A new method based on context for combining statistical language models
Identifieur interne :
000938 ( PascalFrancis/Corpus );
précédent :
000937;
suivant :
000939
A new method based on context for combining statistical language models
Auteurs : David Langlois ;
Kamel Smaïli ;
Jean-Paul HatonSource :
-
Lecture notes in computer science [ 0302-9743 ] ; 2001.
RBID : Pascal:01-0381622
Descripteurs français
English descriptors
Abstract
In this paper we propose a new method to extract from a corpus the histories for which a given language model is better than another one. The decision is based on a measure stemmed from perplexity. This measure allows, for a given history, to compare two language models, and then to choose the best one for this history. Using this principle, and with a 20K vocabulary words, we combined two language models: a bigram and a distant bigram. The contribution of a distant bigram is significant and outperforms a bigram model by 7.5%. Moreover, the performance in Shannon game are improved. We show through this article that we proposed a cheaper framework in comparison to the maximum entropy principle, for combining language models. In addition, the selected histories for which a model is better than another one, have been collected and studied. Almost, all of them are beginnings of very frequently used French phrases. Finally, by using this principle, we achieve a better trigram model in terms of parameters and perplexity. This model is a combination of a bigram and a trigram based on a selected history.
Notice en format standard (ISO 2709)
Pour connaître la documentation sur le format Inist Standard.
pA |
A01 | 01 | 1 | | @0 0302-9743 |
---|
A05 | | | | @2 2116 |
---|
A08 | 01 | 1 | ENG | @1 A new method based on context for combining statistical language models |
---|
A09 | 01 | 1 | ENG | @1 CONTEXT 2001 : modeling and using context : Dundee, 27-30 July 2001 |
---|
A11 | 01 | 1 | | @1 LANGLOIS (David) |
---|
A11 | 02 | 1 | | @1 SMAÏLI (Kamel) |
---|
A11 | 03 | 1 | | @1 HATON (Jean-Paul) |
---|
A12 | 01 | 1 | | @1 AKMAN (Varol) @9 ed. |
---|
A12 | 02 | 1 | | @1 BOUQUET (Paolo) @9 ed. |
---|
A12 | 03 | 1 | | @1 THOMASON (Richmond) @9 ed. |
---|
A12 | 04 | 1 | | @1 YOUNG (Roger A.) @9 ed. |
---|
A14 | 01 | | | @1 LORIA Laboratory, Campus Scientifique BP 239 @2 54506 Vandœuvre-Lès-Nancy @3 FRA @Z 1 aut. @Z 2 aut. @Z 3 aut. |
---|
A20 | | | | @1 235-247 |
---|
A21 | | | | @1 2001 |
---|
A23 | 01 | | | @0 ENG |
---|
A26 | 01 | | | @0 3-540-42379-6 |
---|
A43 | 01 | | | @1 INIST @2 16343 @5 354000092486680180 |
---|
A44 | | | | @0 0000 @1 © 2001 INIST-CNRS. All rights reserved. |
---|
A45 | | | | @0 15 ref. |
---|
A47 | 01 | 1 | | @0 01-0381622 |
---|
A60 | | | | @1 P @2 C |
---|
A61 | | | | @0 A |
---|
A64 | 01 | 1 | | @0 Lecture notes in computer science |
---|
A66 | 01 | | | @0 DEU |
---|
A66 | 02 | | | @0 USA |
---|
C01 | 01 | | ENG | @0 In this paper we propose a new method to extract from a corpus the histories for which a given language model is better than another one. The decision is based on a measure stemmed from perplexity. This measure allows, for a given history, to compare two language models, and then to choose the best one for this history. Using this principle, and with a 20K vocabulary words, we combined two language models: a bigram and a distant bigram. The contribution of a distant bigram is significant and outperforms a bigram model by 7.5%. Moreover, the performance in Shannon game are improved. We show through this article that we proposed a cheaper framework in comparison to the maximum entropy principle, for combining language models. In addition, the selected histories for which a model is better than another one, have been collected and studied. Almost, all of them are beginnings of very frequently used French phrases. Finally, by using this principle, we achieve a better trigram model in terms of parameters and perplexity. This model is a combination of a bigram and a trigram based on a selected history. |
---|
C02 | 01 | X | | @0 001D02C04 |
---|
C03 | 01 | X | FRE | @0 Méthode entropie maximum @5 01 |
---|
C03 | 01 | X | ENG | @0 Method of maximum entropy @5 01 |
---|
C03 | 01 | X | SPA | @0 Método entropía máxima @5 01 |
---|
C03 | 02 | X | FRE | @0 Principe maximum @5 02 |
---|
C03 | 02 | X | ENG | @0 Maximum principle @5 02 |
---|
C03 | 02 | X | SPA | @0 Principio máximo @5 02 |
---|
C03 | 03 | X | FRE | @0 Théorie Shannon @5 03 |
---|
C03 | 03 | X | ENG | @0 Shannon theory @5 03 |
---|
C03 | 03 | X | SPA | @0 Teoría Shannon @5 03 |
---|
C03 | 04 | X | FRE | @0 Vocabulaire @5 04 |
---|
C03 | 04 | X | ENG | @0 Vocabulary @5 04 |
---|
C03 | 04 | X | SPA | @0 Vocabulario @5 04 |
---|
C03 | 05 | X | FRE | @0 Mot @5 05 |
---|
C03 | 05 | X | ENG | @0 Word @5 05 |
---|
C03 | 05 | X | SPA | @0 Palabra @5 05 |
---|
C03 | 06 | X | FRE | @0 Evaluation performance @5 06 |
---|
C03 | 06 | X | ENG | @0 Performance evaluation @5 06 |
---|
C03 | 06 | X | SPA | @0 Evaluación prestación @5 06 |
---|
C03 | 07 | X | FRE | @0 Modélisation @5 07 |
---|
C03 | 07 | X | ENG | @0 Modeling @5 07 |
---|
C03 | 07 | X | SPA | @0 Modelización @5 07 |
---|
C03 | 08 | X | FRE | @0 Méthode statistique @5 08 |
---|
C03 | 08 | X | ENG | @0 Statistical method @5 08 |
---|
C03 | 08 | X | SPA | @0 Método estadístico @5 08 |
---|
C03 | 09 | X | FRE | @0 Reconnaissance parole @5 09 |
---|
C03 | 09 | X | ENG | @0 Speech recognition @5 09 |
---|
C03 | 09 | X | SPA | @0 Reconocimiento palabra @5 09 |
---|
N21 | | | | @1 267 |
---|
|
pR |
A30 | 01 | 1 | ENG | @1 Modeling and using context. International and interdisciplinary conference @2 3 @3 Dundee GBR @4 2001-07-27 |
---|
|
Format Inist (serveur)
NO : | PASCAL 01-0381622 INIST |
ET : | A new method based on context for combining statistical language models |
AU : | LANGLOIS (David); SMAÏLI (Kamel); HATON (Jean-Paul); AKMAN (Varol); BOUQUET (Paolo); THOMASON (Richmond); YOUNG (Roger A.) |
AF : | LORIA Laboratory, Campus Scientifique BP 239/54506 Vandœuvre-Lès-Nancy/France (1 aut., 2 aut., 3 aut.) |
DT : | Publication en série; Congrès; Niveau analytique |
SO : | Lecture notes in computer science; ISSN 0302-9743; Allemagne; Da. 2001; Vol. 2116; Pp. 235-247; Bibl. 15 ref. |
LA : | Anglais |
EA : | In this paper we propose a new method to extract from a corpus the histories for which a given language model is better than another one. The decision is based on a measure stemmed from perplexity. This measure allows, for a given history, to compare two language models, and then to choose the best one for this history. Using this principle, and with a 20K vocabulary words, we combined two language models: a bigram and a distant bigram. The contribution of a distant bigram is significant and outperforms a bigram model by 7.5%. Moreover, the performance in Shannon game are improved. We show through this article that we proposed a cheaper framework in comparison to the maximum entropy principle, for combining language models. In addition, the selected histories for which a model is better than another one, have been collected and studied. Almost, all of them are beginnings of very frequently used French phrases. Finally, by using this principle, we achieve a better trigram model in terms of parameters and perplexity. This model is a combination of a bigram and a trigram based on a selected history. |
CC : | 001D02C04 |
FD : | Méthode entropie maximum; Principe maximum; Théorie Shannon; Vocabulaire; Mot; Evaluation performance; Modélisation; Méthode statistique; Reconnaissance parole |
ED : | Method of maximum entropy; Maximum principle; Shannon theory; Vocabulary; Word; Performance evaluation; Modeling; Statistical method; Speech recognition |
SD : | Método entropía máxima; Principio máximo; Teoría Shannon; Vocabulario; Palabra; Evaluación prestación; Modelización; Método estadístico; Reconocimiento palabra |
LO : | INIST-16343.354000092486680180 |
ID : | 01-0381622 |
Links to Exploration step
Pascal:01-0381622
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">A new method based on context for combining statistical language models</title>
<author><name sortKey="Langlois, David" sort="Langlois, David" uniqKey="Langlois D" first="David" last="Langlois">David Langlois</name>
<affiliation><inist:fA14 i1="01"><s1>LORIA Laboratory, Campus Scientifique BP 239</s1>
<s2>54506 Vandœuvre-Lès-Nancy</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
</affiliation>
</author>
<author><name sortKey="Smaili, Kamel" sort="Smaili, Kamel" uniqKey="Smaili K" first="Kamel" last="Smaïli">Kamel Smaïli</name>
<affiliation><inist:fA14 i1="01"><s1>LORIA Laboratory, Campus Scientifique BP 239</s1>
<s2>54506 Vandœuvre-Lès-Nancy</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
</affiliation>
</author>
<author><name sortKey="Haton, Jean Paul" sort="Haton, Jean Paul" uniqKey="Haton J" first="Jean-Paul" last="Haton">Jean-Paul Haton</name>
<affiliation><inist:fA14 i1="01"><s1>LORIA Laboratory, Campus Scientifique BP 239</s1>
<s2>54506 Vandœuvre-Lès-Nancy</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">INIST</idno>
<idno type="inist">01-0381622</idno>
<date when="2001">2001</date>
<idno type="stanalyst">PASCAL 01-0381622 INIST</idno>
<idno type="RBID">Pascal:01-0381622</idno>
<idno type="wicri:Area/PascalFrancis/Corpus">000938</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">A new method based on context for combining statistical language models</title>
<author><name sortKey="Langlois, David" sort="Langlois, David" uniqKey="Langlois D" first="David" last="Langlois">David Langlois</name>
<affiliation><inist:fA14 i1="01"><s1>LORIA Laboratory, Campus Scientifique BP 239</s1>
<s2>54506 Vandœuvre-Lès-Nancy</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
</affiliation>
</author>
<author><name sortKey="Smaili, Kamel" sort="Smaili, Kamel" uniqKey="Smaili K" first="Kamel" last="Smaïli">Kamel Smaïli</name>
<affiliation><inist:fA14 i1="01"><s1>LORIA Laboratory, Campus Scientifique BP 239</s1>
<s2>54506 Vandœuvre-Lès-Nancy</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
</affiliation>
</author>
<author><name sortKey="Haton, Jean Paul" sort="Haton, Jean Paul" uniqKey="Haton J" first="Jean-Paul" last="Haton">Jean-Paul Haton</name>
<affiliation><inist:fA14 i1="01"><s1>LORIA Laboratory, Campus Scientifique BP 239</s1>
<s2>54506 Vandœuvre-Lès-Nancy</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
</affiliation>
</author>
</analytic>
<series><title level="j" type="main">Lecture notes in computer science</title>
<idno type="ISSN">0302-9743</idno>
<imprint><date when="2001">2001</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><title level="j" type="main">Lecture notes in computer science</title>
<idno type="ISSN">0302-9743</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Maximum principle</term>
<term>Method of maximum entropy</term>
<term>Modeling</term>
<term>Performance evaluation</term>
<term>Shannon theory</term>
<term>Speech recognition</term>
<term>Statistical method</term>
<term>Vocabulary</term>
<term>Word</term>
</keywords>
<keywords scheme="Pascal" xml:lang="fr"><term>Méthode entropie maximum</term>
<term>Principe maximum</term>
<term>Théorie Shannon</term>
<term>Vocabulaire</term>
<term>Mot</term>
<term>Evaluation performance</term>
<term>Modélisation</term>
<term>Méthode statistique</term>
<term>Reconnaissance parole</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">In this paper we propose a new method to extract from a corpus the histories for which a given language model is better than another one. The decision is based on a measure stemmed from perplexity. This measure allows, for a given history, to compare two language models, and then to choose the best one for this history. Using this principle, and with a 20K vocabulary words, we combined two language models: a bigram and a distant bigram. The contribution of a distant bigram is significant and outperforms a bigram model by 7.5%. Moreover, the performance in Shannon game are improved. We show through this article that we proposed a cheaper framework in comparison to the maximum entropy principle, for combining language models. In addition, the selected histories for which a model is better than another one, have been collected and studied. Almost, all of them are beginnings of very frequently used French phrases. Finally, by using this principle, we achieve a better trigram model in terms of parameters and perplexity. This model is a combination of a bigram and a trigram based on a selected history.</div>
</front>
</TEI>
<inist><standard h6="B"><pA><fA01 i1="01" i2="1"><s0>0302-9743</s0>
</fA01>
<fA05><s2>2116</s2>
</fA05>
<fA08 i1="01" i2="1" l="ENG"><s1>A new method based on context for combining statistical language models</s1>
</fA08>
<fA09 i1="01" i2="1" l="ENG"><s1>CONTEXT 2001 : modeling and using context : Dundee, 27-30 July 2001</s1>
</fA09>
<fA11 i1="01" i2="1"><s1>LANGLOIS (David)</s1>
</fA11>
<fA11 i1="02" i2="1"><s1>SMAÏLI (Kamel)</s1>
</fA11>
<fA11 i1="03" i2="1"><s1>HATON (Jean-Paul)</s1>
</fA11>
<fA12 i1="01" i2="1"><s1>AKMAN (Varol)</s1>
<s9>ed.</s9>
</fA12>
<fA12 i1="02" i2="1"><s1>BOUQUET (Paolo)</s1>
<s9>ed.</s9>
</fA12>
<fA12 i1="03" i2="1"><s1>THOMASON (Richmond)</s1>
<s9>ed.</s9>
</fA12>
<fA12 i1="04" i2="1"><s1>YOUNG (Roger A.)</s1>
<s9>ed.</s9>
</fA12>
<fA14 i1="01"><s1>LORIA Laboratory, Campus Scientifique BP 239</s1>
<s2>54506 Vandœuvre-Lès-Nancy</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</fA14>
<fA20><s1>235-247</s1>
</fA20>
<fA21><s1>2001</s1>
</fA21>
<fA23 i1="01"><s0>ENG</s0>
</fA23>
<fA26 i1="01"><s0>3-540-42379-6</s0>
</fA26>
<fA43 i1="01"><s1>INIST</s1>
<s2>16343</s2>
<s5>354000092486680180</s5>
</fA43>
<fA44><s0>0000</s0>
<s1>© 2001 INIST-CNRS. All rights reserved.</s1>
</fA44>
<fA45><s0>15 ref.</s0>
</fA45>
<fA47 i1="01" i2="1"><s0>01-0381622</s0>
</fA47>
<fA60><s1>P</s1>
<s2>C</s2>
</fA60>
<fA64 i1="01" i2="1"><s0>Lecture notes in computer science</s0>
</fA64>
<fA66 i1="01"><s0>DEU</s0>
</fA66>
<fA66 i1="02"><s0>USA</s0>
</fA66>
<fC01 i1="01" l="ENG"><s0>In this paper we propose a new method to extract from a corpus the histories for which a given language model is better than another one. The decision is based on a measure stemmed from perplexity. This measure allows, for a given history, to compare two language models, and then to choose the best one for this history. Using this principle, and with a 20K vocabulary words, we combined two language models: a bigram and a distant bigram. The contribution of a distant bigram is significant and outperforms a bigram model by 7.5%. Moreover, the performance in Shannon game are improved. We show through this article that we proposed a cheaper framework in comparison to the maximum entropy principle, for combining language models. In addition, the selected histories for which a model is better than another one, have been collected and studied. Almost, all of them are beginnings of very frequently used French phrases. Finally, by using this principle, we achieve a better trigram model in terms of parameters and perplexity. This model is a combination of a bigram and a trigram based on a selected history.</s0>
</fC01>
<fC02 i1="01" i2="X"><s0>001D02C04</s0>
</fC02>
<fC03 i1="01" i2="X" l="FRE"><s0>Méthode entropie maximum</s0>
<s5>01</s5>
</fC03>
<fC03 i1="01" i2="X" l="ENG"><s0>Method of maximum entropy</s0>
<s5>01</s5>
</fC03>
<fC03 i1="01" i2="X" l="SPA"><s0>Método entropía máxima</s0>
<s5>01</s5>
</fC03>
<fC03 i1="02" i2="X" l="FRE"><s0>Principe maximum</s0>
<s5>02</s5>
</fC03>
<fC03 i1="02" i2="X" l="ENG"><s0>Maximum principle</s0>
<s5>02</s5>
</fC03>
<fC03 i1="02" i2="X" l="SPA"><s0>Principio máximo</s0>
<s5>02</s5>
</fC03>
<fC03 i1="03" i2="X" l="FRE"><s0>Théorie Shannon</s0>
<s5>03</s5>
</fC03>
<fC03 i1="03" i2="X" l="ENG"><s0>Shannon theory</s0>
<s5>03</s5>
</fC03>
<fC03 i1="03" i2="X" l="SPA"><s0>Teoría Shannon</s0>
<s5>03</s5>
</fC03>
<fC03 i1="04" i2="X" l="FRE"><s0>Vocabulaire</s0>
<s5>04</s5>
</fC03>
<fC03 i1="04" i2="X" l="ENG"><s0>Vocabulary</s0>
<s5>04</s5>
</fC03>
<fC03 i1="04" i2="X" l="SPA"><s0>Vocabulario</s0>
<s5>04</s5>
</fC03>
<fC03 i1="05" i2="X" l="FRE"><s0>Mot</s0>
<s5>05</s5>
</fC03>
<fC03 i1="05" i2="X" l="ENG"><s0>Word</s0>
<s5>05</s5>
</fC03>
<fC03 i1="05" i2="X" l="SPA"><s0>Palabra</s0>
<s5>05</s5>
</fC03>
<fC03 i1="06" i2="X" l="FRE"><s0>Evaluation performance</s0>
<s5>06</s5>
</fC03>
<fC03 i1="06" i2="X" l="ENG"><s0>Performance evaluation</s0>
<s5>06</s5>
</fC03>
<fC03 i1="06" i2="X" l="SPA"><s0>Evaluación prestación</s0>
<s5>06</s5>
</fC03>
<fC03 i1="07" i2="X" l="FRE"><s0>Modélisation</s0>
<s5>07</s5>
</fC03>
<fC03 i1="07" i2="X" l="ENG"><s0>Modeling</s0>
<s5>07</s5>
</fC03>
<fC03 i1="07" i2="X" l="SPA"><s0>Modelización</s0>
<s5>07</s5>
</fC03>
<fC03 i1="08" i2="X" l="FRE"><s0>Méthode statistique</s0>
<s5>08</s5>
</fC03>
<fC03 i1="08" i2="X" l="ENG"><s0>Statistical method</s0>
<s5>08</s5>
</fC03>
<fC03 i1="08" i2="X" l="SPA"><s0>Método estadístico</s0>
<s5>08</s5>
</fC03>
<fC03 i1="09" i2="X" l="FRE"><s0>Reconnaissance parole</s0>
<s5>09</s5>
</fC03>
<fC03 i1="09" i2="X" l="ENG"><s0>Speech recognition</s0>
<s5>09</s5>
</fC03>
<fC03 i1="09" i2="X" l="SPA"><s0>Reconocimiento palabra</s0>
<s5>09</s5>
</fC03>
<fN21><s1>267</s1>
</fN21>
</pA>
<pR><fA30 i1="01" i2="1" l="ENG"><s1>Modeling and using context. International and interdisciplinary conference</s1>
<s2>3</s2>
<s3>Dundee GBR</s3>
<s4>2001-07-27</s4>
</fA30>
</pR>
</standard>
<server><NO>PASCAL 01-0381622 INIST</NO>
<ET>A new method based on context for combining statistical language models</ET>
<AU>LANGLOIS (David); SMAÏLI (Kamel); HATON (Jean-Paul); AKMAN (Varol); BOUQUET (Paolo); THOMASON (Richmond); YOUNG (Roger A.)</AU>
<AF>LORIA Laboratory, Campus Scientifique BP 239/54506 Vandœuvre-Lès-Nancy/France (1 aut., 2 aut., 3 aut.)</AF>
<DT>Publication en série; Congrès; Niveau analytique</DT>
<SO>Lecture notes in computer science; ISSN 0302-9743; Allemagne; Da. 2001; Vol. 2116; Pp. 235-247; Bibl. 15 ref.</SO>
<LA>Anglais</LA>
<EA>In this paper we propose a new method to extract from a corpus the histories for which a given language model is better than another one. The decision is based on a measure stemmed from perplexity. This measure allows, for a given history, to compare two language models, and then to choose the best one for this history. Using this principle, and with a 20K vocabulary words, we combined two language models: a bigram and a distant bigram. The contribution of a distant bigram is significant and outperforms a bigram model by 7.5%. Moreover, the performance in Shannon game are improved. We show through this article that we proposed a cheaper framework in comparison to the maximum entropy principle, for combining language models. In addition, the selected histories for which a model is better than another one, have been collected and studied. Almost, all of them are beginnings of very frequently used French phrases. Finally, by using this principle, we achieve a better trigram model in terms of parameters and perplexity. This model is a combination of a bigram and a trigram based on a selected history.</EA>
<CC>001D02C04</CC>
<FD>Méthode entropie maximum; Principe maximum; Théorie Shannon; Vocabulaire; Mot; Evaluation performance; Modélisation; Méthode statistique; Reconnaissance parole</FD>
<ED>Method of maximum entropy; Maximum principle; Shannon theory; Vocabulary; Word; Performance evaluation; Modeling; Statistical method; Speech recognition</ED>
<SD>Método entropía máxima; Principio máximo; Teoría Shannon; Vocabulario; Palabra; Evaluación prestación; Modelización; Método estadístico; Reconocimiento palabra</SD>
<LO>INIST-16343.354000092486680180</LO>
<ID>01-0381622</ID>
</server>
</inist>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/PascalFrancis/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000938 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/PascalFrancis/Corpus/biblio.hfd -nk 000938 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Lorraine
|area= InforLorV4
|flux= PascalFrancis
|étape= Corpus
|type= RBID
|clé= Pascal:01-0381622
|texte= A new method based on context for combining statistical language models
}}
| This area was generated with Dilib version V0.6.33. Data generation: Mon Jun 10 21:56:28 2019. Site generation: Fri Feb 25 15:29:27 2022 | ![](Common/icons/LogoDilib.gif) |