Relation pruning and discriminative sampling over knowledge graph for long-tail recommendation

Information Sciences, Journal Year: 2024, Volume and Issue: 679, P. 120871 - 120871

Published: June 5, 2024

Language: Английский

---

Improvement of Reading Platforms Assisted by the Spring Framework: A Recommendation Technique Integrating the KGMRA Algorithm and BERT Model

Heliyon, Journal Year: 2025, Volume and Issue: 11(3), P. e42191 - e42191

Published: Jan. 23, 2025

With the widespread adoption of personalized recommendation systems, traditional methods continue to face significant challenges in areas such as accuracy, user experience, and content diversity. Existing approaches struggle effectively integrate behavior data with semantic information article often fall short covering long-tail meet users' diverse needs. To address these limitations, this study introduces a novel system that combines Knowledge Graph-based Multi-Relational Association (KGMRA) algorithm Bidirectional Encoder Representations from Transformers (BERT) model. The KGMRA leverages knowledge graphs extract rich associative employs multi-relational networks capture relationships between articles effectively. Concurrently, BERT model uses deep learning generate robust representations content, enhancing system's capacity understand predict interests greater precision. By integrating advanced technologies, proposed achieves improvements personalization, Specifically, it excels recommending thereby catering more niche or less popular articles. Experimental results highlight superior performance compared existing baseline methods. Key metrics improved substantially, accuracy increasing 72 % 84 %, coverage rising 71 89 click-through rate growing 79 94 %. Additionally, efficiency by 20 resulting faster response times an enhanced experience. This provides practical effective solution for improving systems on reading platforms built Spring framework. It not only offers application value but also contributes new perspectives optimization innovation future systems.

Language: Английский

---

Exploring Causal Learning Through Graph Neural Networks: An In‐Depth Review

Wiley Interdisciplinary Reviews Data Mining and Knowledge Discovery, Journal Year: 2025, Volume and Issue: 15(2)

Published: May 19, 2025

ABSTRACT In machine learning, exploring data correlations to predict outcomes is a fundamental task. Recognizing causal relationships embedded within pivotal for comprehensive understanding of system dynamics, the significance which paramount in data‐driven decision‐making processes. Beyond traditional methods, there has been shift toward using graph neural networks (GNNs) given their capabilities as universal approximators. Thus, thorough review advancements learning GNNs both relevant and timely. To structure this review, we introduce novel taxonomy that encompasses various state‐of‐the‐art GNN methods used studying causality. are further categorized based on applications causality domain. We provide an exhaustive compilation datasets integral with serve resource practical study. This also touches upon application across diverse sectors. conclude insights into potential challenges promising avenues future exploration rapidly evolving field learning.

Language: Английский

---

Causality-aware social recommender system with network homophily informed multi-treatment confounders

Information Sciences, Journal Year: 2024, Volume and Issue: 676, P. 120729 - 120729

Published: May 15, 2024

Language: Английский

---

O‐SCDR: Optimal cluster with attention based shared‐account cross‐domain sequential recommendation using deep reinforcement learning technique

Expert Systems, Journal Year: 2024, Volume and Issue: 41(8)

Published: Feb. 3, 2024

Abstract Sequential recommendation involves suggesting subsequent items in a series of user activities. When recommending relevant to users within the same account, challenge lies discerning diverse behaviours provide tailored recommendations based on individual preferences and timing. Cross‐domain sequential (CDSR) focuses accurately extracting cross‐domain from both within‐sequence between‐sequence interactions among items. Current approaches typically concentrate learning single domain using intra‐sequence item interactions, followed by transfer module for preferences. However, this process implicit method are constrained effectiveness may overlook inter‐sequence associations. In study, we propose an optimal cluster with attention‐based shared‐account (O‐SCSR) system deep reinforcement techniques. Our approach commences formulating modified hummingbird optimization (MHO) algorithm clustering, effectively identifying latent who share account enhance understanding scenarios. Additionally, design filter quantum classic (QCDRL), intelligently selecting contributing O‐SCSR. By quantifying rewards transferred knowledge, QCDRL‐based retains only valuable task SCDR. Finally, validate efficacy our proposed O‐SCDR real‐world datasets, namely HVIDEO HAMAZON. Through simulation results comparing existing state‐of‐the‐art systems, demonstrate its legitimacy.

Language: Английский

---

Ikcs: An Effective Intelligent K-Medoids Clustering Scheme Based on Granularity Rough Entropy for Social Networks

Published: Jan. 1, 2024

Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI

Language: Английский

---

Spatio-Temporal Contrastive Heterogeneous Graph Attention Networks for Session-Based Recommendation

Mathematics, Journal Year: 2024, Volume and Issue: 12(8), P. 1193 - 1193

Published: April 16, 2024

The main goal of session-based recommendation (SBR) is to analyze the list possible next interaction items through user’s historical sequence. existing session models directly model sequence as a graph, and only consider aggregation neighbor based on spatial structure information, ignoring time information items. sparsity sequences also affects accuracy recommendation. This paper proposes spatio-temporal contrastive heterogeneous graph attention network (STC-HGAT). built hypergraph, latent Dirichlet allocation (LDA) algorithm used construct category nodes enhance contextual semantic hypergraph employed capture session. temporal constructed aggregate item. Then, are fused by sumpooling. Meanwhile, modulation factor added cross-entropy loss function adaptive weight (AW) function. Contrastive learning (CL) an auxiliary task further modeling, so alleviate data. A large number experiments real public datasets show that STC-HGAT proposed in this superior baseline metrics such P@20 MRR@20, improving performance certain extent.

Language: Английский

---

Collaborative Long-Tail Session-Based Recommendations

Published: Jan. 1, 2024

Session-Based Recommendation Systems aim to predict the next item in a session based on items current session. However, there exists bias towards popular or `short-head' items, neglecting less `long-tail items', thus reducing recommendation diversity. This limits user exploration and overlooks value of long-tail which are significant catalog.Existing methods mitigate popularity bias, typically ignore collaborative information from sessions with similar diversity profiles, where is defined as ratio unpopular items. We propose neural network architecture enhance performance without sacrificing accuracy. Our approach integrates into embeddings leverages data by selecting nearest neighbors Experiments three datasets show our model outperforms baselines, up 4.73% accuracy 2.51% coverage improvements, demonstrating method's effectiveness maintaining while addressing challenges.

Language: Английский

---

Popularity-Aware Graph Neural Network with Global Context for Session-Based Recommendation

Lecture notes in computer science, Journal Year: 2024, Volume and Issue: unknown, P. 163 - 171

Published: Jan. 1, 2024

Language: Английский

---

CBRec: A causal way balancing multidimensional attraction effect in POI recommendations

Knowledge-Based Systems, Journal Year: 2024, Volume and Issue: 305, P. 112607 - 112607

Published: Oct. 11, 2024

Language: Английский

---