arxiv: 2604.14598 · v2 · submitted 2026-04-16 · 💻 cs.IR

Recognition: unknown

Category-based and Popularity-guided Video Game Recommendation: A Balance-oriented Framework

Xiping Li , Jianghong Ma , Kangzhe Liu , Shanshan Feng , Haijun Zhang , Yutong Wang

Authors on Pith no claims yet

Pith reviewed 2026-05-10 10:14 UTC · model grok-4.3

classification 💻 cs.IR

keywords video game recommendationaccuracy-diversity balancegraph-based recommendationcategory informationpopularity guidanceSteam datasetrecommender systems

0 comments

The pith

CPGRec balances accuracy and diversity in video game recommendations through stricter game connections, category-diverse neighbors, and popularity-guided amplification of long-tail items.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper seeks to build a recommender system for video games that improves both accuracy and diversity at the same time, rather than trading one for the other. It starts from the observation that existing methods either enforce loose connections between games, which limits accuracy, or ignore category labels and popularity signals when selecting neighbors, which limits variety. The authors therefore construct a three-module system that first tightens game-to-game links for accuracy, then selects category-diverse neighbors and routes influence from popular games to rare ones for diversity, and finally merges the two with a reweighting step on negative samples. If the approach works, players facing thousands of titles would receive suggestions that are both relevant and varied without extra tuning after the fact.

Core claim

The CPGRec framework improves game recommendations by extending accuracy-focused methods with more stringent game connections in an accuracy-driven module, connecting neighbors that span diverse categories while amplifying long-tail games through popular nodes in a diversity-driven module, and fusing the two signals in a comprehensive module via a new negative-sample rating score reweighting technique, with experiments on the Steam dataset confirming gains in both accuracy and diversity metrics.

What carries the argument

The CPGRec framework and its three modules: an accuracy-driven module that enforces stricter game connections, a diversity-driven module that selects category-diverse neighbors in the game graph and amplifies long-tail items via popular nodes in the bipartite player-game graph, and a comprehensive module that combines the signals with negative-sample reweighting.

If this is right

Stricter connections between games will produce higher accuracy scores than methods using looser links.
Selecting neighbors across multiple categories while boosting long-tail games via popularity will increase recommendation variety.
The combined module with negative-sample reweighting will achieve the balance without one objective dominating the other.
Overall performance on the Steam dataset will exceed that of prior accuracy-only or diversity-only game recommenders.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same separation of a game graph for accuracy and a bipartite graph for diversity could be reused in other domains where items carry category labels and popularity skews are strong.
The approach implies that explicit modeling of item-to-item relations alongside user-to-item relations is necessary when both precision and coverage matter.
If the reweighting step proves robust, it could reduce the need for separate hyper-parameter searches that often hide accuracy-diversity conflicts.

Load-bearing premise

The proposed strict game connections, category-diverse neighbors, and popularity amplification will simultaneously raise accuracy and diversity without requiring post-hoc tuning that masks trade-offs.

What would settle it

Evaluating CPGRec on the Steam dataset and finding no joint improvement in standard accuracy metrics such as precision or recall and diversity metrics such as intra-list diversity or coverage, relative to strong baselines, would falsify the central claim.

Figures

Figures reproduced from arXiv: 2604.14598 by Haijun Zhang, Jianghong Ma, Kangzhe Liu, Shanshan Feng, Xiping Li, Yutong Wang.

**Figure 2.** Figure 2: Illustration of the proposed framework of CPGRec. The left module is designed to emphasize accuracy, which is [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

**Figure 3.** Figure 3: Number of edges on Steam game graphs, where the [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

**Figure 4.** Figure 4: The comparative functions of negative-sample rat [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗

**Figure 5.** Figure 5: Parameter sensitivity on 𝜃 ℎ𝑜𝑡 𝑒 ,𝜃 ℎ𝑜𝑡 𝑛 , and 𝜃 𝑐𝑜𝑙𝑑 𝑛 . Regarding 𝜃 𝑐𝑜𝑙𝑑 𝑛 , enhancing the node weight of long-tail games gradually from 1 to 5 results in the expected outcome: the model improves diversity significantly at the cost of accuracy by assigning higher weights to long-tail games. However, it is important to note that when long-tail games are given excessively high weights (e.g., 7, 9), both a… view at source ↗

**Figure 6.** Figure 6: As the parameter 𝑤𝐶𝑎 gradually increases, CPGRec consistently improves its accuracy performance. Regarding diversity, as measured by Coverage@5, CPGRec shows fluctuations but generally maintains around 8.5. This behavior contrasts with the typical accuracy-diversity dilemma, where diversity often sharply decreases as accuracy increases. CPGRec effectively manages to maintain diversity while enhancing acc… view at source ↗

**Figure 7.** Figure 7: The average number of deceptive games recom [PITH_FULL_IMAGE:figures/full_fig_p011_7.png] view at source ↗

**Figure 8.** Figure 8: The average number of long-tail games recom [PITH_FULL_IMAGE:figures/full_fig_p011_8.png] view at source ↗

read the original abstract

In recent years, the video game industry has experienced substantial growth, presenting players with a vast array of game choices. This surge in options has spurred the need for a specialized recommender system tailored for video games. However, current video game recommendation approaches tend to prioritize accuracy over diversity, potentially leading to unvaried game suggestions. In addition, the existing game recommendation methods commonly lack the ability to establish strict connections between games to enhance accuracy. Furthermore, many existing diversity-focused methods fail to leverage crucial item information, such as item category and popularity during neighbor modeling and message propagation. To address these challenges, we introduce a novel framework, called CPGRec, comprising three modules, namely accuracy-driven, diversity-driven, and comprehensive modules. The first module extends the state-of-the-art accuracy-focused game recommendation method by connecting games in a more stringent manner to enhance recommendation accuracy. The second module connects neighbors with diverse categories within the proposed game graph and harnesses the advantages of popular game nodes to amplify the influence of long-tail games within the player-game bipartite graph, thereby enriching recommendation diversity. The third module combines the above two modules and employs a new negative-sample rating score reweighting method to balance accuracy and diversity. Experimental results on the Steam dataset demonstrate the effectiveness of our proposed method in improving game recommendations. The dataset and source codes are anonymously released at: https://github.com/CPGRec2024/CPGRec.git.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

CPGRec adds strict connections, category-diverse neighbors, popularity boosting, and negative reweighting to graph recs for games, with code released, but the balance claim needs ablations to show it isn't just tuned trade-offs on Steam.

read the letter

The paper's core contribution is a three-module graph framework called CPGRec that tightens game-to-game links for accuracy, picks category-diverse neighbors while using popular nodes to lift long-tail games, and then combines them through a negative-sample reweighting step to push both accuracy and diversity at once. It reports gains on the Steam dataset and releases the code and data anonymously on GitHub, which is the most immediately useful part for anyone who wants to test or extend it.

Referee Report

2 major / 2 minor

Summary. The paper proposes CPGRec, a three-module framework for video game recommendation. The accuracy-driven module extends prior accuracy-focused methods by imposing stricter connections between games. The diversity-driven module constructs neighbors with diverse categories in a game graph and amplifies popular nodes to boost long-tail games in the player-game bipartite graph. The comprehensive module integrates the two via a negative-sample rating reweighting scheme intended to balance accuracy and diversity. Experiments on the Steam dataset are reported to demonstrate effectiveness, with dataset and code released anonymously.

Significance. If the balance claim holds under rigorous validation, the work provides a graph-based approach to jointly improving accuracy and diversity in game recommendations by explicitly incorporating category information and popularity signals. The release of code and data is a clear strength that enables direct reproduction and extension.

major comments (2)

[Experimental evaluation section] Experimental evaluation section: the central claim that the negative-sample rating reweighting simultaneously lifts both accuracy and diversity metrics lacks any ablation isolating its contribution, any Pareto-front analysis across hyperparameter settings in the comprehensive module, or statistical significance tests on the reported gains. This directly affects the load-bearing assumption that the three modules produce genuine joint improvements rather than tuned trade-offs.
[Framework description (modules 1-3)] Framework description (modules 1-3): no explicit comparison is made to the strongest recent accuracy-only and diversity-only baselines on Steam, nor is it shown that the stricter game connections and category-diverse neighbors are not simply recovering performance already achievable by standard GNN message-passing variants.

minor comments (2)

[Abstract] The abstract states that 'existing diversity-focused methods fail to leverage crucial item information such as item category and popularity during neighbor modeling,' but does not cite the specific prior works being critiqued.
[Comprehensive module] Notation for the reweighting formula in the comprehensive module should be introduced with a clear equation number and variable definitions to avoid ambiguity when readers attempt to re-implement.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We address each major comment below and indicate planned revisions to the manuscript.

read point-by-point responses

Referee: [Experimental evaluation section] Experimental evaluation section: the central claim that the negative-sample rating reweighting simultaneously lifts both accuracy and diversity metrics lacks any ablation isolating its contribution, any Pareto-front analysis across hyperparameter settings in the comprehensive module, or statistical significance tests on the reported gains. This directly affects the load-bearing assumption that the three modules produce genuine joint improvements rather than tuned trade-offs.

Authors: We acknowledge the absence of these specific analyses in the current experimental section. To directly support the claim of joint improvement via the negative-sample reweighting, the revised manuscript will add: (1) an ablation that isolates the reweighting by disabling it in the comprehensive module and reports the resulting accuracy and diversity metrics; (2) a Pareto-front plot obtained by sweeping the relevant hyperparameters of the comprehensive module; and (3) statistical significance tests (paired t-tests with p-values) on the reported gains. These additions will clarify whether the observed balance is robust rather than a tuned trade-off. revision: yes
Referee: [Framework description (modules 1-3)] Framework description (modules 1-3): no explicit comparison is made to the strongest recent accuracy-only and diversity-only baselines on Steam, nor is it shown that the stricter game connections and category-diverse neighbors are not simply recovering performance already achievable by standard GNN message-passing variants.

Authors: The current experiments already include multiple accuracy- and diversity-oriented baselines, yet we agree that the strongest recent methods specific to Steam should be added for completeness. In revision we will incorporate the most recent accuracy-only and diversity-only GNN-based recommenders applicable to the Steam dataset. In addition, we will include a controlled comparison against vanilla GNN message-passing on the identical game graph and bipartite graph structures to demonstrate that the stricter connections and category-diverse neighbor selection yield measurable gains beyond standard propagation. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical framework validated externally with no self-referential derivations.

full rationale

The paper introduces CPGRec as a three-module architecture (accuracy-driven strict connections, diversity-driven category/popularity neighbors, and comprehensive negative-sample reweighting) whose claimed improvements are demonstrated solely through experiments on the external Steam dataset. No equations or derivations are presented that reduce the reported accuracy/diversity gains to quantities defined by the modules themselves or by fitted parameters renamed as predictions. The reweighting step is a design choice whose joint effect is measured empirically rather than assumed by construction, and no self-citation chain or uniqueness theorem is invoked to force the framework. The derivation chain is therefore self-contained against the dataset benchmark.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

Abstract-only review limits visibility into exact hyperparameters; the framework implicitly relies on standard graph-neural-network assumptions and introduces balancing weights whose values are not specified here.

free parameters (1)

module balancing weights
Weights that combine accuracy and diversity modules and the negative-sample reweighting factor are expected to be tuned on data.

axioms (1)

domain assumption Games can be meaningfully connected via category labels and popularity signals in a bipartite player-game graph
Invoked in the diversity module description.

pith-pipeline@v0.9.0 · 5569 in / 1220 out tokens · 39381 ms · 2026-05-10T10:14:20.254018+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

60 extracted references · 4 canonical work pages · 2 internal anchors

[1]

Chris Anderson, Christopher Nissley, and Chris Anderson. 2006. The long tail

2006
[2]

Syed Muhammad Anwar, Talha Shahzad, Zunaira Sattar, Rahma Khan, and Muhammad Majid. 2017. A game recommender system using collaborative filtering (GAMBIT). In 2017 14th International Bhurban Conference on Applied Sciences and Technology (IBCAST). 328–332

2017
[3]

BharathiPriya, Akash Sreenivasu, and Sampath Kumar

C. BharathiPriya, Akash Sreenivasu, and Sampath Kumar. 2021. Online Video Game Recommendation System Using Content And Collaborative Filtering Tech- niques. In 2021 International Conference on Advancements in Electrical, Electronics, Communication, Computing and Automation (ICAECA) . 1–7

2021
[4]

Jaime Carbonell and Jade Goldstein. 1998. The use of MMR, diversity-based reranking for reordering documents and producing summaries. In Proceedings of the 21st annual international ACM SIGIR conference on Research and development in information retrieval. 335–336

1998
[5]

Loïc Caroux. 2023. Presence in video games: A systematic review and meta- analysis of the effects of game design choices. Applied Ergonomics 107 (2023), 103936

2023
[6]

Peizhe Cheng, Shuaiqiang Wang, Jun Ma, Jiankai Sun, and Hui Xiong. 2017. Learning to recommend accurate and diverse items. In Proceedings of the World Wide Web. 183–192

2017
[7]

Germán Cheuque, Jose Antonio Guzman Gomez, and Denis Parra. 2019. Recom- mender Systems for Online Video Game Platforms: the Case of STEAM. 763–771

2019
[8]

Jiasheng Duan, Peng-Fei Zhang, Ruihong Qiu, and Zi Huang. 2023. Long short- term enhanced memory for sequential recommendation. World Wide Web 26, 2 (2023), 561–583

2023
[9]

Chen Gao, Chao Huang, Donghan Yu, Haohao Fu, Tzh-Heng Lin, Depeng Jin, and Yong Li. 2022. Item Recommendation for Word-of-Mouth Scenario in Social E-Commerce. IEEE Transactions on Knowledge and Data Engineering 34, 6 (2022), 2798–2809

2022
[10]

Chen Gao, Tzu-Heng Lin, Nian Li, Depeng Jin, and Yong Li. 2023. Cross-Platform Item Recommendation for Online Social E-Commerce. IEEE Transactions on Knowledge and Data Engineering 35, 2 (2023), 1351–1364

2023
[11]

Valerio La Gatta, Vincenzo Moscato, Mirko Pennone, Marco Postiglione, and Giancarlo Sperlí. 2023. Music Recommendation via Hypergraph Embedding.IEEE Transactions on Neural Networks and Learning Systems 34, 10 (2023), 7887–7899

2023
[12]

Inductive Representation Learning on Large Graphs

William L. Hamilton, Rex Ying, and Jure Leskovec. 2018. Inductive Representation Learning on Large Graphs. arXiv:1706.02216 [cs.SI]

work page Pith review arXiv 2018
[13]

Han Han, Can Wang, Yunwei Zhao, Min Shu, Wenlei Wang, and Yong Min
[14]

Filter Bubble

SSLE: A framework for evaluating the “Filter Bubble” effect on the news aggregator and recommenders. World Wide Web 25, 3 (2022), 1169–1195

2022
[15]

Xiangnan He, Kuan Deng, Xiang Wang, Yan Li, Yongdong Zhang, and Meng Wang. 2020. Lightgcn: Simplifying and powering graph convolution network for recommendation. In Proceedings of the 43rd International ACM SIGIR conference on research and development in Information Retrieval . 639–648

2020
[16]

Fasiha Ikram, Humera Farooq, et al . 2022. Multimedia Recommendation Sys- tem for Video Game Based on High-Level Visual Semantic Features. Scientific Programming 2022 (2022)

2022
[17]

Nan Jiang, Zihao Hu, Jie Wen, Jiahui Zhao, Weihao Gu, Ziang Tu, Ximeng Liu, Yuanyuan Li, Jianfei Gong, and Fengtao Lin. 2023. NAH: neighbor-aware attention-based heterogeneous relation network model in E-commerce recom- mendation. World Wide Web (2023), 1–22

2023
[18]

Sichen Jin, Yijia Zhang, Xingwang Li, and Mingyu Lu. 2023. Heterogeneous Graph Convolutional Network for E-commerce Product Recommendation with Adaptive Denoising Training. IEEE Transactions on Consumer Electronics (2023), 1–1

2023
[19]

Zafran Khan, Muhammad Ishfaq Hussain, Naima Iltaf, Joonmo Kim, and Moongu Jeon. 2021. Contextual recommender system for E-commerce applications. Ap- plied Soft Computing 109 (2021), 107552

2021
[20]

Semi-Supervised Classification with Graph Convolutional Networks

Thomas N. Kipf and Max Welling. 2017. Semi-Supervised Classification with Graph Convolutional Networks. arXiv:1609.02907 [cs.LG]

work page internal anchor Pith review arXiv 2017
[21]

Silvia Knobloch-Westerwick and Axel Westerwick. 2023. Algorithmic personal- ization of source cues in the filter bubble: Self-esteem and self-construal impact information exposure. New Media & Society 25, 8 (2023), 2095–2117

2023
[22]

Valerio La Gatta, Vincenzo Moscato, Mirko Pennone, Marco Postiglione, and Giancarlo Sperlí. 2022. Music recommendation via hypergraph embedding. IEEE Transactions on Neural Networks and Learning Systems (2022)

2022
[23]

Yile Liang, Tieyun Qian, Qing Li, and Hongzhi Yin. 2021. Enhancing domain- level and user-level adaptivity in diversified recommendation. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. 747–756

2021
[24]

Peiyang Liu. 2024. Unsupervised corrupt data detection for text training. Expert Systems with Applications 248 (2024), 123335

2024
[25]

Peiyang Liu, Xi Wang, Ziqiang Cui, and Wei Ye. 2025. Queries Are Not Alone: Clustering Text Embeddings for Video Search. In The 48th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2025). Association for Computing Machinery, 874–883

2025
[26]

Peiyang Liu, Jinyu Yang, Lin Wang, Sen Wang, Yunlai Hao, and Huihui Bai. 2023. Retrieval-Based Unsupervised Noisy Label Detection on Text Data. In Proceed- ings of the 32nd ACM International Conference on Information and Knowledge Management. 4099–4104

2023
[27]

Wenpeng Lu, Rongyao Wang, Shoujin Wang, Xueping Peng, Hao Wu, and Qian Zhang. 2022. Aspect-Driven User Preference and News Representation Learn- ing for News Recommendation. IEEE Transactions on Intelligent Transportation Systems 23, 12 (2022), 25297–25307

2022
[28]

Lien Michiels, Jens Leysen, Annelien Smets, and Bart Goethals. 2022. What are filter bubbles really? A review of the conceptual and empirical work. In Adjunct Proceedings of the 30th ACM Conference on User Modeling, Adaptation and Personalization. 274–279

2022
[29]

Vincenzo Moscato, Antonio Picariello, and Giancarlo Sperlí. 2021. An Emotional Recommender System for Music. IEEE Intelligent Systems 36, 5 (2021), 57–68

2021
[30]

Mark O’Neill, Elham Vaziripour, Justin Wu, and Daniel Zappala. 2016. Condensing steam: Distilling the diversity of gamer behavior. InProceedings of the 2016 internet measurement conference. 81–95

2016
[31]

Jiménez-Bravo, Vivian F

Javier Pérez-Marcos, Lucía Martín-Gómez, Diego M. Jiménez-Bravo, Vivian F. López, and María N. Moreno-García. 2020. Hybrid system for video game rec- ommendation based on implicit ratings and social networks. Journal of Ambient Intelligence and Humanized Computing 11, 11 (01 Nov 2020), 4525–4535

2020
[32]

Laure Prétet, Gaël Richard, Clément Souchier, and Geoffroy Peeters. 2023. Video- to-Music Recommendation Using Temporal Alignment of Segments. IEEE Trans- actions on Multimedia 25 (2023), 2898–2911

2023
[33]

Steffen Rendle, Christoph Freudenthaler, Zeno Gantner, and Lars Schmidt- Thieme. 2012. BPR: Bayesian Personalized Ranking from Implicit Feedback. arXiv:1205.2618 [cs.IR]

work page internal anchor Pith review arXiv 2012
[34]

Michail Salampasis, Alkiviadis Katsalis, Theodosios Siomos, Marina Delianidi, Dimitrios Tektonidis, Konstantinos Christantonis, Pantelis Kaplanoglou, Ifigeneia Karaveli, Chrysostomos Bourlis, and Konstantinos Diamantaras. 2023. A Flexible Session-Based Recommender System for e-Commerce. Applied Sciences 13, 5 (2023)

2023
[35]

Alexander Semenov, Maciej Rysz, Gaurav Pandey, and Guanglin Xu. 2022. Di- versity in news recommendations using contextual bandits. Expert Systems with Applications 195 (2022), 116478

2022
[36]

Heng-Shiou Sheu, Zhixuan Chu, Daiqing Qi, and Sheng Li. 2022. Knowledge- Guided Article Embedding Refinement for Session-Based News Recommendation. IEEE Transactions on Neural Networks and Learning Systems 33, 12 (2022), 7921– 7927

2022
[37]

Lei Shi. 2013. Trading-off among accuracy, similarity, diversity, and long-tail: a graph-based recommendation approach. InProceedings of the 7th ACM Conference on Recommender Systems. 57–64

2013
[38]

Hao Tang, Guoshuai Zhao, Yujiao He, Yuxia Wu, and Xueming Qian. 2023. Ranking-based contrastive loss for recommendation systems. Knowledge-Based Systems 261 (2023), 110180

2023
[39]

Douglas Zanatta Ulian, João Luiz Becker, Carla Bonato Marcolin, and Eusebio Scornavacca. 2021. Exploring the effects of different Clustering Methods on a News Recommender System. Expert Systems with Applications 183 (2021), 115341

2021
[40]

Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. Advances in neural information processing systems 30 (2017)

2017
[41]

Dongjing Wang, Xin Zhang, Yao Wan, Dongjin Yu, Guandong Xu, and Shuiguang Deng. 2022. Modeling Sequential Listening Behaviors With Attentive Temporal Point Process for Next and Next New Music Recommendation. IEEE Transactions on Multimedia 24 (2022), 4170–4182

2022
[42]

Wenqiang Wang, Yan XIAO, Xiaojun Jia, Yangshijie Zhang, Peng Chen, Bin Zeng, and Xiaochun Cao. 2026. Dynamic $k$-shot In-Context Learning. https: //openreview.net/forum?id=6XdT4NuIMz WWW ’24, May 13–17, 2024, Singapore, Singapore. Xiping Li et al

2026
[43]

Wenqiang Wang, XIAO Yan, Huiyu Zhou, Peng Chen, Si-Yuan Liang, Xiaochun Cao, et al. [n. d.]. Simplify In-Context Learning. ([n. d.])
[44]

Wenqiang Wang and Yangshijie Zhang. 2025. Incomplete In-context Learning. arXiv preprint arXiv:2505.07251 (2025)

work page arXiv 2025
[45]

Xiang Wang, Xiangnan He, Meng Wang, Fuli Feng, and Tat-Seng Chua. 2019. Neural graph collaborative filtering. In Proceedings of the ACM SIGIR . 165–174

2019
[46]

Youquan Wang, Zhiwen Dai, Jie Cao, Jia Wu, Haicheng Tao, and Guixiang Zhu
[47]

World Wide Web 26, 1 (2023), 71–94

Intra-and inter-association attention network-enhanced policy learning for social group recommendation. World Wide Web 26, 1 (2023), 71–94

2023
[48]

Michael Wolfowicz, David Weisburd, and Badi Hasisi. 2023. Examining the interactive effects of the filter bubble and the echo chamber on radicalization. Journal of Experimental Criminology 19, 1 (2023), 119–141

2023
[49]

Chuhan Wu, Fangzhao Wu, Yongfeng Huang, and Xing Xie. 2023. Personal- ized news recommendation: Methods and challenges. ACM Transactions on Information Systems 41, 1 (2023), 1–50

2023
[50]

Liangwei Yang, Zhiwei Liu, Yu Wang, Chen Wang, Ziwei Fan, and Philip S. Yu
[51]

In Proceedings of the Web Conference

Large-scale Personalized Video Game Recommendation via Social-aware Contextualized Graph Neural Network. In Proceedings of the Web Conference
[52]

Yu, and Taiqing Wang

Liangwei Yang, Shengjie Wang, Yunzhe Tao, Jiankai Sun, Xiaolong Liu, Philip S. Yu, and Taiqing Wang. 2023. DGRec: Graph Neural Network for Recommendation with Diversified Embedding Generation. In Proceedings of the WSDM

2023
[53]

Rui Ye, Yuqing Hou, Te Lei, Yunxing Zhang, Qing Zhang, Jiale Guo, Huaiwen Wu, and Hengliang Luo. 2021. Dynamic Graph Construction for Improving Diversity of Recommendation. In Proceedings of the 15th ACM Conference on Recommender Systems. 651–655

2021
[54]

Jing Yi, Yaochen Zhu, Jiayi Xie, and Zhenzhong Chen. 2023. Cross-Modal Varia- tional Auto-Encoder for Content-Based Micro-Video Background Music Recom- mendation. IEEE Transactions on Multimedia 25 (2023), 515–528

2023
[55]

Qing Yin, Hui Fang, Zhu Sun, and Yew-Soon Ong. 2023. Understanding Diversity in Session-based Recommendation. ACM Transactions on Information Systems 42, 1 (aug 2023), 1–34

2023
[56]

Rex Ying, Ruining He, Kaifeng Chen, Pong Eksombatchai, William L Hamilton, and Jure Leskovec. 2018. Graph convolutional neural networks for web-scale recommender systems. In Proceedings of the 24th ACM SIGKDD . 974–983

2018
[57]

Mingwei Zhang, Guiping Wang, Lanlan Ren, Jianxin Li, Ke Deng, and Bin Zhang
[58]

Knowledge-Based Systems 238 (2022), 107922

METoNR: A meta explanation triplet oriented news recommendation model. Knowledge-Based Systems 238 (2022), 107922

2022
[59]

Yu Zheng, Chen Gao, Liang Chen, Depeng Jin, and Yong Li. 2021. DGCN: Diver- sified Recommendation with Graph Convolutional Networks. In Proceedings of the Web Conference 2021. ACM

2021
[60]

McNee, Joseph A

Cai-Nicolas Ziegler, Sean M. McNee, Joseph A. Konstan, and Georg Lausen. 2005. Improving Recommendation Lists through Topic Diversification. In Proceddings of the World Wide Web 2005. 22–32. A IMPLEMENTATION DETAILS In the experiments, we implement CPGRec using Pytorch and the DGL, which is tailored for deep learning on graphs. More details, we employ the...

2005