Towards Interpretation of Recommender Systems with Sorted Explanation Paths

Fan Yang, Ninghao Liu, Suhang Wang, Xia Hu

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

4 Citations (Scopus)

Abstract

Despite the wide application in recent years, most recommender systems are not capable of providing interpretations together with recommendation results, which impedes both deployers and customers from understanding or trusting the results. Recent advances in recommendation models, such as deep learning models, usually involve extracting latent representations of users and items. However, the representation space is not directly comprehensible since each dimension usually does not have any specific meaning. In addition, recommender systems incorporate various sources of information, such as user behaviors, item information, and other side content information. Properly organizing different types of information, as well as effectively selecting important information for interpretation, is challenging and has not been fully tackled by conventional interpretation methods. In this paper, we propose a post-hoc method called Sorted Explanation Paths (SEP) to interpret recommendation results. Specifically, we first build a unified heterogeneous information network to incorporate multiple types of objects and relations based on representations from the recommender system and information from the dataset. Then, we search for explanation paths between given recommendation pairs, and use the set of simple paths to construct semantic explanations. Next, three heuristic metrics, i.e., credibility, readability and diversity, are designed to measure the validity of each explanation path, and to sort all the paths comprehensively. The top-ranked explanation paths are selected as the final interpretation. After that, practical issues on computation and efficiency of the proposed SEP method are also handled by corresponding approaches. Finally, we conduct experiments on three real-world benchmark datasets, and demonstrate the applicability and effectiveness of the proposed SEP method.

Original languageEnglish (US)
Title of host publication2018 IEEE International Conference on Data Mining, ICDM 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages667-676
Number of pages10
ISBN (Electronic)9781538691588
DOIs
StatePublished - Dec 27 2018
Event18th IEEE International Conference on Data Mining, ICDM 2018 - Singapore, Singapore
Duration: Nov 17 2018Nov 20 2018

Publication series

NameProceedings - IEEE International Conference on Data Mining, ICDM
Volume2018-November
ISSN (Print)1550-4786

Conference

Conference18th IEEE International Conference on Data Mining, ICDM 2018
CountrySingapore
CitySingapore
Period11/17/1811/20/18

Fingerprint

Recommender systems
Semantics
Experiments

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

Yang, F., Liu, N., Wang, S., & Hu, X. (2018). Towards Interpretation of Recommender Systems with Sorted Explanation Paths. In 2018 IEEE International Conference on Data Mining, ICDM 2018 (pp. 667-676). [8594891] (Proceedings - IEEE International Conference on Data Mining, ICDM; Vol. 2018-November). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICDM.2018.00082
Yang, Fan ; Liu, Ninghao ; Wang, Suhang ; Hu, Xia. / Towards Interpretation of Recommender Systems with Sorted Explanation Paths. 2018 IEEE International Conference on Data Mining, ICDM 2018. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 667-676 (Proceedings - IEEE International Conference on Data Mining, ICDM).
@inproceedings{71494c375d3c4ac29c8646d417b0822b,
title = "Towards Interpretation of Recommender Systems with Sorted Explanation Paths",
abstract = "Despite the wide application in recent years, most recommender systems are not capable of providing interpretations together with recommendation results, which impedes both deployers and customers from understanding or trusting the results. Recent advances in recommendation models, such as deep learning models, usually involve extracting latent representations of users and items. However, the representation space is not directly comprehensible since each dimension usually does not have any specific meaning. In addition, recommender systems incorporate various sources of information, such as user behaviors, item information, and other side content information. Properly organizing different types of information, as well as effectively selecting important information for interpretation, is challenging and has not been fully tackled by conventional interpretation methods. In this paper, we propose a post-hoc method called Sorted Explanation Paths (SEP) to interpret recommendation results. Specifically, we first build a unified heterogeneous information network to incorporate multiple types of objects and relations based on representations from the recommender system and information from the dataset. Then, we search for explanation paths between given recommendation pairs, and use the set of simple paths to construct semantic explanations. Next, three heuristic metrics, i.e., credibility, readability and diversity, are designed to measure the validity of each explanation path, and to sort all the paths comprehensively. The top-ranked explanation paths are selected as the final interpretation. After that, practical issues on computation and efficiency of the proposed SEP method are also handled by corresponding approaches. Finally, we conduct experiments on three real-world benchmark datasets, and demonstrate the applicability and effectiveness of the proposed SEP method.",
author = "Fan Yang and Ninghao Liu and Suhang Wang and Xia Hu",
year = "2018",
month = "12",
day = "27",
doi = "10.1109/ICDM.2018.00082",
language = "English (US)",
series = "Proceedings - IEEE International Conference on Data Mining, ICDM",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "667--676",
booktitle = "2018 IEEE International Conference on Data Mining, ICDM 2018",
address = "United States",

}

Yang, F, Liu, N, Wang, S & Hu, X 2018, Towards Interpretation of Recommender Systems with Sorted Explanation Paths. in 2018 IEEE International Conference on Data Mining, ICDM 2018., 8594891, Proceedings - IEEE International Conference on Data Mining, ICDM, vol. 2018-November, Institute of Electrical and Electronics Engineers Inc., pp. 667-676, 18th IEEE International Conference on Data Mining, ICDM 2018, Singapore, Singapore, 11/17/18. https://doi.org/10.1109/ICDM.2018.00082

Towards Interpretation of Recommender Systems with Sorted Explanation Paths. / Yang, Fan; Liu, Ninghao; Wang, Suhang; Hu, Xia.

2018 IEEE International Conference on Data Mining, ICDM 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 667-676 8594891 (Proceedings - IEEE International Conference on Data Mining, ICDM; Vol. 2018-November).

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

TY - GEN

T1 - Towards Interpretation of Recommender Systems with Sorted Explanation Paths

AU - Yang, Fan

AU - Liu, Ninghao

AU - Wang, Suhang

AU - Hu, Xia

PY - 2018/12/27

Y1 - 2018/12/27

N2 - Despite the wide application in recent years, most recommender systems are not capable of providing interpretations together with recommendation results, which impedes both deployers and customers from understanding or trusting the results. Recent advances in recommendation models, such as deep learning models, usually involve extracting latent representations of users and items. However, the representation space is not directly comprehensible since each dimension usually does not have any specific meaning. In addition, recommender systems incorporate various sources of information, such as user behaviors, item information, and other side content information. Properly organizing different types of information, as well as effectively selecting important information for interpretation, is challenging and has not been fully tackled by conventional interpretation methods. In this paper, we propose a post-hoc method called Sorted Explanation Paths (SEP) to interpret recommendation results. Specifically, we first build a unified heterogeneous information network to incorporate multiple types of objects and relations based on representations from the recommender system and information from the dataset. Then, we search for explanation paths between given recommendation pairs, and use the set of simple paths to construct semantic explanations. Next, three heuristic metrics, i.e., credibility, readability and diversity, are designed to measure the validity of each explanation path, and to sort all the paths comprehensively. The top-ranked explanation paths are selected as the final interpretation. After that, practical issues on computation and efficiency of the proposed SEP method are also handled by corresponding approaches. Finally, we conduct experiments on three real-world benchmark datasets, and demonstrate the applicability and effectiveness of the proposed SEP method.

AB - Despite the wide application in recent years, most recommender systems are not capable of providing interpretations together with recommendation results, which impedes both deployers and customers from understanding or trusting the results. Recent advances in recommendation models, such as deep learning models, usually involve extracting latent representations of users and items. However, the representation space is not directly comprehensible since each dimension usually does not have any specific meaning. In addition, recommender systems incorporate various sources of information, such as user behaviors, item information, and other side content information. Properly organizing different types of information, as well as effectively selecting important information for interpretation, is challenging and has not been fully tackled by conventional interpretation methods. In this paper, we propose a post-hoc method called Sorted Explanation Paths (SEP) to interpret recommendation results. Specifically, we first build a unified heterogeneous information network to incorporate multiple types of objects and relations based on representations from the recommender system and information from the dataset. Then, we search for explanation paths between given recommendation pairs, and use the set of simple paths to construct semantic explanations. Next, three heuristic metrics, i.e., credibility, readability and diversity, are designed to measure the validity of each explanation path, and to sort all the paths comprehensively. The top-ranked explanation paths are selected as the final interpretation. After that, practical issues on computation and efficiency of the proposed SEP method are also handled by corresponding approaches. Finally, we conduct experiments on three real-world benchmark datasets, and demonstrate the applicability and effectiveness of the proposed SEP method.

UR - http://www.scopus.com/inward/record.url?scp=85061397470&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061397470&partnerID=8YFLogxK

U2 - 10.1109/ICDM.2018.00082

DO - 10.1109/ICDM.2018.00082

M3 - Conference contribution

AN - SCOPUS:85061397470

T3 - Proceedings - IEEE International Conference on Data Mining, ICDM

SP - 667

EP - 676

BT - 2018 IEEE International Conference on Data Mining, ICDM 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Yang F, Liu N, Wang S, Hu X. Towards Interpretation of Recommender Systems with Sorted Explanation Paths. In 2018 IEEE International Conference on Data Mining, ICDM 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 667-676. 8594891. (Proceedings - IEEE International Conference on Data Mining, ICDM). https://doi.org/10.1109/ICDM.2018.00082