Learning to Rank for
Information Retrieval
and Natural Language
Processing
Second Edition
Hang Li
LI LEARNING TO RANK FOR INFORMATION RETRIEVAL AND NATURAL LANGUAGE PROCESSING: SECOND EDITION MORGAN & CLAYPOOL
Learning to Rank for Information Retrieval
and Natural Language Processing,
Second Edition
Hang Li, Huawei Technologies
Learning to rank refers to machine learning techniques for training a model in a ranking task. Learning to rank is
useful for many applications in information retrieval, natural language processing, and data mining. Intensive studies
have been conducted on its problems recently, and signicant progress has been made. This lecture gives an intro-
duction to the area including the fundamental problems, major approaches, theories, applications, and future work.
The author begins by showing that various ranking problems in information retrieval and natural language pro-
cessing can be formalized as two basic ranking tasks, namely ranking creation (or simply ranking) and ranking ag-
gregation. In ranking creation, given a request, one wants to generate a ranking list of oerings based on the features
derived from the request and the oerings. In ranking aggregation, given a request, as well as a number of ranking
lists of oerings, one wants to generate a new ranking list of the oerings.
Ranking creation (or ranking) is the major problem in learning to rank. It is usually formalized as a supervised
learning task. The author gives detailed explanations on learning for ranking creation and ranking aggregation, in-
cluding training and testing, evaluation, feature creation, and major approaches. Many methods have been proposed
for ranking creation. The methods can be categorized as the pointwise, pairwise, and listwise approaches according
to the loss functions they employ. They can also be categorized according to the techniques they employ, such as the
SVM based, Boosting based, and Neural Network based approaches.
The author also introduces some popular learning to rank methods in details. These include: PRank, OC SVM,
McRank, Ranking SVM, IR SVM, GBRank, RankNet, ListNet & ListMLE, AdaRank, SVM MAP, SoftRank,
LambdaRank, LambdaMART, Borda Count, Markov Chain, and CRanking.
The author explains several example applications of learning to rank including web search, collaborative ltering,
denition search, keyphrase extraction, query dependent summarization, and re-ranking in machine translation.
A formulation of learning for ranking creation is given in the statistical learning framework. Ongoing and future
research directions for learning to rank are also discussed.
ISBN: 978-1-62705-584-0
9 781627 055840
90000
Series Editor: Graeme Hirst, University of Toronto
SyntheSiS LectureS on
human Language technoLogieS
ABOUT SYNTHESIS
is volume is a printed version of a work that appears in the Synthesis
Digital
Library of Engineering and Computer Science. Synthesis Lectures
provide concise,
original presentations of important research and development
topics, published quickly,
in digital and print formats. For more information visit www.morganclaypool.com
www.morganclaypool.com
MORGAN
&
CLAYPOOL PUBLISHERS
Series ISSN: 1947-4040
SyntheSiS LectureS on
h
uman Language technoLogieS
Graeme Hirst, Series Editor
MORGAN
&
CLAYPOOL PUBLISHERS
Learning to Rank for
Information Retrieval
and Natural Language
Processing
Second Edition
Hang Li
LI LEARNING TO RANK FOR INFORMATION RETRIEVAL AND NATURAL LANGUAGE PROCESSING: SECOND EDITION MORGAN & CLAYPOOL
Learning to Rank for Information Retrieval
and Natural Language Processing,
Second Edition
Hang Li, Huawei Technologies
Learning to rank refers to machine learning techniques for training a model in a ranking task. Learning to rank is
useful for many applications in information retrieval, natural language processing, and data mining. Intensive studies
have been conducted on its problems recently, and signicant progress has been made. This lecture gives an intro-
duction to the area including the fundamental problems, major approaches, theories, applications, and future work.
The author begins by showing that various ranking problems in information retrieval and natural language pro-
cessing can be formalized as two basic ranking tasks, namely ranking creation (or simply ranking) and ranking ag-
gregation. In ranking creation, given a request, one wants to generate a ranking list of oerings based on the features
derived from the request and the oerings. In ranking aggregation, given a request, as well as a number of ranking
lists of oerings, one wants to generate a new ranking list of the oerings.
Ranking creation (or ranking) is the major problem in learning to rank. It is usually formalized as a supervised
learning task. The author gives detailed explanations on learning for ranking creation and ranking aggregation, in-
cluding training and testing, evaluation, feature creation, and major approaches. Many methods have been proposed
for ranking creation. The methods can be categorized as the pointwise, pairwise, and listwise approaches according
to the loss functions they employ. They can also be categorized according to the techniques they employ, such as the
SVM based, Boosting based, and Neural Network based approaches.
The author also introduces some popular learning to rank methods in details. These include: PRank, OC SVM,
McRank, Ranking SVM, IR SVM, GBRank, RankNet, ListNet & ListMLE, AdaRank, SVM MAP, SoftRank,
LambdaRank, LambdaMART, Borda Count, Markov Chain, and CRanking.
The author explains several example applications of learning to rank including web search, collaborative ltering,
denition search, keyphrase extraction, query dependent summarization, and re-ranking in machine translation.
A formulation of learning for ranking creation is given in the statistical learning framework. Ongoing and future
research directions for learning to rank are also discussed.
ISBN: 978-1-62705-584-0
9 781627 055840
90000
Series Editor: Graeme Hirst, University of Toronto
SyntheSiS LectureS on
human Language technoLogieS
ABOUT SYNTHESIS
is volume is a printed version of a work that appears in the Synthesis
Digital
Library of Engineering and Computer Science. Synthesis Lectures
provide concise,
original presentations of important research and development
topics, published quickly,
in digital and print formats. For more information visit www.morganclaypool.com
www.morganclaypool.com
MORGAN
&
CLAYPOOL PUBLISHERS
Series ISSN: 1947-4040
SyntheSiS LectureS on
human Language technoLogieS
Graeme Hirst, Series Editor
MORGAN
&
CLAYPOOL PUBLISHERS
Learning to Rank for
Information Retrieval
and Natural Language
Processing
Second Edition
Hang Li
LI LEARNING TO RANK FOR INFORMATION RETRIEVAL AND NATURAL LANGUAGE PROCESSING: SECOND EDITION MORGAN & CLAYPOOL
Learning to Rank for Information Retrieval
and Natural Language Processing,
Second Edition
Hang Li, Huawei Technologies
Learning to rank refers to machine learning techniques for training a model in a ranking task. Learning to rank is
useful for many applications in information retrieval, natural language processing, and data mining. Intensive studies
have been conducted on its problems recently, and signicant progress has been made. This lecture gives an intro-
duction to the area including the fundamental problems, major approaches, theories, applications, and future work.
The author begins by showing that various ranking problems in information retrieval and natural language pro-
cessing can be formalized as two basic ranking tasks, namely ranking creation (or simply ranking) and ranking ag-
gregation. In ranking creation, given a request, one wants to generate a ranking list of oerings based on the features
derived from the request and the oerings. In ranking aggregation, given a request, as well as a number of ranking
lists of oerings, one wants to generate a new ranking list of the oerings.
Ranking creation (or ranking) is the major problem in learning to rank. It is usually formalized as a supervised
learning task. The author gives detailed explanations on learning for ranking creation and ranking aggregation, in-
cluding training and testing, evaluation, feature creation, and major approaches. Many methods have been proposed
for ranking creation. The methods can be categorized as the pointwise, pairwise, and listwise approaches according
to the loss functions they employ. They can also be categorized according to the techniques they employ, such as the
SVM based, Boosting based, and Neural Network based approaches.
The author also introduces some popular learning to rank methods in details. These include: PRank, OC SVM,
McRank, Ranking SVM, IR SVM, GBRank, RankNet, ListNet & ListMLE, AdaRank, SVM MAP, SoftRank,
LambdaRank, LambdaMART, Borda Count, Markov Chain, and CRanking.
The author explains several example applications of learning to rank including web search, collaborative ltering,
denition search, keyphrase extraction, query dependent summarization, and re-ranking in machine translation.
A formulation of learning for ranking creation is given in the statistical learning framework. Ongoing and future
research directions for learning to rank are also discussed.
ISBN: 978-1-62705-584-0
9 781627 055840
90000
Series Editor: Graeme Hirst, University of Toronto
SyntheSiS LectureS on
h
uman Language technoLogieS
ABOUT SYNTHESIS
is volume is a printed version of a work that appears in the Synthesis
Digital
Library of Engineering and Computer Science. Synthesis Lectures
provide concise,
original presentations of important research and development
topics, published quickly,
in digital and print formats. For more information visit www.morganclaypool.com
www.morganclaypool.com
MORGAN
&
CLAYPOOL PUBLISHERS
Series ISSN: 1947-4040
SyntheSiS LectureS on
human Language technoLogieS
Graeme Hirst, Series Editor
MORGAN
&
CLAYPOOL PUBLISHERS
剩余122页未读，继续阅读
评论0