A Decision Tree Based Recommender System

Amir Gershman; Amnon Meisels; Karl Heinz Lüke; Lior Rokach; Alon Schclar; Arnon Sturm

A Decision Tree Based Recommender System

Amir Gershman, Amnon Meisels, Karl Heinz Lüke, Lior Rokach, Alon Schclar, Arnon Sturm

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

21 Scopus citations

Abstract

A new method for decision-tree-based recommender systems is proposed. The proposed method includes two new major innovations. First, the decision tree produces lists of recommended items at its leaf nodes, instead of single items. This leads to reduced amount of search, when using the tree to compile a recommendation list for a user and consequently enables a scaling of the recommendation system. The second major contribution of the paper is the splitting method for constructing the decision tree. Splitting is based on a new criterion - the least probable intersection size. The new criterion computes the probability for getting the intersection for each potential split in a random split and selects the split that generates the least probable size of intersection. The proposed decision tree based recommendation system was evaluated on a large sample of the MovieLens dataset and is shown to outperform the quality of recommendations produced by the well known information gain splitting criterion.

Original language	English
Title of host publication	10th International Conference on Innovative Internet Community Services, I2CS 2010 - Jubilee Edition 2010
Subtitle of host publication	Proceedings
Editors	Gerald Eichler, Peter Kropf, Ulrike Lechner, Phayung Meesad, Herwig Unger
Publisher	Gesellschaft fur Informatik (GI)
Pages	170-179
Number of pages	10
ISBN (Electronic)	9783885792598
State	Published - 2010
Externally published	Yes
Event	10th International Conference on Innovative Internet Community Services, I2CS 2010 - Bangkok, Thailand Duration: 3 Jun 2010 → 5 Jun 2010

Publication series

Name	Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)
Volume	P-165
ISSN (Print)	1617-5468
ISSN (Electronic)	2944-7682

Conference

Conference	10th International Conference on Innovative Internet Community Services, I2CS 2010
Country/Territory	Thailand
City	Bangkok
Period	3/06/10 → 5/06/10

Bibliographical note

Publisher Copyright:
© 2010 Gesellschaft fur Informatik (GI). All rights reserved.

Funding

The project was funded and managed by Deutsche Telekom Laboratories as part of the Context-Aware Service Offering and Usage (CASOU) project.

Funders	Funder number
Context-Aware Service Offering and Usage
Deutsche Telekom Laboratories

Cite this

Gershman, A., Meisels, A., Lüke, K. H., Rokach, L., Schclar, A., & Sturm, A. (2010). A Decision Tree Based Recommender System. In G. Eichler, P. Kropf, U. Lechner, P. Meesad, & H. Unger (Eds.), 10th International Conference on Innovative Internet Community Services, I2CS 2010 - Jubilee Edition 2010: Proceedings (pp. 170-179). (Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI); Vol. P-165). Gesellschaft fur Informatik (GI).

Gershman, Amir ; Meisels, Amnon ; Lüke, Karl Heinz et al. / A Decision Tree Based Recommender System. 10th International Conference on Innovative Internet Community Services, I2CS 2010 - Jubilee Edition 2010: Proceedings. editor / Gerald Eichler ; Peter Kropf ; Ulrike Lechner ; Phayung Meesad ; Herwig Unger. Gesellschaft fur Informatik (GI), 2010. pp. 170-179 (Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)).

@inproceedings{fef24641186c4439918012388ab24168,

title = "A Decision Tree Based Recommender System",

abstract = "A new method for decision-tree-based recommender systems is proposed. The proposed method includes two new major innovations. First, the decision tree produces lists of recommended items at its leaf nodes, instead of single items. This leads to reduced amount of search, when using the tree to compile a recommendation list for a user and consequently enables a scaling of the recommendation system. The second major contribution of the paper is the splitting method for constructing the decision tree. Splitting is based on a new criterion - the least probable intersection size. The new criterion computes the probability for getting the intersection for each potential split in a random split and selects the split that generates the least probable size of intersection. The proposed decision tree based recommendation system was evaluated on a large sample of the MovieLens dataset and is shown to outperform the quality of recommendations produced by the well known information gain splitting criterion.",

author = "Amir Gershman and Amnon Meisels and L{\"u}ke, {Karl Heinz} and Lior Rokach and Alon Schclar and Arnon Sturm",

note = "Publisher Copyright: {\textcopyright} 2010 Gesellschaft fur Informatik (GI). All rights reserved.; 10th International Conference on Innovative Internet Community Services, I2CS 2010 ; Conference date: 03-06-2010 Through 05-06-2010",

year = "2010",

language = "אנגלית",

series = "Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)",

publisher = "Gesellschaft fur Informatik (GI)",

pages = "170--179",

editor = "Gerald Eichler and Peter Kropf and Ulrike Lechner and Phayung Meesad and Herwig Unger",

booktitle = "10th International Conference on Innovative Internet Community Services, I2CS 2010 - Jubilee Edition 2010",

address = "גרמניה",

}

Gershman, A, Meisels, A, Lüke, KH, Rokach, L, Schclar, A & Sturm, A 2010, A Decision Tree Based Recommender System. in G Eichler, P Kropf, U Lechner, P Meesad & H Unger (eds), 10th International Conference on Innovative Internet Community Services, I2CS 2010 - Jubilee Edition 2010: Proceedings. Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI), vol. P-165, Gesellschaft fur Informatik (GI), pp. 170-179, 10th International Conference on Innovative Internet Community Services, I2CS 2010, Bangkok, Thailand, 3/06/10.

A Decision Tree Based Recommender System. / Gershman, Amir; Meisels, Amnon; Lüke, Karl Heinz et al.
10th International Conference on Innovative Internet Community Services, I2CS 2010 - Jubilee Edition 2010: Proceedings. ed. / Gerald Eichler; Peter Kropf; Ulrike Lechner; Phayung Meesad; Herwig Unger. Gesellschaft fur Informatik (GI), 2010. p. 170-179 (Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI); Vol. P-165).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A Decision Tree Based Recommender System

AU - Gershman, Amir

AU - Meisels, Amnon

AU - Lüke, Karl Heinz

AU - Rokach, Lior

AU - Schclar, Alon

AU - Sturm, Arnon

PY - 2010

Y1 - 2010

N2 - A new method for decision-tree-based recommender systems is proposed. The proposed method includes two new major innovations. First, the decision tree produces lists of recommended items at its leaf nodes, instead of single items. This leads to reduced amount of search, when using the tree to compile a recommendation list for a user and consequently enables a scaling of the recommendation system. The second major contribution of the paper is the splitting method for constructing the decision tree. Splitting is based on a new criterion - the least probable intersection size. The new criterion computes the probability for getting the intersection for each potential split in a random split and selects the split that generates the least probable size of intersection. The proposed decision tree based recommendation system was evaluated on a large sample of the MovieLens dataset and is shown to outperform the quality of recommendations produced by the well known information gain splitting criterion.

AB - A new method for decision-tree-based recommender systems is proposed. The proposed method includes two new major innovations. First, the decision tree produces lists of recommended items at its leaf nodes, instead of single items. This leads to reduced amount of search, when using the tree to compile a recommendation list for a user and consequently enables a scaling of the recommendation system. The second major contribution of the paper is the splitting method for constructing the decision tree. Splitting is based on a new criterion - the least probable intersection size. The new criterion computes the probability for getting the intersection for each potential split in a random split and selects the split that generates the least probable size of intersection. The proposed decision tree based recommendation system was evaluated on a large sample of the MovieLens dataset and is shown to outperform the quality of recommendations produced by the well known information gain splitting criterion.

UR - http://www.scopus.com/inward/record.url?scp=84926659728&partnerID=8YFLogxK

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:84926659728

T3 - Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)

SP - 170

EP - 179

BT - 10th International Conference on Innovative Internet Community Services, I2CS 2010 - Jubilee Edition 2010

A2 - Eichler, Gerald

A2 - Kropf, Peter

A2 - Lechner, Ulrike

A2 - Meesad, Phayung

A2 - Unger, Herwig

PB - Gesellschaft fur Informatik (GI)

T2 - 10th International Conference on Innovative Internet Community Services, I2CS 2010

Y2 - 3 June 2010 through 5 June 2010

ER -

Gershman A, Meisels A, Lüke KH, Rokach L, Schclar A, Sturm A. A Decision Tree Based Recommender System. In Eichler G, Kropf P, Lechner U, Meesad P, Unger H, editors, 10th International Conference on Innovative Internet Community Services, I2CS 2010 - Jubilee Edition 2010: Proceedings. Gesellschaft fur Informatik (GI). 2010. p. 170-179. (Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)).

A Decision Tree Based Recommender System

Abstract

Publication series

Conference

Bibliographical note

Funding

Other files and links

Fingerprint

Cite this