Inducing safer oblique trees without costs
Abstract: Decision tree induction has been widely studied and applied. In safety applications, such as determining whether a chemical process is safe or whether a person has a medical condition, the cost of misclassification in one of the classes is significantly higher than in the other class. Seve...
Ausführliche Beschreibung
Gespeichert in:
| Autor*in: |
Vadera, Sunil [verfasserIn] |
|---|
| Format: |
E-Artikel |
|---|
| Erschienen: |
Oxford, UK: Blackwell Publishing ; 2005 |
|---|
| Schlagwörter: |
|---|
| Umfang: |
Online-Ressource |
|---|
| Reproduktion: |
2005 ; Blackwell Publishing Journal Backfiles 1879-2005 |
|---|---|
| Übergeordnetes Werk: |
In: Expert systems - Oxford [u.a.] : Wiley-Blackwell, 1997, 22(2005), 4, Seite 0 |
| Übergeordnetes Werk: |
volume:22 ; year:2005 ; number:4 ; pages:0 |
| Links: |
|---|
| DOI / URN: |
10.1111/j.1468-0394.2005.00311.x |
|---|
| Katalog-ID: |
NLEJ242374468 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLEJ242374468 | ||
| 003 | DE-627 | ||
| 005 | 20210707154126.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 120427s2005 xx |||||o 00| ||und c | ||
| 024 | 7 | |a 10.1111/j.1468-0394.2005.00311.x |2 doi | |
| 035 | |a (DE-627)NLEJ242374468 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 100 | 1 | |a Vadera, Sunil |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Inducing safer oblique trees without costs |
| 264 | 1 | |a Oxford, UK |b Blackwell Publishing |c 2005 | |
| 300 | |a Online-Ressource | ||
| 336 | |a nicht spezifiziert |b zzz |2 rdacontent | ||
| 337 | |a nicht spezifiziert |b z |2 rdamedia | ||
| 338 | |a nicht spezifiziert |b zu |2 rdacarrier | ||
| 520 | |a Abstract: Decision tree induction has been widely studied and applied. In safety applications, such as determining whether a chemical process is safe or whether a person has a medical condition, the cost of misclassification in one of the classes is significantly higher than in the other class. Several authors have tackled this problem by developing cost-sensitive decision tree learning algorithms or have suggested ways of changing the distribution of training examples to bias the decision tree learning process so as to take account of costs. A prerequisite for applying such algorithms is the availability of costs of misclassification. Although this may be possible for some applications, obtaining reasonable estimates of costs of misclassification is not easy in the area of safety.This paper presents a new algorithm for applications where the cost of misclassifications cannot be quantified, although the cost of misclassification in one class is known to be significantly higher than in another class. The algorithm utilizes linear discriminant analysis to identify oblique relationships between continuous attributes and then carries out an appropriate modification to ensure that the resulting tree errs on the side of safety. The algorithm is evaluated with respect to one of the best known cost-sensitive algorithms (ICET), a well-known oblique decision tree algorithm (OC1) and an algorithm that utilizes robust linear programming. | ||
| 533 | |d 2005 |f Blackwell Publishing Journal Backfiles 1879-2005 |7 |2005|||||||||| | ||
| 650 | 4 | |a decision tree learning | |
| 773 | 0 | 8 | |i In |t Expert systems |d Oxford [u.a.] : Wiley-Blackwell, 1997 |g 22(2005), 4, Seite 0 |h Online-Ressource |w (DE-627)NLEJ243925662 |w (DE-600)2016958-9 |x 1468-0394 |7 nnns |
| 773 | 1 | 8 | |g volume:22 |g year:2005 |g number:4 |g pages:0 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1111/j.1468-0394.2005.00311.x |q text/html |x Verlag |z Deutschlandweit zugänglich |3 Volltext |
| 912 | |a GBV_USEFLAG_U | ||
| 912 | |a ZDB-1-DJB | ||
| 912 | |a GBV_NL_ARTICLE | ||
| 951 | |a AR | ||
| 952 | |d 22 |j 2005 |e 4 |h 0 | ||
| author_variant |
s v sv |
|---|---|
| matchkey_str |
article:14680394:2005----::nuigaeolqerew |
| hierarchy_sort_str |
2005 |
| publishDate |
2005 |
| allfields |
10.1111/j.1468-0394.2005.00311.x doi (DE-627)NLEJ242374468 DE-627 ger DE-627 rakwb Vadera, Sunil verfasserin aut Inducing safer oblique trees without costs Oxford, UK Blackwell Publishing 2005 Online-Ressource nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier Abstract: Decision tree induction has been widely studied and applied. In safety applications, such as determining whether a chemical process is safe or whether a person has a medical condition, the cost of misclassification in one of the classes is significantly higher than in the other class. Several authors have tackled this problem by developing cost-sensitive decision tree learning algorithms or have suggested ways of changing the distribution of training examples to bias the decision tree learning process so as to take account of costs. A prerequisite for applying such algorithms is the availability of costs of misclassification. Although this may be possible for some applications, obtaining reasonable estimates of costs of misclassification is not easy in the area of safety.This paper presents a new algorithm for applications where the cost of misclassifications cannot be quantified, although the cost of misclassification in one class is known to be significantly higher than in another class. The algorithm utilizes linear discriminant analysis to identify oblique relationships between continuous attributes and then carries out an appropriate modification to ensure that the resulting tree errs on the side of safety. The algorithm is evaluated with respect to one of the best known cost-sensitive algorithms (ICET), a well-known oblique decision tree algorithm (OC1) and an algorithm that utilizes robust linear programming. 2005 Blackwell Publishing Journal Backfiles 1879-2005 |2005|||||||||| decision tree learning In Expert systems Oxford [u.a.] : Wiley-Blackwell, 1997 22(2005), 4, Seite 0 Online-Ressource (DE-627)NLEJ243925662 (DE-600)2016958-9 1468-0394 nnns volume:22 year:2005 number:4 pages:0 http://dx.doi.org/10.1111/j.1468-0394.2005.00311.x text/html Verlag Deutschlandweit zugänglich Volltext GBV_USEFLAG_U ZDB-1-DJB GBV_NL_ARTICLE AR 22 2005 4 0 |
| spelling |
10.1111/j.1468-0394.2005.00311.x doi (DE-627)NLEJ242374468 DE-627 ger DE-627 rakwb Vadera, Sunil verfasserin aut Inducing safer oblique trees without costs Oxford, UK Blackwell Publishing 2005 Online-Ressource nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier Abstract: Decision tree induction has been widely studied and applied. In safety applications, such as determining whether a chemical process is safe or whether a person has a medical condition, the cost of misclassification in one of the classes is significantly higher than in the other class. Several authors have tackled this problem by developing cost-sensitive decision tree learning algorithms or have suggested ways of changing the distribution of training examples to bias the decision tree learning process so as to take account of costs. A prerequisite for applying such algorithms is the availability of costs of misclassification. Although this may be possible for some applications, obtaining reasonable estimates of costs of misclassification is not easy in the area of safety.This paper presents a new algorithm for applications where the cost of misclassifications cannot be quantified, although the cost of misclassification in one class is known to be significantly higher than in another class. The algorithm utilizes linear discriminant analysis to identify oblique relationships between continuous attributes and then carries out an appropriate modification to ensure that the resulting tree errs on the side of safety. The algorithm is evaluated with respect to one of the best known cost-sensitive algorithms (ICET), a well-known oblique decision tree algorithm (OC1) and an algorithm that utilizes robust linear programming. 2005 Blackwell Publishing Journal Backfiles 1879-2005 |2005|||||||||| decision tree learning In Expert systems Oxford [u.a.] : Wiley-Blackwell, 1997 22(2005), 4, Seite 0 Online-Ressource (DE-627)NLEJ243925662 (DE-600)2016958-9 1468-0394 nnns volume:22 year:2005 number:4 pages:0 http://dx.doi.org/10.1111/j.1468-0394.2005.00311.x text/html Verlag Deutschlandweit zugänglich Volltext GBV_USEFLAG_U ZDB-1-DJB GBV_NL_ARTICLE AR 22 2005 4 0 |
| allfields_unstemmed |
10.1111/j.1468-0394.2005.00311.x doi (DE-627)NLEJ242374468 DE-627 ger DE-627 rakwb Vadera, Sunil verfasserin aut Inducing safer oblique trees without costs Oxford, UK Blackwell Publishing 2005 Online-Ressource nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier Abstract: Decision tree induction has been widely studied and applied. In safety applications, such as determining whether a chemical process is safe or whether a person has a medical condition, the cost of misclassification in one of the classes is significantly higher than in the other class. Several authors have tackled this problem by developing cost-sensitive decision tree learning algorithms or have suggested ways of changing the distribution of training examples to bias the decision tree learning process so as to take account of costs. A prerequisite for applying such algorithms is the availability of costs of misclassification. Although this may be possible for some applications, obtaining reasonable estimates of costs of misclassification is not easy in the area of safety.This paper presents a new algorithm for applications where the cost of misclassifications cannot be quantified, although the cost of misclassification in one class is known to be significantly higher than in another class. The algorithm utilizes linear discriminant analysis to identify oblique relationships between continuous attributes and then carries out an appropriate modification to ensure that the resulting tree errs on the side of safety. The algorithm is evaluated with respect to one of the best known cost-sensitive algorithms (ICET), a well-known oblique decision tree algorithm (OC1) and an algorithm that utilizes robust linear programming. 2005 Blackwell Publishing Journal Backfiles 1879-2005 |2005|||||||||| decision tree learning In Expert systems Oxford [u.a.] : Wiley-Blackwell, 1997 22(2005), 4, Seite 0 Online-Ressource (DE-627)NLEJ243925662 (DE-600)2016958-9 1468-0394 nnns volume:22 year:2005 number:4 pages:0 http://dx.doi.org/10.1111/j.1468-0394.2005.00311.x text/html Verlag Deutschlandweit zugänglich Volltext GBV_USEFLAG_U ZDB-1-DJB GBV_NL_ARTICLE AR 22 2005 4 0 |
| allfieldsGer |
10.1111/j.1468-0394.2005.00311.x doi (DE-627)NLEJ242374468 DE-627 ger DE-627 rakwb Vadera, Sunil verfasserin aut Inducing safer oblique trees without costs Oxford, UK Blackwell Publishing 2005 Online-Ressource nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier Abstract: Decision tree induction has been widely studied and applied. In safety applications, such as determining whether a chemical process is safe or whether a person has a medical condition, the cost of misclassification in one of the classes is significantly higher than in the other class. Several authors have tackled this problem by developing cost-sensitive decision tree learning algorithms or have suggested ways of changing the distribution of training examples to bias the decision tree learning process so as to take account of costs. A prerequisite for applying such algorithms is the availability of costs of misclassification. Although this may be possible for some applications, obtaining reasonable estimates of costs of misclassification is not easy in the area of safety.This paper presents a new algorithm for applications where the cost of misclassifications cannot be quantified, although the cost of misclassification in one class is known to be significantly higher than in another class. The algorithm utilizes linear discriminant analysis to identify oblique relationships between continuous attributes and then carries out an appropriate modification to ensure that the resulting tree errs on the side of safety. The algorithm is evaluated with respect to one of the best known cost-sensitive algorithms (ICET), a well-known oblique decision tree algorithm (OC1) and an algorithm that utilizes robust linear programming. 2005 Blackwell Publishing Journal Backfiles 1879-2005 |2005|||||||||| decision tree learning In Expert systems Oxford [u.a.] : Wiley-Blackwell, 1997 22(2005), 4, Seite 0 Online-Ressource (DE-627)NLEJ243925662 (DE-600)2016958-9 1468-0394 nnns volume:22 year:2005 number:4 pages:0 http://dx.doi.org/10.1111/j.1468-0394.2005.00311.x text/html Verlag Deutschlandweit zugänglich Volltext GBV_USEFLAG_U ZDB-1-DJB GBV_NL_ARTICLE AR 22 2005 4 0 |
| allfieldsSound |
10.1111/j.1468-0394.2005.00311.x doi (DE-627)NLEJ242374468 DE-627 ger DE-627 rakwb Vadera, Sunil verfasserin aut Inducing safer oblique trees without costs Oxford, UK Blackwell Publishing 2005 Online-Ressource nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier Abstract: Decision tree induction has been widely studied and applied. In safety applications, such as determining whether a chemical process is safe or whether a person has a medical condition, the cost of misclassification in one of the classes is significantly higher than in the other class. Several authors have tackled this problem by developing cost-sensitive decision tree learning algorithms or have suggested ways of changing the distribution of training examples to bias the decision tree learning process so as to take account of costs. A prerequisite for applying such algorithms is the availability of costs of misclassification. Although this may be possible for some applications, obtaining reasonable estimates of costs of misclassification is not easy in the area of safety.This paper presents a new algorithm for applications where the cost of misclassifications cannot be quantified, although the cost of misclassification in one class is known to be significantly higher than in another class. The algorithm utilizes linear discriminant analysis to identify oblique relationships between continuous attributes and then carries out an appropriate modification to ensure that the resulting tree errs on the side of safety. The algorithm is evaluated with respect to one of the best known cost-sensitive algorithms (ICET), a well-known oblique decision tree algorithm (OC1) and an algorithm that utilizes robust linear programming. 2005 Blackwell Publishing Journal Backfiles 1879-2005 |2005|||||||||| decision tree learning In Expert systems Oxford [u.a.] : Wiley-Blackwell, 1997 22(2005), 4, Seite 0 Online-Ressource (DE-627)NLEJ243925662 (DE-600)2016958-9 1468-0394 nnns volume:22 year:2005 number:4 pages:0 http://dx.doi.org/10.1111/j.1468-0394.2005.00311.x text/html Verlag Deutschlandweit zugänglich Volltext GBV_USEFLAG_U ZDB-1-DJB GBV_NL_ARTICLE AR 22 2005 4 0 |
| source |
In Expert systems 22(2005), 4, Seite 0 volume:22 year:2005 number:4 pages:0 |
| sourceStr |
In Expert systems 22(2005), 4, Seite 0 volume:22 year:2005 number:4 pages:0 |
| format_phy_str_mv |
Article |
| institution |
findex.gbv.de |
| topic_facet |
decision tree learning |
| isfreeaccess_bool |
false |
| container_title |
Expert systems |
| authorswithroles_txt_mv |
Vadera, Sunil @@aut@@ |
| publishDateDaySort_date |
2005-01-01T00:00:00Z |
| hierarchy_top_id |
NLEJ243925662 |
| id |
NLEJ242374468 |
| fullrecord |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">NLEJ242374468</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20210707154126.0</controlfield><controlfield tag="007">cr uuu---uuuuu</controlfield><controlfield tag="008">120427s2005 xx |||||o 00| ||und c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1111/j.1468-0394.2005.00311.x</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)NLEJ242374468</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Vadera, Sunil</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Inducing safer oblique trees without costs</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="a">Oxford, UK</subfield><subfield code="b">Blackwell Publishing</subfield><subfield code="c">2005</subfield></datafield><datafield tag="300" ind1=" " ind2=" "><subfield code="a">Online-Ressource</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">zzz</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">z</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">zu</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Abstract: Decision tree induction has been widely studied and applied. In safety applications, such as determining whether a chemical process is safe or whether a person has a medical condition, the cost of misclassification in one of the classes is significantly higher than in the other class. Several authors have tackled this problem by developing cost-sensitive decision tree learning algorithms or have suggested ways of changing the distribution of training examples to bias the decision tree learning process so as to take account of costs. A prerequisite for applying such algorithms is the availability of costs of misclassification. Although this may be possible for some applications, obtaining reasonable estimates of costs of misclassification is not easy in the area of safety.This paper presents a new algorithm for applications where the cost of misclassifications cannot be quantified, although the cost of misclassification in one class is known to be significantly higher than in another class. The algorithm utilizes linear discriminant analysis to identify oblique relationships between continuous attributes and then carries out an appropriate modification to ensure that the resulting tree errs on the side of safety. The algorithm is evaluated with respect to one of the best known cost-sensitive algorithms (ICET), a well-known oblique decision tree algorithm (OC1) and an algorithm that utilizes robust linear programming.</subfield></datafield><datafield tag="533" ind1=" " ind2=" "><subfield code="d">2005</subfield><subfield code="f">Blackwell Publishing Journal Backfiles 1879-2005</subfield><subfield code="7">|2005||||||||||</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">decision tree learning</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">In</subfield><subfield code="t">Expert systems</subfield><subfield code="d">Oxford [u.a.] : Wiley-Blackwell, 1997</subfield><subfield code="g">22(2005), 4, Seite 0</subfield><subfield code="h">Online-Ressource</subfield><subfield code="w">(DE-627)NLEJ243925662</subfield><subfield code="w">(DE-600)2016958-9</subfield><subfield code="x">1468-0394</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:22</subfield><subfield code="g">year:2005</subfield><subfield code="g">number:4</subfield><subfield code="g">pages:0</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">http://dx.doi.org/10.1111/j.1468-0394.2005.00311.x</subfield><subfield code="q">text/html</subfield><subfield code="x">Verlag</subfield><subfield code="z">Deutschlandweit zugänglich</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_U</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">ZDB-1-DJB</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_NL_ARTICLE</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">22</subfield><subfield code="j">2005</subfield><subfield code="e">4</subfield><subfield code="h">0</subfield></datafield></record></collection>
|
| series2 |
Blackwell Publishing Journal Backfiles 1879-2005 |
| author |
Vadera, Sunil |
| spellingShingle |
Vadera, Sunil misc decision tree learning Inducing safer oblique trees without costs |
| authorStr |
Vadera, Sunil |
| ppnlink_with_tag_str_mv |
@@773@@(DE-627)NLEJ243925662 |
| format |
electronic Article |
| delete_txt_mv |
keep |
| author_role |
aut |
| collection |
NL |
| publishPlace |
Oxford, UK |
| remote_str |
true |
| illustrated |
Not Illustrated |
| issn |
1468-0394 |
| topic_title |
Inducing safer oblique trees without costs decision tree learning |
| publisher |
Blackwell Publishing |
| publisherStr |
Blackwell Publishing |
| topic |
misc decision tree learning |
| topic_unstemmed |
misc decision tree learning |
| topic_browse |
misc decision tree learning |
| format_facet |
Elektronische Aufsätze Aufsätze Elektronische Ressource |
| format_main_str_mv |
Text Zeitschrift/Artikel |
| carriertype_str_mv |
zu |
| hierarchy_parent_title |
Expert systems |
| hierarchy_parent_id |
NLEJ243925662 |
| hierarchy_top_title |
Expert systems |
| isfreeaccess_txt |
false |
| familylinks_str_mv |
(DE-627)NLEJ243925662 (DE-600)2016958-9 |
| title |
Inducing safer oblique trees without costs |
| ctrlnum |
(DE-627)NLEJ242374468 |
| title_full |
Inducing safer oblique trees without costs |
| author_sort |
Vadera, Sunil |
| journal |
Expert systems |
| journalStr |
Expert systems |
| isOA_bool |
false |
| recordtype |
marc |
| publishDateSort |
2005 |
| contenttype_str_mv |
zzz |
| container_start_page |
0 |
| author_browse |
Vadera, Sunil |
| container_volume |
22 |
| physical |
Online-Ressource |
| format_se |
Elektronische Aufsätze |
| author-letter |
Vadera, Sunil |
| doi_str_mv |
10.1111/j.1468-0394.2005.00311.x |
| title_sort |
inducing safer oblique trees without costs |
| title_auth |
Inducing safer oblique trees without costs |
| abstract |
Abstract: Decision tree induction has been widely studied and applied. In safety applications, such as determining whether a chemical process is safe or whether a person has a medical condition, the cost of misclassification in one of the classes is significantly higher than in the other class. Several authors have tackled this problem by developing cost-sensitive decision tree learning algorithms or have suggested ways of changing the distribution of training examples to bias the decision tree learning process so as to take account of costs. A prerequisite for applying such algorithms is the availability of costs of misclassification. Although this may be possible for some applications, obtaining reasonable estimates of costs of misclassification is not easy in the area of safety.This paper presents a new algorithm for applications where the cost of misclassifications cannot be quantified, although the cost of misclassification in one class is known to be significantly higher than in another class. The algorithm utilizes linear discriminant analysis to identify oblique relationships between continuous attributes and then carries out an appropriate modification to ensure that the resulting tree errs on the side of safety. The algorithm is evaluated with respect to one of the best known cost-sensitive algorithms (ICET), a well-known oblique decision tree algorithm (OC1) and an algorithm that utilizes robust linear programming. |
| abstractGer |
Abstract: Decision tree induction has been widely studied and applied. In safety applications, such as determining whether a chemical process is safe or whether a person has a medical condition, the cost of misclassification in one of the classes is significantly higher than in the other class. Several authors have tackled this problem by developing cost-sensitive decision tree learning algorithms or have suggested ways of changing the distribution of training examples to bias the decision tree learning process so as to take account of costs. A prerequisite for applying such algorithms is the availability of costs of misclassification. Although this may be possible for some applications, obtaining reasonable estimates of costs of misclassification is not easy in the area of safety.This paper presents a new algorithm for applications where the cost of misclassifications cannot be quantified, although the cost of misclassification in one class is known to be significantly higher than in another class. The algorithm utilizes linear discriminant analysis to identify oblique relationships between continuous attributes and then carries out an appropriate modification to ensure that the resulting tree errs on the side of safety. The algorithm is evaluated with respect to one of the best known cost-sensitive algorithms (ICET), a well-known oblique decision tree algorithm (OC1) and an algorithm that utilizes robust linear programming. |
| abstract_unstemmed |
Abstract: Decision tree induction has been widely studied and applied. In safety applications, such as determining whether a chemical process is safe or whether a person has a medical condition, the cost of misclassification in one of the classes is significantly higher than in the other class. Several authors have tackled this problem by developing cost-sensitive decision tree learning algorithms or have suggested ways of changing the distribution of training examples to bias the decision tree learning process so as to take account of costs. A prerequisite for applying such algorithms is the availability of costs of misclassification. Although this may be possible for some applications, obtaining reasonable estimates of costs of misclassification is not easy in the area of safety.This paper presents a new algorithm for applications where the cost of misclassifications cannot be quantified, although the cost of misclassification in one class is known to be significantly higher than in another class. The algorithm utilizes linear discriminant analysis to identify oblique relationships between continuous attributes and then carries out an appropriate modification to ensure that the resulting tree errs on the side of safety. The algorithm is evaluated with respect to one of the best known cost-sensitive algorithms (ICET), a well-known oblique decision tree algorithm (OC1) and an algorithm that utilizes robust linear programming. |
| collection_details |
GBV_USEFLAG_U ZDB-1-DJB GBV_NL_ARTICLE |
| container_issue |
4 |
| title_short |
Inducing safer oblique trees without costs |
| url |
http://dx.doi.org/10.1111/j.1468-0394.2005.00311.x |
| remote_bool |
true |
| ppnlink |
NLEJ243925662 |
| mediatype_str_mv |
z |
| isOA_txt |
false |
| hochschulschrift_bool |
false |
| doi_str |
10.1111/j.1468-0394.2005.00311.x |
| up_date |
2024-07-06T01:45:39.395Z |
| _version_ |
1803792248361451520 |
| fullrecord_marcxml |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">NLEJ242374468</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20210707154126.0</controlfield><controlfield tag="007">cr uuu---uuuuu</controlfield><controlfield tag="008">120427s2005 xx |||||o 00| ||und c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1111/j.1468-0394.2005.00311.x</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)NLEJ242374468</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Vadera, Sunil</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Inducing safer oblique trees without costs</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="a">Oxford, UK</subfield><subfield code="b">Blackwell Publishing</subfield><subfield code="c">2005</subfield></datafield><datafield tag="300" ind1=" " ind2=" "><subfield code="a">Online-Ressource</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">zzz</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">z</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">zu</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Abstract: Decision tree induction has been widely studied and applied. In safety applications, such as determining whether a chemical process is safe or whether a person has a medical condition, the cost of misclassification in one of the classes is significantly higher than in the other class. Several authors have tackled this problem by developing cost-sensitive decision tree learning algorithms or have suggested ways of changing the distribution of training examples to bias the decision tree learning process so as to take account of costs. A prerequisite for applying such algorithms is the availability of costs of misclassification. Although this may be possible for some applications, obtaining reasonable estimates of costs of misclassification is not easy in the area of safety.This paper presents a new algorithm for applications where the cost of misclassifications cannot be quantified, although the cost of misclassification in one class is known to be significantly higher than in another class. The algorithm utilizes linear discriminant analysis to identify oblique relationships between continuous attributes and then carries out an appropriate modification to ensure that the resulting tree errs on the side of safety. The algorithm is evaluated with respect to one of the best known cost-sensitive algorithms (ICET), a well-known oblique decision tree algorithm (OC1) and an algorithm that utilizes robust linear programming.</subfield></datafield><datafield tag="533" ind1=" " ind2=" "><subfield code="d">2005</subfield><subfield code="f">Blackwell Publishing Journal Backfiles 1879-2005</subfield><subfield code="7">|2005||||||||||</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">decision tree learning</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">In</subfield><subfield code="t">Expert systems</subfield><subfield code="d">Oxford [u.a.] : Wiley-Blackwell, 1997</subfield><subfield code="g">22(2005), 4, Seite 0</subfield><subfield code="h">Online-Ressource</subfield><subfield code="w">(DE-627)NLEJ243925662</subfield><subfield code="w">(DE-600)2016958-9</subfield><subfield code="x">1468-0394</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:22</subfield><subfield code="g">year:2005</subfield><subfield code="g">number:4</subfield><subfield code="g">pages:0</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">http://dx.doi.org/10.1111/j.1468-0394.2005.00311.x</subfield><subfield code="q">text/html</subfield><subfield code="x">Verlag</subfield><subfield code="z">Deutschlandweit zugänglich</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_U</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">ZDB-1-DJB</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_NL_ARTICLE</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">22</subfield><subfield code="j">2005</subfield><subfield code="e">4</subfield><subfield code="h">0</subfield></datafield></record></collection>
|
| score |
7.399764 |