The need for efficient decentralized recommender systems has been appreciated for some time, both for the intrinsic advantages of decentralization and the necessity of integrating recommender systems into P2P applications. On the other hand, the accuracy of recommender systems is often hurt by data sparsity. In this paper, we compare different decentralized user-based and item-based Collaborative Filtering (CF) algorithms with each other, and propose a new user-based random walk approach customized for decentralized systems, specifically designed to handle sparse data. We show how the application of random walks to decentralized environments is different from the centralized version. We examine the performance of our random walk approach in different settings by varying the sparsity, the similarity measure and the neighborhood size. In addition, we introduce the popularizing disadvantage of the significance weighting term traditionally used to increase the precision of similarity measures, and elaborate how it can affect the performance of the random walk algorithm. The simulations on MovieLens 10,000,000 ratings dataset demonstrate that over a wide range of sparsity, our algorithm outperforms other decentralized CF schemes. Moreover, our results show decentralized user-based approaches perform better than their item-based counterparts in P2P recommender applications.
We propose FlexGD, a force-directed algorithm for straightline undirected graph drawing. The algorithm strives to draw graph layouts encompassing from uniform vertex distribution to extreme structure abstraction. It is flexible for it is parameterized so that the emphasis can be put on either of the two drawing criteria. The parameter determines how much the edges are shorter than the average distance between vertices. Extending the clustering property of the LinLog model, FlexGD is efficient for cluster visualization in an adjustable level. The energy function of FlexGD is minimized through a multilevel approach, particularly designed to work in contexts where edge length distribution is not uniform. Applying FlexGD on several real datasets, we illustrate both the good quality of the layout on various topologies, and the ability of the algorithm to meet the addressed drawing criteria.
Much efforts have lately been concentrated on increasing the precision of recommendations following the Netflix Prize competition. Recently, many researchers and industries have noted that other factors like adequate presentation of the results can add more utility to a recommender system than slight improvement in the precision. In this paper, we suggest a methodology for user-friendly representation of recommendations to the end users. Our scheme unifies the two objectives of prediction and visualization in the core of a unique approach. Users and items are first embedded into a high dimensional latent feature space according to a predictor function, particularly designated to meet visualization requirements. The data is then projected into a 2-dimensional space by Curvilinear Component Analysis (CCA). CCA draws personalized Item Maps (PIMs) representing a small subset of items to the active user. The intra-item semantic correlations are preserved in PIMs which is inherited from the clustering property of the high-dimensional embedding space. Our prediction function and the projection method are both non-linear to increase the clarity of the maps and to limit the effect of projection error. The algorithms are tested on three versions of the MovieLens dataset and the Netflix dataset to show they combine good accuracy with satisfactory visual properties. Moreover, a novel offline methodology is suggested to measure the quality of PIMs.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.