A Weighted Meta-graph Based Approach for Mobile Application Recommendation on Heterogeneous Information Networks

Xie, Fei; Chen, Liang; Ye, Yongjian; Yang, Liu; Zheng, Zibin; Lin, Xiaola

doi:10.1007/978-3-030-03596-9_29

“…Further considering the interactions of categories and other side information of apps, Liang et al [9] utilized a tensor-based framework to effectively integrate app category information and multi-view features of users and apps to do context-aware app recommendation. Focusing on the complex semantics among different kinds of side information, Xie et al [10] exploited weighted meta-graph and heterogeneous information network for mobile app recommendation, mainly considering user review information. However, it is not an end-toend method.…”

Section: Mobile App Recommendationmentioning

confidence: 99%

“…As detailed in Sect. 5, most of them [2][3][4][5][6][7][8][9][10] only exploited limited types of side information. In addition, they usually treated different kinds of side information as isolated features of users and apps, and neglected the relations and semantics of them.…”

Section: Introductionmentioning

confidence: 99%

A Knowledge Graph Based Approach for Mobile Application Recommendation

Zhang

¹

,

Zhao

²

,

Dong

³

et al. 2020

Service-Oriented Computing

5

0

View full text Add to dashboard Cite

With the rapid prevalence of mobile devices and the dramatic proliferation of mobile applications (apps), app recommendation becomes an emergent task that would benefit both app users and stockholders. How to effectively organize and make full use of rich side information of users and apps is a key challenge to address the sparsity issue for traditional approaches. To meet this challenge, we proposed a novel endto-end Knowledge Graph Convolutional Embedding Propagation Model (KGEP) for app recommendation. Specifically, we first designed a knowledge graph construction method to model the user and app side information, then adopted KG embedding techniques to capture the factual triplet-focused semantics of the side information related to the first-order structure of the KG, and finally proposed a relation-weighted convolutional embedding propagation model to capture the recommendationfocused semantics related to high-order structure of the KG. Extensive experiments conducted on a real-world dataset validate the effectiveness of the proposed approach compared to the state-of-the-art recommendation approaches.

show abstract

“…Further considering the interactions of categories and other side information of apps, Liang et al [9] utilized a tensor-based framework to effectively integrate app category information and multi-view features of users and apps to do context-aware app recommendation. Focusing on the complex semantics among different kinds of side information, Xie et al [10] exploited weighted meta-graph and heterogeneous information network for mobile app recommendation, mainly considering user review information. However, it is not an end-to-end method.…”

Section: Related Workmentioning

confidence: 99%

“…To address the sparsity problem of user-app interactions, researchers usually turn to feature-rich scenarios, where side information of users and apps is used to compensate for the sparsity and improve the performance of recommendation. As detailed in section 5, most of them [2,3,4,5,6,7,8,9,10] only exploited limited types of side information. In addition, they usually treated different kinds of side information as isolated features of users and apps, and neglected the relations and semantics of them.…”

Section: Introductionmentioning

confidence: 99%

A Knowledge Graph based Approach for Mobile Application Recommendation

Zhang¹,

Zhao²,

Dong³

et al. 2020

Preprint

0

View full text Add to dashboard Cite

With the rapid prevalence of mobile devices and the dramatic proliferation of mobile applications (apps), app recommendation becomes an emergent task that would benefit both app users and stockholders. How to effectively organize and make full use of rich side information of users and apps is a key challenge to address the sparsity issue for traditional approaches. To meet this challenge, we proposed a novel endto-end Knowledge Graph Convolutional Embedding Propagation Model (KGEP) for app recommendation. Specifically, we first designed a knowledge graph construction method to model the user and app side information, then adopted KG embedding techniques to capture the factual triplet-focused semantics of the side information related to the first-order structure of the KG, and finally proposed a relation-weighted convolutional embedding propagation model to capture the recommendationfocused semantics related to high-order structure of the KG. Extensive experiments conducted on a real-world dataset validate the effectiveness of the proposed approach compared to the state-of-the-art recommendation approaches.

show abstract

“…Among these studies, one of the most widely used technique in recommender systems is collaborative filtering (CF) 4 , which focuses on the users' historical information. Depending on the methods utilized to learn the correlation between users' historical information, CF recommender systems can be divided into four categories: matrix-factorization-based methods 5 , graph-based methods 6 , association-rule-based methods 7,8 and neighbourhood-based methods 9 . Numerous commercial companies (e.g., Netflix 1 , YouTube 2 , eBay 3 and Taobao 4 ) have already applied CF recommender systems to their products, such as web and app, to alleviate the information overload, improve users' experience, and bring them tremendous economic benefits.…”

Section: Introductionmentioning

confidence: 99%

Data poisoning attacks on neighborhood‐based recommender systems

Chen

¹

,

Xu

²

,

Xie

³

et al. 2020

Trans Emerging Tel Tech

Self Cite

View full text Add to dashboard Cite

Nowadays, collaborative filtering recommender systems have been widely deployed in many commercial companies to make profit. Neighbourhood-based collaborative filtering is common and effective. To date, despite its effectiveness, there has been little effort to explore their robustness and the impact of data poisoning attacks on their performance. Can the neighbourhood-based recommender systems be easily fooled? To this end, we shed light on the robustness of neighbourhood-based recommender systems and propose a novel data poisoning attack framework encoding the purpose of attack and constraint against them. We firstly illustrate how to calculate the optimal data poisoning attack, namely UNAttack. We inject a few well-designed fake users into the recommender systems such that target items will be recommended to as many normal users as possible. Extensive experiments are conducted on three real-world datasets to validate the effectiveness and the transferability of our proposed method. Besides, some interesting phenomenons can be found. For example, 1) neighbourhood-based recommender systems with Euclidean Distance-based similarity have strong robustness. 2) the fake users can be transferred to attack the state-of-the-art collaborative filtering recommender systems such as Neural Collaborative Filtering and Bayesian Personalized Ranking Matrix Factorization.

show abstract

A Weighted Meta-graph Based Approach for Mobile Application Recommendation on Heterogeneous Information Networks

Cited by 17 publications

References 18 publications

A Knowledge Graph Based Approach for Mobile Application Recommendation

A Knowledge Graph Based Approach for Mobile Application Recommendation

A Knowledge Graph based Approach for Mobile Application Recommendation

Data poisoning attacks on neighborhood‐based recommender systems

Contact Info

Product

Resources

About