Figure 8 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Download
View publication
LastFM top ten artists.

LastFM top ten artists.

Source publication
Figure 7. LastFM gender distribution.
Figure 8. LastFM top ten artists.
Figure 9. LastFM distribution of the records on the map.
Figure 19 shows the results of differential fairness for the sensitive...
+3 Figure 19. Results of differential fairness of sensitive attribute...
A Comparative Analysis of Bias Amplification in Graph Neural Network Approaches for Recommender Systems
Preprint
Full-text available
  • Jan 2023
Recommender Systems (RSs) are used to provide users with personalized item recommendations and help them overcome the problem of information overload. Currently, recommendation methods based on deep learning are gaining ground over traditional methods such as matrix factorization due to their ability to represent the complex relationships between u...
Cite
Download full-text

Context in source publication

Context 1
... in the MovieLens dataset, the number of male users is considerably higher than the female users. Figure 8 illustrates the top ten most popular artists regarding the number of records. Figure 9 shows the distribution of records throughout the world map. ...

Similar publications

Fig. 1 Performance change of LightGCN with different graph convolution...
Fig. 2 Average ∥θ θ θ∥ and d 3 2 ∥θ θ θ∥ in each items group. Items are...
Fig. 3 The average aggregation weight of one-order neighbor users in...
Fig. 4 Performance comparison between LightGCN and LightGCN-DAP with...
+4 Fig. 5 Ablation study of DAP with different hyper-parameters α and β on...
How Graph Convolutions Amplify Popularity Bias for Recommendation?
Preprint
Full-text available
  • May 2023
Graph convolutional networks (GCNs) have become prevalent in recommender system (RS) due to their superiority in modeling collaborative patterns. Although improving the overall accuracy, GCNs unfortunately amplify popularity bias -- tail items are less likely to be recommended. This effect prevents the GCN-based RS from making precise and fair reco...
View