An Examination of Implicit Trust and Influence in Social Recommendation Through Graph Convolutional Networks
Bibliographic record
Abstract
In the realm of social recommendation, the utilization of Graph Convolution Networks (GCNs) has proven effective for embedding propagation, allowing user-item implicit social relations to be modeled efficiently.However, the shortcomings of these existing models lie in their focus on local neighbors and their lack of simulation of recursive diffusion in a broader social network, thus limiting their performance potential.To address this gap, a novel GCN framework, herein referred to as the Trust and Influence Graph Convolution Network (TIGCN), is proposed.This framework aims to improve the robustness of social recommendation systems by utilizing implicit social relations and user-item interactions.Through the construction of user-user trust and influence graphs derived from a bipartite social network, influential users are identified using the Structural Holes method.The TIGCN framework then employs these inter-user relationships, including trust and influence features, to collectively navigate the propagation of user interests and social relations.The effectiveness of the TIGCN is demonstrated through experiments conducted on real-world datasets such as Ciao, Epinions, and FilmTrust.Results show that the TIGCN offers significant performance improvements over other state-of-the-art baselines like FST, FSTID, and SocialLGN.The metrics indicating these improvements include increased Precision@5 (up to 2.08%), Recall@5 (up to 2.7%), NDCG@5 (up to 2.28%), Precision@10 (up to 2.65%), Recall@10 (up to 2.87%), and NDCG@10 (up to 2.82%).In conclusion, the introduction of the TIGCN has opened up promising avenues for the enhancement of social recommendation through the incorporation of implicit social relations and the innovative use of GCNs.Future studies should focus on improving the scalability and effectiveness of the TIGCN framework to maximize its contribution to social recommendation systems.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".