Behavior patterns of online users and the effect on information filtering
arXiv:1107.1900 · doi:10.1016/j.physa.2011.09.038
Abstract
Understanding the structure and evolution of web-based user-object bipartite networks is an important task since they play a fundamental role in online information filtering. In this paper, we focus on investigating the patterns of online users' behavior and the effect on recommendation process. Empirical analysis on the e-commercial systems show that users have significant taste diversity and their interests for niche items highly overlap. Additionally, recommendation process are investigated on both the real networks and the reshuffled networks in which real users' behavior patterns can be gradually destroyed. Our results shows that the performance of personalized recommendation methods is strongly related to the real network structure. Detail study on each item shows that recommendation accuracy for hot items is almost maximum and quite robust to the reshuffling process. However, niche items cannot be accurately recommended after removing users' behavior patterns. Our work also is meaningful in practical sense since it reveals an effective direction to improve the accuracy and the robustness of the existing recommender systems.
arXiv comment8 pages, 6 figures
References in corpus (28)
- Community structure in social and biological networks
- Assortative mixing in networks
- Link Prediction in Complex Networks: A Survey
- The worldwide air transportation network: Anomalous centrality, community structure, and cities' global roles
- Structural Vulnerability of the North American Power Grid
- Modeling the scaling properties of human mobility
- How to project a bipartite network?
- Detecting rich-club ordering in complex networks
- Solving the apparent diversity-accuracy dilemma of recommender systems
- Link Prediction Based on Local Random Walk
- Statistical Analysis of Airport Network of China
- Prominence and control: The weighted rich-club effect
- Diffusion of scientific credits and the ranking of scientists
- The Spontaneous Emergence of Social Influence in Online Systems
- Heat Conduction Process on Community Networks as a Recommendation Model
- Driven polymer translocation through nanopores: slow versus fast dynamics
- Information filtering via biased heat conduction
- Local structure of directed networks
- Empirical analysis of web-based user-object bipartite networks
- Extracting Hidden Information from Knowledge Networks
- Information filtering via preferential diffusion
- Information Filtering via Self-Consistent Refinement
- Exact Solution for the Time Evolution of Network Rewiring Models
- Highly accurate recommendation algorithm based on high-order similarities
- An improved HeatS+ProbS hybrid recommendation algorithm based on heterogeneous initial resource configurations
- Scaling properties in spatial networks and its effects on topology and traffic dynamics
- Dynamics on Spatial Networks and the Effect of Distance Coarse Graining
- Unevenness of Loop Location in Complex Networks
Cited by in corpus (14)
- Recommender Systems
- Query expansion techniques for information retrieval: A survey
- An agent-based model for emotion contagion and competition in online social media
- Extracting the information backbone in online system
- Evaluating user reputation in online rating systems via an iterative group-based ranking method
- Anchoring Bias in Online Voting
- Promoting cold-start items in recommender systems
- Strong ties promote the epidemic prevalence in susceptible-infected-susceptible spreading dynamics
- A trust-based recommendation method using network diffusion processes
- An improved HeatS+ProbS hybrid recommendation algorithm based on heterogeneous initial resource configurations
- Improving information filtering via network manipulation
- Spectral coarse graining for random walk in bipartite networks
- Adaptive social recommendation in a multiple category landscape
- Membership in social networks and the application in information filtering