SwePub
Sök i LIBRIS databas

  Utökad sökning

onr:"swepub:oai:DiVA.org:his-2096"
 

Sökning: onr:"swepub:oai:DiVA.org:his-2096" > Maximizing the Area...

LIBRIS Formathandbok  (Information om MARC21)
FältnamnIndikatorerMetadata
00003571naa a2200397 4500
001oai:DiVA.org:his-2096
003SwePub
008080530s2007 | |||||||||||000 ||eng|
009oai:DiVA.org:kth-221459
024a https://urn.kb.se/resolve?urn=urn:nbn:se:his:diva-20962 URI
024a https://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-2214592 URI
040 a (SwePub)hisd (SwePub)kth
041 a engb eng
042 9 SwePub
072 7a ref2 swepub-contenttype
072 7a kon2 swepub-publicationtype
100a Boström, Henriku Högskolan i Skövde,Institutionen för kommunikation och information,Forskningscentrum för Informationsteknologi,Högskolan i Skövde, Institutionen för kommunikation och information4 aut0 (Swepub:kth)u1r0rr47
2451 0a Maximizing the Area under the ROC Curve with Decision Lists and Rule Sets
264 1b Society for Industrial and Applied Mathematics,c 2007
338 a print2 rdacarrier
500 a QC 20180119
520 a Decision lists (or ordered rule sets) have two attractive properties compared to unordered rule sets: they require a simpler classi¯cation procedure and they allow for a more compact representation. However, it is an open question what effect these properties have on the area under the ROC curve (AUC). Two ways of forming decision lists are considered in this study: by generating a sequence of rules, with a default rule for one of the classes, and by imposing an order upon rules that have been generated for all classes. An empirical investigation shows that the latter method gives a significantly higher AUC than the former, demonstrating that the compactness obtained by using one of the classes as a default is indeed associated with a cost. Furthermore, by using all applicable rules rather than the first in an ordered set, an even further significant improvement in AUC is obtained, demonstrating that the simple classification procedure is also associated with a cost. The observed gains in AUC for unordered rule sets compared to decision lists can be explained by that learning rules for all classes as well as combining multiple rules allow for examples to be ranked according to a more fine-grained scale compared to when applying rules in a fixed order and providing a default rule for one of the classes.
650 7a NATURVETENSKAPx Data- och informationsvetenskapx Datorseende och robotik0 (SwePub)102072 hsv//swe
650 7a NATURAL SCIENCESx Computer and Information Sciencesx Computer Vision and Robotics0 (SwePub)102072 hsv//eng
650 7a NATURVETENSKAPx Matematikx Diskret matematik0 (SwePub)101042 hsv//swe
650 7a NATURAL SCIENCESx Mathematicsx Discrete Mathematics0 (SwePub)101042 hsv//eng
650 7a NATURVETENSKAPx Data- och informationsvetenskapx Datavetenskap0 (SwePub)102012 hsv//swe
650 7a NATURAL SCIENCESx Computer and Information Sciencesx Computer Sciences0 (SwePub)102012 hsv//eng
650 7a NATURVETENSKAPx Data- och informationsvetenskap0 (SwePub)1022 hsv//swe
650 7a NATURAL SCIENCESx Computer and Information Sciences0 (SwePub)1022 hsv//eng
653 a Technology
653 a Teknik
710a Högskolan i Skövdeb Institutionen för kommunikation och information4 org
773t Proceedings of the 7th SIAM International Conference on Data Miningd : Society for Industrial and Applied Mathematicsg , s. 27-34q <27-34z 9780898716306
8564 8u https://urn.kb.se/resolve?urn=urn:nbn:se:his:diva-2096
8564 8u https://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-221459

Hitta via bibliotek

Till lärosätets databas

Sök utanför SwePub

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.

 
pil uppåt Stäng

Kopiera och spara länken för att återkomma till aktuell vy