Solving the apparent diversity-accuracy dilemma of recommender systems

Abstract

Recommender systems use data on past user preferences to predict possible future likes and interests. A key challenge is that while the most useful individual recommendations are to be found among diverse niche objects, the most reliably accurate results are obtained by methods that recommend objects based on user or object similarity. In this paper we introduce a new algorithm specifically to address the challenge of diversity and show how it can be used to resolve this apparent dilemma when combined in an elegant hybrid with an accuracy-focused algorithm. By tuning the hybrid appropriately we are able to obtain, without relying on any semantic or context-specific information, simultaneous gains in both accuracy and diversity of recommendations.

10 pages, 9 figures, 4 tables (final version with supporting information included)

References in corpus (8)

• How to project a bipartite network?
• Collaborative Tagging and Semiotic Dynamics
• Heat Conduction Process on Community Networks as a Recommendation Model
• Personalized Recommendation via Integrated Diffusion on User-Item-Tag Tripartite Graphs
• Effect of initial configuration on network-based recommendation
• Information filtering via Iterative Refinement
• Information Filtering via Self-Consistent Refinement
• Information Flow in Interaction Networks

Cited by in corpus (137)

• Link Prediction in Complex Networks: A Survey
• Vital nodes identification in complex networks
• Recommender Systems
• How Algorithmic Confounding in Recommendation Systems Increases Homogeneity and Decreases Utility
• Hybrid Recommender Systems: A Systematic Literature Review
• Current Challenges and Visions in Music Recommender Systems Research
• Identifying Influential Spreaders by Weighted LeaderRank
• Measuring the Business Value of Recommender Systems
• Ranking in evolving complex networks
• Tag-Aware Recommender Systems: A State-of-the-art Survey
• DGRec: Graph Neural Network for Recommendation with Diversified Embedding Generation
• DGCN: Diversified Recommendation with Graph Convolutional Networks
• Managing Popularity Bias in Recommender Systems with Personalized Re-ranking
• Leveraging Large Language Models for Sequential Recommendation
• Information filtering via biased heat conduction
• Common neighbours and the local-community-paradigm for link prediction in bipartite networks
• Empirical analysis of web-based user-object bipartite networks
• Contextual Hybrid Session-based News Recommendation with Recurrent Neural Networks
• Network-based recommendation algorithms: A review
• Information filtering via preferential diffusion
• Black-Box Testing of Deep Neural Networks Through Test Case Diversity
• A probabilistic model to resolve diversity-accuracy challenge of recommendation systems
• Personalized Bundle List Recommendation
• Investigating Accuracy-Novelty Performance for Graph-based Collaborative Filtering
• Recommender Systems for Online and Mobile Social Networks: A survey
• Diversity and Inclusion Metrics in Subset Selection
• Emergence of scale-free leadership structure in social recommender systems
• Extracting the information backbone in online system
• A Survey on Hyperlink Prediction
• Prediction in complex systems: the case of the international trade network
• Uncovering missing links with cold ends
• Behavior patterns of online users and the effect on information filtering
• Solving the accuracy-diversity dilemma via directed random walks
• Graph-based Collaborative Ranking
• Feature Selection for Recommender Systems with Quantum Computing
• Effective Mechanism for Social Recommendation of News
• Information Filtering via Collaborative User Clustering Modeling
• Recommender systems based on graph embedding techniques: A comprehensive review
• How to be Fair and Diverse?
• A vertex similarity index for better personalized recommendation
• Promoting cold-start items in recommender systems
• Intelligent social bots uncover the link between user preference and diversity of news consumption
• Strong ties promote the epidemic prevalence in susceptible-infected-susceptible spreading dynamics
• Information filtering via hybridization of similarity preferential diffusion processes
• Heter-LP: A heterogeneous label propagation algorithm and its application in drug repositioning
• Efficient Sampling for k-Determinantal Point Processes
• A trust-based recommendation method using network diffusion processes
• Multi-objective Contextual Multi-armed Bandit with a Dominant Objective
• An improved HeatS+ProbS hybrid recommendation algorithm based on heterogeneous initial resource configurations
• The reinforcing influence of recommendations on global diversification
• DeepFM: An End-to-End Wide & Deep Learning Framework for CTR Prediction
• Personalised Travel Recommendation based on Location Co-occurrence
• Analysis of a heterogeneous social network of humans and cultural objects
• Improving information filtering via network manipulation
• Enhancing topology adaptation in information-sharing social networks
• Personalized recommendation with corrected similarity
• Understanding Diversity in Session-Based Recommendation
• OtherTube: Facilitating Content Discovery and Reflection by Exchanging YouTube Recommendations with Strangers
• The essential role of time in network-based recommendation
• A Personalized Recommender System for Pervasive Social Networks
• Heterogenous scaling in interevent time of on-line bookmarking
• Information filtering in sparse online systems: recommendation via semi-local diffusion
• Fixed-point algorithms for learning determinantal point processes
• Ultra accurate collaborative information filtering via directed user similarity
• The long-term impact of ranking algorithms in growing networks
• Diffusion-like recommendation with enhanced similarity of objects
• Influence, originality and similarity in directed acyclic graphs
• Collaborative Filtering Approach to Link Prediction
• Improving personalized link prediction by hybrid diffusion
• Statistical Mechanics of Competitive Resource Allocation using Agent-based Models
• Information Filtering on Coupled Social Networks
• Boosting Item-based Collaborative Filtering via Nearly Uncoupled Random Walks
• Kronecker Determinantal Point Processes
• Modeling mutual feedback between users and recommender systems
• Variation Control and Evaluation for Generative SlateRecommendations
• Click-aware purchase prediction with push at the top
• Improving Sequential Recommendations with LLMs
• Uncovering the Deep Filter Bubble: Narrow Exposure in Short-Video Recommendation
• Practitioners Versus Users: A Value-Sensitive Evaluation of Current Industrial Recommender System Design
• Dynamics underlying Box-office: Movie Competition on Recommender Systems
• Spectral coarse graining for random walk in bipartite networks
• Gravity Effects on Information Filtering and Network Evolving
• A Fast Recommendation Algorithm for Social Tagging Systems : A Delicious Case
• Do recommender systems benefit users?
• Alleviating the recommendation bias via rank aggregation
• Crowd Avoidance and Diversity in Socio-Economic Systems and Recommendation
• PLIERS: a Popularity-Based Recommender System for Content Dissemination in Online Social Networks
• A Flexible Recommendation System for Cable TV
• Adaptive social recommendation in a multiple category landscape
• An Adjustable Heat Conduction based KNN Approach for Session-based Recommendation
• Hypergraph Clustering for Finding Diverse and Experienced Groups
• Predictability of diffusion-based recommender systems
• Collaborative Residual Metric Learning
• Enhancing the long-term performance of recommender system
• Modeling correlated human dynamics
• A Hybrid Approach to Web Service Recommendation Based on QoS-Aware Rating and Ranking
• A survey of food recommenders
• Reducing Popularity Bias in Recommendation Over Time
• Probabilistic Generating Circuits
• Detecting and modelling real percolation and phase transitions of information on social media
• Beyond Accuracy Optimization: On the Value of Item Embeddings for Student Job Recommendations
• One Size Does Not Fit All: Modeling Users' Personal Curiosity in Recommender Systems
• Optimal timescale for community detection in growing networks
• An Evaluation Study of Generative Adversarial Networks for Collaborative Filtering
• Stability of similarity measurements for bipartite networks
• A General Framework for Complex Network Applications
• Membership in social networks and the application in information filtering
• DUM: Diversity-Weighted Utility Maximization for Recommendations
• Improving Recommendation Diversity by Highlighting the ExTrA Fabricated Experts
• Hop Sampling: A Simple Regularized Graph Learning for Non-Stationary Environments
• Assessing Fashion Recommendations: A Multifaceted Offline Evaluation Approach
• Weighted Random Walk Sampling for Multi-Relational Recommendation
• DPPNet: Approximating Determinantal Point Processes with Deep Networks
• Detection of Shilling Attack Based on T-distribution on the Dynamic Time Intervals in Recommendation Systems
• On the Relationship Between Probabilistic Circuits and Determinantal Point Processes
• Reconstructing propagation networks with temporal similarity metrics
• A two-step Recommendation Algorithm via Iterative Local Least Squares
• Learning Instance-Aware Object Detection Using Determinantal Point Processes
• PURS: Personalized Unexpected Recommender System for Improving User Satisfaction
• Information Cocoons in Online Navigation
• Recommending Given Names
• An Empirical Analysis on Transparent Algorithmic Exploration in Recommender Systems
• Memory effect of the online user preference
• Understanding the Effects of Adversarial Personalized Ranking Optimization Method on Recommendation Quality
• Efficient and Optimal Algorithms for Tree Summarization with Weighted Terminologies
• Lexicographic Multiarmed Bandit
• A Survey on GNN-based Link Prediction: Techniques, Applications, and Challenges
• A Fast Parallel Approach for Neighborhood-based Link Prediction by Disregarding Large Hubs
• Data Optimisation for a Deep Learning Recommender System
• Conquering the rating bound problem in neighborhood-based collaborative filtering: a function recovery approach
• Learning Hierarchical Item Categories from Implicit Feedback Data for Efficient Recommendations and Browsing
• Latent Unexpected Recommendations
• Network-based models for social recommender systems
• Enhancing countries' fitness with recommender systems on the international trade network
• Inferring users' preferences through leveraging their social relationships
• Fashion and art cycles are driven by counter-dominance signals of elite competition: quantitative evidence from music styles
• Bridge the gap between network-based inference method and global ranking method in personal recommendation