PerRank: Personalized rank retrieval with categorical and numerical attributes

Sangkyum Kim, Jaebum Kim, Younhee Ko, Seung Won Hwang, Jiawei Han

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Ranking has been popularly used for intelligent data retrieval in both database and machine learning communities. Recently, there were studies on integrating these two approaches to support soft queries, based on a user's sense of relevance and preference, for ranking with numerical attributes. However, in real life, it is desirable to use categorical attributes together with numerical ones in ranking. For example, when buying a car, categorical attributes, such as make, model, color, and equipments, are considered as significant factors as numerical attributes, such as price and year. Meanwhile, users often do not have sufficient domain knowledge at formulating an effective selection query over categories, whereas rank formulation is even more challenging as categories have no inherent ordering. In this paper, we propose a framework PerRank (Personalized Ranking with Categorical and Numerical Attributes) to support personalized ranking with both categorical and numerical attributes for soft queries. For an efficient computation, we developed an algorithm CAC (Clustering-based Attribute Construction) which makes use of a clustering method. Extensive experiments show CAC is effective and efficient at supporting ranking with both categorical and numerical attributes for soft queries.

Original languageEnglish
Title of host publicationProceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008
Pages270-277
Number of pages8
DOIs
Publication statusPublished - 2008 Sep 22
Event9th International Conference on Web-Age Information Management, WAIM 2008 - Zhangjiajie, China
Duration: 2008 Jul 202008 Jul 22

Publication series

NameProceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008

Other

Other9th International Conference on Web-Age Information Management, WAIM 2008
CountryChina
CityZhangjiajie
Period08/7/2008/7/22

Fingerprint

Clustering algorithms
Learning systems
Railroad cars
Color
Experiments
Ranking
Query
Clustering

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Information Systems and Management

Cite this

Kim, S., Kim, J., Ko, Y., Hwang, S. W., & Han, J. (2008). PerRank: Personalized rank retrieval with categorical and numerical attributes. In Proceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008 (pp. 270-277). [4597024] (Proceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008). https://doi.org/10.1109/WAIM.2008.88
Kim, Sangkyum ; Kim, Jaebum ; Ko, Younhee ; Hwang, Seung Won ; Han, Jiawei. / PerRank : Personalized rank retrieval with categorical and numerical attributes. Proceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008. 2008. pp. 270-277 (Proceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008).
@inproceedings{179c3ff2e62744bdad8bf36d1960464f,
title = "PerRank: Personalized rank retrieval with categorical and numerical attributes",
abstract = "Ranking has been popularly used for intelligent data retrieval in both database and machine learning communities. Recently, there were studies on integrating these two approaches to support soft queries, based on a user's sense of relevance and preference, for ranking with numerical attributes. However, in real life, it is desirable to use categorical attributes together with numerical ones in ranking. For example, when buying a car, categorical attributes, such as make, model, color, and equipments, are considered as significant factors as numerical attributes, such as price and year. Meanwhile, users often do not have sufficient domain knowledge at formulating an effective selection query over categories, whereas rank formulation is even more challenging as categories have no inherent ordering. In this paper, we propose a framework PerRank (Personalized Ranking with Categorical and Numerical Attributes) to support personalized ranking with both categorical and numerical attributes for soft queries. For an efficient computation, we developed an algorithm CAC (Clustering-based Attribute Construction) which makes use of a clustering method. Extensive experiments show CAC is effective and efficient at supporting ranking with both categorical and numerical attributes for soft queries.",
author = "Sangkyum Kim and Jaebum Kim and Younhee Ko and Hwang, {Seung Won} and Jiawei Han",
year = "2008",
month = "9",
day = "22",
doi = "10.1109/WAIM.2008.88",
language = "English",
isbn = "9780769531854",
series = "Proceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008",
pages = "270--277",
booktitle = "Proceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008",

}

Kim, S, Kim, J, Ko, Y, Hwang, SW & Han, J 2008, PerRank: Personalized rank retrieval with categorical and numerical attributes. in Proceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008., 4597024, Proceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008, pp. 270-277, 9th International Conference on Web-Age Information Management, WAIM 2008, Zhangjiajie, China, 08/7/20. https://doi.org/10.1109/WAIM.2008.88

PerRank : Personalized rank retrieval with categorical and numerical attributes. / Kim, Sangkyum; Kim, Jaebum; Ko, Younhee; Hwang, Seung Won; Han, Jiawei.

Proceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008. 2008. p. 270-277 4597024 (Proceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - PerRank

T2 - Personalized rank retrieval with categorical and numerical attributes

AU - Kim, Sangkyum

AU - Kim, Jaebum

AU - Ko, Younhee

AU - Hwang, Seung Won

AU - Han, Jiawei

PY - 2008/9/22

Y1 - 2008/9/22

N2 - Ranking has been popularly used for intelligent data retrieval in both database and machine learning communities. Recently, there were studies on integrating these two approaches to support soft queries, based on a user's sense of relevance and preference, for ranking with numerical attributes. However, in real life, it is desirable to use categorical attributes together with numerical ones in ranking. For example, when buying a car, categorical attributes, such as make, model, color, and equipments, are considered as significant factors as numerical attributes, such as price and year. Meanwhile, users often do not have sufficient domain knowledge at formulating an effective selection query over categories, whereas rank formulation is even more challenging as categories have no inherent ordering. In this paper, we propose a framework PerRank (Personalized Ranking with Categorical and Numerical Attributes) to support personalized ranking with both categorical and numerical attributes for soft queries. For an efficient computation, we developed an algorithm CAC (Clustering-based Attribute Construction) which makes use of a clustering method. Extensive experiments show CAC is effective and efficient at supporting ranking with both categorical and numerical attributes for soft queries.

AB - Ranking has been popularly used for intelligent data retrieval in both database and machine learning communities. Recently, there were studies on integrating these two approaches to support soft queries, based on a user's sense of relevance and preference, for ranking with numerical attributes. However, in real life, it is desirable to use categorical attributes together with numerical ones in ranking. For example, when buying a car, categorical attributes, such as make, model, color, and equipments, are considered as significant factors as numerical attributes, such as price and year. Meanwhile, users often do not have sufficient domain knowledge at formulating an effective selection query over categories, whereas rank formulation is even more challenging as categories have no inherent ordering. In this paper, we propose a framework PerRank (Personalized Ranking with Categorical and Numerical Attributes) to support personalized ranking with both categorical and numerical attributes for soft queries. For an efficient computation, we developed an algorithm CAC (Clustering-based Attribute Construction) which makes use of a clustering method. Extensive experiments show CAC is effective and efficient at supporting ranking with both categorical and numerical attributes for soft queries.

UR - http://www.scopus.com/inward/record.url?scp=51849136548&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51849136548&partnerID=8YFLogxK

U2 - 10.1109/WAIM.2008.88

DO - 10.1109/WAIM.2008.88

M3 - Conference contribution

AN - SCOPUS:51849136548

SN - 9780769531854

T3 - Proceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008

SP - 270

EP - 277

BT - Proceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008

ER -

Kim S, Kim J, Ko Y, Hwang SW, Han J. PerRank: Personalized rank retrieval with categorical and numerical attributes. In Proceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008. 2008. p. 270-277. 4597024. (Proceedings - The 9th International Conference on Web-Age Information Management, WAIM 2008). https://doi.org/10.1109/WAIM.2008.88