Detection of semantically incorrect rules in knowledge-based systems
Identifieur interne : 000182 ( Main/Exploration ); précédent : 000181; suivant : 000183Detection of semantically incorrect rules in knowledge-based systems
Auteurs : L.-M. Fu [États-Unis]Source :
- Knowledge-Based Systems [ 0950-7051 ] ; 1992.
Abstract
A novel approach that applies the technique of back propagation to the recognition of semantically incorrect rules is presented. When the rule strengths of most rules are semantically correct, semantically incorrect rules can be recognized if their strengths are weakened or change signs after training with correct samples. In each training cycle, the discrepancies in the belief values of goal hypotheses are propagated backwards, and the strengths of rules responsible for such discrepancies are modified appropriately. A function called consistent-shift is defined for measuring the shift of a rule strength in the direction that is consistent with the strength assigned before training, and this is a critical component of this technique. A formal analysis of this approach is provided. The viability of this technique has been demonstrated in a practical domain.
Url:
DOI: 10.1016/0950-7051(92)90002-W
Affiliations:
Links toward previous steps (curation, corpus...)
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title>Detection of semantically incorrect rules in knowledge-based systems</title><author><name sortKey="Fu, L M" sort="Fu, L M" uniqKey="Fu L" first="L.-M." last="Fu">L.-M. Fu</name></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:512438D4B8F4CFC492E4D493ED0B3F65544A59CD</idno><date when="1992" year="1992">1992</date><idno type="doi">10.1016/0950-7051(92)90002-W</idno><idno type="url">https://api.istex.fr/document/512438D4B8F4CFC492E4D493ED0B3F65544A59CD/fulltext/pdf</idno><idno type="wicri:Area/Main/Corpus">000252</idno><idno type="wicri:Area/Main/Curation">000219</idno><idno type="wicri:Area/Main/Exploration">000182</idno><idno type="wicri:explorRef" wicri:stream="Main" wicri:step="Exploration">000182</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a">Detection of semantically incorrect rules in knowledge-based systems</title><author><name sortKey="Fu, L M" sort="Fu, L M" uniqKey="Fu L" first="L.-M." last="Fu">L.-M. Fu</name><affiliation wicri:level="1"><country xml:lang="fr">États-Unis</country><wicri:regionArea>Computer and Information Sciences Department, 301 CSE, University of Florida, Gainesville, Florida 32611</wicri:regionArea><wicri:noRegion>Florida 32611</wicri:noRegion></affiliation></author></analytic><monogr></monogr><series><title level="j">Knowledge-Based Systems</title><title level="j" type="abbrev">KNOSYS</title><idno type="ISSN">0950-7051</idno><imprint><publisher>ELSEVIER</publisher><date type="published" when="1992">1992</date><biblScope unit="volume">5</biblScope><biblScope unit="issue">2</biblScope><biblScope unit="page" from="117">117</biblScope><biblScope unit="page" to="124">124</biblScope></imprint><idno type="ISSN">0950-7051</idno></series><idno type="istex">512438D4B8F4CFC492E4D493ED0B3F65544A59CD</idno><idno type="DOI">10.1016/0950-7051(92)90002-W</idno><idno type="PII">0950-7051(92)90002-W</idno></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">0950-7051</idno></seriesStmt></fileDesc><profileDesc><textClass></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">A novel approach that applies the technique of back propagation to the recognition of semantically incorrect rules is presented. When the rule strengths of most rules are semantically correct, semantically incorrect rules can be recognized if their strengths are weakened or change signs after training with correct samples. In each training cycle, the discrepancies in the belief values of goal hypotheses are propagated backwards, and the strengths of rules responsible for such discrepancies are modified appropriately. A function called consistent-shift is defined for measuring the shift of a rule strength in the direction that is consistent with the strength assigned before training, and this is a critical component of this technique. A formal analysis of this approach is provided. The viability of this technique has been demonstrated in a practical domain.</div></front></TEI><affiliations><list><country><li>États-Unis</li></country></list><tree><country name="États-Unis"><noRegion><name sortKey="Fu, L M" sort="Fu, L M" uniqKey="Fu L" first="L.-M." last="Fu">L.-M. Fu</name></noRegion></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/TlfNancyV1/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000182 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000182 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Lorraine
|area= TlfNancyV1
|flux= Main
|étape= Exploration
|type= RBID
|clé= ISTEX:512438D4B8F4CFC492E4D493ED0B3F65544A59CD
|texte= Detection of semantically incorrect rules in knowledge-based systems
}}
| This area was generated with Dilib version V0.6.39. |  |