" >
Exploring new possi...
-
Green, MichaelLund University,Lunds universitet,Beräkningsbiologi och biologisk fysik - Genomgår omorganisation,Institutionen för astronomi och teoretisk fysik - Genomgår omorganisation,Naturvetenskapliga fakulteten,Computational Biology and Biological Physics - Undergoing reorganization,Department of Astronomy and Theoretical Physics - Undergoing reorganization,Faculty of Science
(författare)
Exploring new possibilities for case based explanation of artificial neural network ensembles
- Artikel/kapitelEngelska2009
Förlag, utgivningsår, omfång ...
Nummerbeteckningar
-
LIBRIS-ID:oai:lup.lub.lu.se:a3527bd8-8ffc-4c59-a221-9c90a1227ac1
-
https://lup.lub.lu.se/record/779530URI
-
https://doi.org/10.1016/j.neunet.2008.09.014DOI
Kompletterande språkuppgifter
-
Språk:engelska
-
Sammanfattning på:engelska
Ingår i deldatabas
Klassifikation
-
Ämneskategori:art swepub-publicationtype
-
Ämneskategori:ref swepub-contenttype
Anmärkningar
-
Artificial neural network (ANN) ensembles have long suffered from a lack of interpretability. This has severely limited the practical usability of ANNs in settings where an erroneous decision can be disastrous. Several attempts have been made to alleviate this problem. Many of them are based on decomposing the decision boundary of the ANN into a set of rules. We explore and compare a set of new methods for this explanation process on two artificial data sets (Monks 1 and 3), and one acute coronary syndrome data set consisting of 861 electrocardiograms (ECG) collected retrospectively at the emergency department at Lund University Hospital. The algorithms managed to extract good explanations in more than 84% of the cases. More to the point, the best method provided 99% and 91% good explanations in Monks data 1 and 3 respectively. Also there was a significant overlap between the algorithms. Furthermore, when explaining a given ECG, the overlap between this method and one of the physicians was the same as the one between the two physicians in this study. Still the physicians were significantly, p-value <0.001, more similar to each other than to any of the methods. The algorithms have the potential to be used as an explanatory aid when using ANN ensembles in clinical decision support systems.
Ämnesord och genrebeteckningar
Biuppslag (personer, institutioner, konferenser, titlar ...)
-
Ekelund, UlfLund University,Lunds universitet,Medicin, Lund,Sektion II,Institutionen för kliniska vetenskaper, Lund,Medicinska fakulteten,Medicine, Lund,Section II,Department of Clinical Sciences, Lund,Faculty of Medicine(Swepub:lu)mphy-uek
(författare)
-
Edenbrandt, LarsLund University,Lunds universitet,Nuklearmedicin, Malmö,Forskargrupper vid Lunds universitet,Nuclear medicine, Malmö,Lund University Research Groups(Swepub:lu)klfy-led
(författare)
-
Björk, JonasLund University,Lunds universitet,Centrum för ekonomisk demografi,Ekonomihögskolan,Avdelningen för arbets- och miljömedicin,Institutionen för laboratoriemedicin,Medicinska fakulteten,Centre for Economic Demography,Lund University School of Economics and Management, LUSEM,Division of Occupational and Environmental Medicine, Lund University,Department of Laboratory Medicine,Faculty of Medicine(Swepub:lu)ymed-jbj
(författare)
-
Lundager Hansen, JakobSkåne University Hospital(Swepub:lu)ja8218lu
(författare)
-
Ohlsson, MattiasLund University,Lunds universitet,Beräkningsbiologi och biologisk fysik - Genomgår omorganisation,Institutionen för astronomi och teoretisk fysik - Genomgår omorganisation,Naturvetenskapliga fakulteten,Computational Biology and Biological Physics - Undergoing reorganization,Department of Astronomy and Theoretical Physics - Undergoing reorganization,Faculty of Science(Swepub:lu)thep-moh
(författare)
-
Beräkningsbiologi och biologisk fysik - Genomgår omorganisationInstitutionen för astronomi och teoretisk fysik - Genomgår omorganisation
(creator_code:org_t)
Sammanhörande titlar
-
Ingår i:Neural Networks: Elsevier BV22:1, s. 75-811879-27820893-6080
Internetlänk
Hitta via bibliotek
Till lärosätets databas