Promoting cold-start items in recommender systems

Abstract

As one of major challenges, cold-start problem plagues nearly all recommender systems. In particular, new items will be overlooked, impeding the development of new products online. Given limited resources, how to utilize the knowledge of recommender systems and design efficient marketing strategy for new items is extremely important. In this paper, we convert this ticklish issue into a clear mathematical problem based on a bipartite network representation. Under the most widely used algorithm in real e-commerce recommender systems, so-called the item-based collaborative filtering, we show that to simply push new items to active users is not a good strategy. To our surprise, experiments on real recommender systems indicate that to connect new items with some less active users will statistically yield better performance, namely these new items will have more chance to appear in other users' recommendation lists. Further analysis suggests that the disassortative nature of recommender systems contributes to such observation. In a word, getting in-depth understanding on recommender systems could pave the way for the owners to popularize their cold-start products with low costs.

6 pages, 6 figures

References in corpus (19)

• Emergence of scaling in random networks
• Link Prediction in Complex Networks: A Survey
• Dynamical and correlation properties of the Internet
• How to project a bipartite network?
• Recommender Systems
• Solving the apparent diversity-accuracy dilemma of recommender systems
• Heat Conduction Process on Community Networks as a Recommendation Model
• Human dynamics revealed through Web analytics
• Tag-Aware Recommender Systems: A State-of-the-art Survey
• Uncovering collective listening habits and music genres in bipartite networks
• Information filtering via biased heat conduction
• Empirical analysis of web-based user-object bipartite networks
• Ultra accurate personalized recommendation via eliminating redundant correlations
• Recommendation model based on opinion diffusion
• Information filtering via preferential diffusion
• Extracting the information backbone in online system
• Behavior patterns of online users and the effect on information filtering
• Anchoring Bias in Online Voting
• Link power coordination for energy conservation in complex communication networks

Cited by in corpus (4)

• A vertex similarity index for better personalized recommendation
• Diffusion-like recommendation with enhanced similarity of objects
• Improving personalized link prediction by hybrid diffusion
• Coupled effects of local movement and global interaction on contagion