Predicting trading interactions in an online marketplace through location-based and online social networks

Lukas Eberhard, Christoph Trattner, Martin Atzmueller

Research output: Contribution to journalArticleResearchpeer-review

Abstract

Link prediction is a prominent research direction e.g., for inferring upcoming interactions to be used in recommender systems. Although this problem of predicting links between users has been extensively studied in the past, research investigating this issue simultaneously in multiplex networks is rather rare so far. This is the focus of this paper. We investigate the extent to which trading interactions between sellers and buyers within an online marketplace platform can be predicted based on three different but overlapping networks—an online social network, a location-based social network and a trading network. In particular, we conducted the study in the context of the virtual world Second Life. For that, we crawled according data of the online social network, user information of the location-based social network obtained by specialized bots, and we extracted purchases of the trading network. Overall, we generated and used 57 topological and homophilic features in different constellations to predict trading interactions between user pairs. We focused on both unsupervised as well as supervised learning methods. For supervised learning, we achieved accuracy values up to (Formula presented.), for unsupervised learning we obtained nDCG values up to over (Formula presented.) and MAP values up to (Formula presented.).

Original languageEnglish
Pages (from-to)1-38
Number of pages38
JournalInformation Retrieval Journal
DOIs
Publication statusE-pub ahead of print - 9 Jul 2018

Fingerprint

Supervised learning
social network
Unsupervised learning
Recommender systems
interaction
Values
learning method
learning
purchase

Keywords

  • Buyer
  • Link prediction
  • Location-based and online social networks
  • Second life
  • Seller
  • Supervised and unsupervised learning

ASJC Scopus subject areas

  • Information Systems
  • Library and Information Sciences

Cite this

Predicting trading interactions in an online marketplace through location-based and online social networks. / Eberhard, Lukas; Trattner, Christoph; Atzmueller, Martin.

In: Information Retrieval Journal, 09.07.2018, p. 1-38.

Research output: Contribution to journalArticleResearchpeer-review

@article{a5fcbf38a06c493a86faf8bec517c70e,
title = "Predicting trading interactions in an online marketplace through location-based and online social networks",
abstract = "Link prediction is a prominent research direction e.g., for inferring upcoming interactions to be used in recommender systems. Although this problem of predicting links between users has been extensively studied in the past, research investigating this issue simultaneously in multiplex networks is rather rare so far. This is the focus of this paper. We investigate the extent to which trading interactions between sellers and buyers within an online marketplace platform can be predicted based on three different but overlapping networks—an online social network, a location-based social network and a trading network. In particular, we conducted the study in the context of the virtual world Second Life. For that, we crawled according data of the online social network, user information of the location-based social network obtained by specialized bots, and we extracted purchases of the trading network. Overall, we generated and used 57 topological and homophilic features in different constellations to predict trading interactions between user pairs. We focused on both unsupervised as well as supervised learning methods. For supervised learning, we achieved accuracy values up to (Formula presented.), for unsupervised learning we obtained nDCG values up to over (Formula presented.) and MAP values up to (Formula presented.).",
keywords = "Buyer, Link prediction, Location-based and online social networks, Second life, Seller, Supervised and unsupervised learning",
author = "Lukas Eberhard and Christoph Trattner and Martin Atzmueller",
year = "2018",
month = "7",
day = "9",
doi = "10.1007/s10791-018-9336-z",
language = "English",
pages = "1--38",
journal = "Information Retrieval Journal",
issn = "1386-4564",
publisher = "Springer Science+Business Media B.V",

}

TY - JOUR

T1 - Predicting trading interactions in an online marketplace through location-based and online social networks

AU - Eberhard, Lukas

AU - Trattner, Christoph

AU - Atzmueller, Martin

PY - 2018/7/9

Y1 - 2018/7/9

N2 - Link prediction is a prominent research direction e.g., for inferring upcoming interactions to be used in recommender systems. Although this problem of predicting links between users has been extensively studied in the past, research investigating this issue simultaneously in multiplex networks is rather rare so far. This is the focus of this paper. We investigate the extent to which trading interactions between sellers and buyers within an online marketplace platform can be predicted based on three different but overlapping networks—an online social network, a location-based social network and a trading network. In particular, we conducted the study in the context of the virtual world Second Life. For that, we crawled according data of the online social network, user information of the location-based social network obtained by specialized bots, and we extracted purchases of the trading network. Overall, we generated and used 57 topological and homophilic features in different constellations to predict trading interactions between user pairs. We focused on both unsupervised as well as supervised learning methods. For supervised learning, we achieved accuracy values up to (Formula presented.), for unsupervised learning we obtained nDCG values up to over (Formula presented.) and MAP values up to (Formula presented.).

AB - Link prediction is a prominent research direction e.g., for inferring upcoming interactions to be used in recommender systems. Although this problem of predicting links between users has been extensively studied in the past, research investigating this issue simultaneously in multiplex networks is rather rare so far. This is the focus of this paper. We investigate the extent to which trading interactions between sellers and buyers within an online marketplace platform can be predicted based on three different but overlapping networks—an online social network, a location-based social network and a trading network. In particular, we conducted the study in the context of the virtual world Second Life. For that, we crawled according data of the online social network, user information of the location-based social network obtained by specialized bots, and we extracted purchases of the trading network. Overall, we generated and used 57 topological and homophilic features in different constellations to predict trading interactions between user pairs. We focused on both unsupervised as well as supervised learning methods. For supervised learning, we achieved accuracy values up to (Formula presented.), for unsupervised learning we obtained nDCG values up to over (Formula presented.) and MAP values up to (Formula presented.).

KW - Buyer

KW - Link prediction

KW - Location-based and online social networks

KW - Second life

KW - Seller

KW - Supervised and unsupervised learning

UR - http://www.scopus.com/inward/record.url?scp=85049621864&partnerID=8YFLogxK

U2 - 10.1007/s10791-018-9336-z

DO - 10.1007/s10791-018-9336-z

M3 - Article

SP - 1

EP - 38

JO - Information Retrieval Journal

JF - Information Retrieval Journal

SN - 1386-4564

ER -