Serendipitous recommendations through ontology-based contextual pre-filtering

Aleksandra Karpus; Iacopo Vagliano; Krzysztof Goczyła

doi:https://doi.org/10.1007/978-3-319-58274-0_21

Serendipitous recommendations through ontology-based contextual pre-filtering

Aleksandra Karpus, Iacopo Vagliano, Krzysztof Goczyła

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

10 Citations (Scopus)

Abstract

Context-aware Recommender Systems aim to provide users with better recommendations for their current situation. Although evaluations of recommender systems often focus on accuracy, it is not the only important aspect. Often recommendations are overspecialized, i.e. all of the same kind. To deal with this problem, other properties can be considered, such as serendipity. In this paper, we study how an ontology-based and context-aware pre-filtering technique which can be combined with existing recommendation algorithm performs in ranking tasks. We also investigate the impact of our method on the serendipity of the recommendations. We evaluated our approach through an offline study which showed that when used with well-known recommendation algorithms it can improve the accuracy and serendipity.

Original language	English
Title of host publication	Beyond Databases, Architectures and Structures
Subtitle of host publication	Towards Efficient Solutions for Data Analysis and Knowledge Representation - 13th International Conference, BDAS 2017, Proceedings
Editors	Stanislaw Kozielski, Dariusz Mrozek, Pawel Kasprowski, Bozena Malysiak-Mrozek, Daniel Kostrzewa
Publisher	Springer - Verlag
Pages	246-259
Number of pages	14
ISBN (Print)	9783319582733
DOIs	https://doi.org/10.1007/978-3-319-58274-0_21
Publication status	Published - 2017
Externally published	Yes
Event	13th International Conference on Beyond Databases, Architectures and Structures, BDAS 2017 - Ustron, Poland Duration: 30 May 2017 → 2 Jun 2017

Publication series

Name	Communications in Computer and Information Science
Volume	716

Conference

Conference	13th International Conference on Beyond Databases, Architectures and Structures, BDAS 2017
Country/Territory	Poland
City	Ustron
Period	30/05/2017 → 2/06/2017

Keywords

Context-awareness
Ontologies
Recommender systems
Serendipity

Access to Document

https://doi.org/10.1007/978-3-319-58274-0_21

Cite this

Karpus, A., Vagliano, I., & Goczyła, K. (2017). Serendipitous recommendations through ontology-based contextual pre-filtering. In S. Kozielski, D. Mrozek, P. Kasprowski, B. Malysiak-Mrozek, & D. Kostrzewa (Eds.), Beyond Databases, Architectures and Structures: Towards Efficient Solutions for Data Analysis and Knowledge Representation - 13th International Conference, BDAS 2017, Proceedings (pp. 246-259). (Communications in Computer and Information Science; Vol. 716). Springer - Verlag. https://doi.org/10.1007/978-3-319-58274-0_21

Karpus, Aleksandra ; Vagliano, Iacopo ; Goczyła, Krzysztof. / Serendipitous recommendations through ontology-based contextual pre-filtering. Beyond Databases, Architectures and Structures: Towards Efficient Solutions for Data Analysis and Knowledge Representation - 13th International Conference, BDAS 2017, Proceedings. editor / Stanislaw Kozielski ; Dariusz Mrozek ; Pawel Kasprowski ; Bozena Malysiak-Mrozek ; Daniel Kostrzewa. Springer - Verlag, 2017. pp. 246-259 (Communications in Computer and Information Science).

@inproceedings{da8f760eee7749358a754c6be0b9af51,

title = "Serendipitous recommendations through ontology-based contextual pre-filtering",

abstract = "Context-aware Recommender Systems aim to provide users with better recommendations for their current situation. Although evaluations of recommender systems often focus on accuracy, it is not the only important aspect. Often recommendations are overspecialized, i.e. all of the same kind. To deal with this problem, other properties can be considered, such as serendipity. In this paper, we study how an ontology-based and context-aware pre-filtering technique which can be combined with existing recommendation algorithm performs in ranking tasks. We also investigate the impact of our method on the serendipity of the recommendations. We evaluated our approach through an offline study which showed that when used with well-known recommendation algorithms it can improve the accuracy and serendipity.",

keywords = "Context-awareness, Ontologies, Recommender systems, Serendipity",

author = "Aleksandra Karpus and Iacopo Vagliano and Krzysztof Goczy{\l}a",

note = "Funding Information: The second author was supported by a fellowship from TIM. Publisher Copyright: {\textcopyright} Springer International Publishing AG 2017.; 13th International Conference on Beyond Databases, Architectures and Structures, BDAS 2017 ; Conference date: 30-05-2017 Through 02-06-2017",

year = "2017",

doi = "https://doi.org/10.1007/978-3-319-58274-0_21",

language = "English",

isbn = "9783319582733",

series = "Communications in Computer and Information Science",

publisher = "Springer - Verlag",

pages = "246--259",

editor = "Stanislaw Kozielski and Dariusz Mrozek and Pawel Kasprowski and Bozena Malysiak-Mrozek and Daniel Kostrzewa",

booktitle = "Beyond Databases, Architectures and Structures",

}

Karpus, A, Vagliano, I & Goczyła, K 2017, Serendipitous recommendations through ontology-based contextual pre-filtering. in S Kozielski, D Mrozek, P Kasprowski, B Malysiak-Mrozek & D Kostrzewa (eds), Beyond Databases, Architectures and Structures: Towards Efficient Solutions for Data Analysis and Knowledge Representation - 13th International Conference, BDAS 2017, Proceedings. Communications in Computer and Information Science, vol. 716, Springer - Verlag, pp. 246-259, 13th International Conference on Beyond Databases, Architectures and Structures, BDAS 2017, Ustron, Poland, 30/05/2017. https://doi.org/10.1007/978-3-319-58274-0_21

Serendipitous recommendations through ontology-based contextual pre-filtering. / Karpus, Aleksandra; Vagliano, Iacopo; Goczyła, Krzysztof.
Beyond Databases, Architectures and Structures: Towards Efficient Solutions for Data Analysis and Knowledge Representation - 13th International Conference, BDAS 2017, Proceedings. ed. / Stanislaw Kozielski; Dariusz Mrozek; Pawel Kasprowski; Bozena Malysiak-Mrozek; Daniel Kostrzewa. Springer - Verlag, 2017. p. 246-259 (Communications in Computer and Information Science; Vol. 716).

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

TY - GEN

T1 - Serendipitous recommendations through ontology-based contextual pre-filtering

AU - Karpus, Aleksandra

AU - Vagliano, Iacopo

AU - Goczyła, Krzysztof

N1 - Funding Information: The second author was supported by a fellowship from TIM. Publisher Copyright: © Springer International Publishing AG 2017.

PY - 2017

Y1 - 2017

N2 - Context-aware Recommender Systems aim to provide users with better recommendations for their current situation. Although evaluations of recommender systems often focus on accuracy, it is not the only important aspect. Often recommendations are overspecialized, i.e. all of the same kind. To deal with this problem, other properties can be considered, such as serendipity. In this paper, we study how an ontology-based and context-aware pre-filtering technique which can be combined with existing recommendation algorithm performs in ranking tasks. We also investigate the impact of our method on the serendipity of the recommendations. We evaluated our approach through an offline study which showed that when used with well-known recommendation algorithms it can improve the accuracy and serendipity.

AB - Context-aware Recommender Systems aim to provide users with better recommendations for their current situation. Although evaluations of recommender systems often focus on accuracy, it is not the only important aspect. Often recommendations are overspecialized, i.e. all of the same kind. To deal with this problem, other properties can be considered, such as serendipity. In this paper, we study how an ontology-based and context-aware pre-filtering technique which can be combined with existing recommendation algorithm performs in ranking tasks. We also investigate the impact of our method on the serendipity of the recommendations. We evaluated our approach through an offline study which showed that when used with well-known recommendation algorithms it can improve the accuracy and serendipity.

KW - Context-awareness

KW - Ontologies

KW - Recommender systems

KW - Serendipity

UR - http://www.scopus.com/inward/record.url?scp=85019715098&partnerID=8YFLogxK

U2 - https://doi.org/10.1007/978-3-319-58274-0_21

DO - https://doi.org/10.1007/978-3-319-58274-0_21

M3 - Conference contribution

SN - 9783319582733

T3 - Communications in Computer and Information Science

SP - 246

EP - 259

BT - Beyond Databases, Architectures and Structures

A2 - Kozielski, Stanislaw

A2 - Mrozek, Dariusz

A2 - Kasprowski, Pawel

A2 - Malysiak-Mrozek, Bozena

A2 - Kostrzewa, Daniel

PB - Springer - Verlag

T2 - 13th International Conference on Beyond Databases, Architectures and Structures, BDAS 2017

Y2 - 30 May 2017 through 2 June 2017

ER -

Karpus A, Vagliano I, Goczyła K. Serendipitous recommendations through ontology-based contextual pre-filtering. In Kozielski S, Mrozek D, Kasprowski P, Malysiak-Mrozek B, Kostrzewa D, editors, Beyond Databases, Architectures and Structures: Towards Efficient Solutions for Data Analysis and Knowledge Representation - 13th International Conference, BDAS 2017, Proceedings. Springer - Verlag. 2017. p. 246-259. (Communications in Computer and Information Science). doi: https://doi.org/10.1007/978-3-319-58274-0_21

Serendipitous recommendations through ontology-based contextual pre-filtering

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this