Abstract

Measures of node ranking, such as personalized PageRank, are utilized in many web and social-network based prediction and recommendation applications. Despite their e'ectiveness when the underlying graph is certain, however, these measures become difficult to apply in the presence of uncertainties, as they are not designed for graphs that include uncertain information, such as edges that mutually exclude each other. While there are several ways to naively extend existing techniques (such as trying to encode uncertainties as edge weights or computing all possible scenarios), as we discuss in this paper, these either lead to large degrees of errors or are very expensive to compute, as the number of possible worlds can grow exponentially with the amount of uncertainty. To tackle with this challenge, in this paper, we propose an effcient Uncertain Personalized PageRank (UPPR) algorithm to approximately compute personalized PageRank values on an uncertain graph with edge uncertainties. UPPR avoids enumeration of all possible worlds, yet it is able to achieve comparable accuracy by carefully encoding edge uncertainties in a data structure that leads to fast approximations. Experimental results show that UPPR is very effcient in terms of execution time and its accuracy is comparable or beffer than more costly alternatives.

Original languageEnglish (US)
Title of host publicationSIGIR 2017 - Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval
PublisherAssociation for Computing Machinery, Inc
Pages525-534
Number of pages10
ISBN (Electronic)9781450350228
DOIs
StatePublished - Aug 7 2017
Event40th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2017 - Tokyo, Shinjuku, Japan
Duration: Aug 7 2017Aug 11 2017

Other

Other40th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2017
CountryJapan
CityTokyo, Shinjuku
Period8/7/178/11/17

Fingerprint

Data structures
Uncertainty

ASJC Scopus subject areas

  • Information Systems
  • Software
  • Computer Graphics and Computer-Aided Design

Cite this

Kim, J. H., Li, M. L., Candan, K., & Sapino, M. L. (2017). Personalized pagerank in uncertain graphs with mutually exclusive edges. In SIGIR 2017 - Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 525-534). Association for Computing Machinery, Inc. https://doi.org/10.1145/3077136.3080794

Personalized pagerank in uncertain graphs with mutually exclusive edges. / Kim, Jung Hyun; Li, Mao Lin; Candan, Kasim; Sapino, Maria Luisa.

SIGIR 2017 - Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc, 2017. p. 525-534.

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

Kim, JH, Li, ML, Candan, K & Sapino, ML 2017, Personalized pagerank in uncertain graphs with mutually exclusive edges. in SIGIR 2017 - Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc, pp. 525-534, 40th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2017, Tokyo, Shinjuku, Japan, 8/7/17. https://doi.org/10.1145/3077136.3080794
Kim JH, Li ML, Candan K, Sapino ML. Personalized pagerank in uncertain graphs with mutually exclusive edges. In SIGIR 2017 - Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc. 2017. p. 525-534 https://doi.org/10.1145/3077136.3080794
Kim, Jung Hyun ; Li, Mao Lin ; Candan, Kasim ; Sapino, Maria Luisa. / Personalized pagerank in uncertain graphs with mutually exclusive edges. SIGIR 2017 - Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc, 2017. pp. 525-534
@inproceedings{41287333968648f4bcefc29c09ef4889,
title = "Personalized pagerank in uncertain graphs with mutually exclusive edges",
abstract = "Measures of node ranking, such as personalized PageRank, are utilized in many web and social-network based prediction and recommendation applications. Despite their e'ectiveness when the underlying graph is certain, however, these measures become difficult to apply in the presence of uncertainties, as they are not designed for graphs that include uncertain information, such as edges that mutually exclude each other. While there are several ways to naively extend existing techniques (such as trying to encode uncertainties as edge weights or computing all possible scenarios), as we discuss in this paper, these either lead to large degrees of errors or are very expensive to compute, as the number of possible worlds can grow exponentially with the amount of uncertainty. To tackle with this challenge, in this paper, we propose an effcient Uncertain Personalized PageRank (UPPR) algorithm to approximately compute personalized PageRank values on an uncertain graph with edge uncertainties. UPPR avoids enumeration of all possible worlds, yet it is able to achieve comparable accuracy by carefully encoding edge uncertainties in a data structure that leads to fast approximations. Experimental results show that UPPR is very effcient in terms of execution time and its accuracy is comparable or beffer than more costly alternatives.",
author = "Kim, {Jung Hyun} and Li, {Mao Lin} and Kasim Candan and Sapino, {Maria Luisa}",
year = "2017",
month = "8",
day = "7",
doi = "10.1145/3077136.3080794",
language = "English (US)",
pages = "525--534",
booktitle = "SIGIR 2017 - Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval",
publisher = "Association for Computing Machinery, Inc",

}

TY - GEN

T1 - Personalized pagerank in uncertain graphs with mutually exclusive edges

AU - Kim, Jung Hyun

AU - Li, Mao Lin

AU - Candan, Kasim

AU - Sapino, Maria Luisa

PY - 2017/8/7

Y1 - 2017/8/7

N2 - Measures of node ranking, such as personalized PageRank, are utilized in many web and social-network based prediction and recommendation applications. Despite their e'ectiveness when the underlying graph is certain, however, these measures become difficult to apply in the presence of uncertainties, as they are not designed for graphs that include uncertain information, such as edges that mutually exclude each other. While there are several ways to naively extend existing techniques (such as trying to encode uncertainties as edge weights or computing all possible scenarios), as we discuss in this paper, these either lead to large degrees of errors or are very expensive to compute, as the number of possible worlds can grow exponentially with the amount of uncertainty. To tackle with this challenge, in this paper, we propose an effcient Uncertain Personalized PageRank (UPPR) algorithm to approximately compute personalized PageRank values on an uncertain graph with edge uncertainties. UPPR avoids enumeration of all possible worlds, yet it is able to achieve comparable accuracy by carefully encoding edge uncertainties in a data structure that leads to fast approximations. Experimental results show that UPPR is very effcient in terms of execution time and its accuracy is comparable or beffer than more costly alternatives.

AB - Measures of node ranking, such as personalized PageRank, are utilized in many web and social-network based prediction and recommendation applications. Despite their e'ectiveness when the underlying graph is certain, however, these measures become difficult to apply in the presence of uncertainties, as they are not designed for graphs that include uncertain information, such as edges that mutually exclude each other. While there are several ways to naively extend existing techniques (such as trying to encode uncertainties as edge weights or computing all possible scenarios), as we discuss in this paper, these either lead to large degrees of errors or are very expensive to compute, as the number of possible worlds can grow exponentially with the amount of uncertainty. To tackle with this challenge, in this paper, we propose an effcient Uncertain Personalized PageRank (UPPR) algorithm to approximately compute personalized PageRank values on an uncertain graph with edge uncertainties. UPPR avoids enumeration of all possible worlds, yet it is able to achieve comparable accuracy by carefully encoding edge uncertainties in a data structure that leads to fast approximations. Experimental results show that UPPR is very effcient in terms of execution time and its accuracy is comparable or beffer than more costly alternatives.

UR - http://www.scopus.com/inward/record.url?scp=85029375216&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029375216&partnerID=8YFLogxK

U2 - 10.1145/3077136.3080794

DO - 10.1145/3077136.3080794

M3 - Conference contribution

SP - 525

EP - 534

BT - SIGIR 2017 - Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval

PB - Association for Computing Machinery, Inc

ER -