RECORD
Uncovering multi-order Popularity and Similarity Mechanisms in Link Prediction by graphlet predictors
Social and Information Networks · Submitted 18 Aug 2024
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim The proposed approach improves the accuracy and interpretability of link prediction, thus facilitating the analysis of complex networks.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Link prediction has become a critical problem in network science and has thus attracted increasing research interest. Popularity and similarity are two primary mechanisms in the formation of real networks. However, the roles of popularity and similarity mechanisms in link prediction across various domain networks remain poorly understood. Accordingly, this study used orbit degrees of graphlets to construct multi-order popularity- and similarity-based network link predictors, demonstrating that traditional popularity- and similarity-based indices can be efficiently represented in terms of orbit degrees. Moreover, we designed a supervised learning model that fuses multiple orbit-degree-based features and validated its link prediction performance. We also evaluated the mean absolute Shapley additive explanations of each feature within this model across 550 real-world networks from six domains. We observed that the homophily mechanism, which is a similarity-based feature, dominated social networks, with its win rate being 91\%. Moreover, a different similarity-based feature was prominent in economic, technological, and information networks. Finally, no single feature dominated the biological and transportation networks. The proposed approach improves the accuracy and interpretability of link prediction, thus facilitating the analysis of complex networks.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Link prediction has become a critical problem in network science and has thus attracted increasing research interest. | background 1.00 | 0.04 | 0.01 | asserted 1.00 |
| s1 | Popularity and similarity are two primary mechanisms in the formation of real networks. | background 1.00 | 0.04 | 0.02 | asserted 1.00 |
| s2 | However, the roles of popularity and similarity mechanisms in link prediction across various domain networks remain poorly understood. | gap 1.00 | 0.05 | 0.02 | asserted 0.98 |
| s3 | Accordingly, this study used orbit degrees of graphlets to construct multi-order popularity- and similarity-based network link predictors, demonstrating that traditional popularity- and similarity-based indices can be efficiently represented in terms of orbit degrees. | method 0.62 | 0.96 | 0.18 | asserted 1.00 |
| s4 | Moreover, we designed a supervised learning model that fuses multiple orbit-degree-based features and validated its link prediction performance. | method 0.89 | 0.97 | 0.06 | asserted 1.00 |
| s5 | We also evaluated the mean absolute Shapley additive explanations of each feature within this model across 550 real-world networks from six domains. | method 0.89 | 0.97 | 0.04 | asserted 0.96 |
| s6 | We observed that the homophily mechanism, which is a similarity-based feature, dominated social networks, with its win rate being 91\%. | result 1.00 | 0.96 | 0.20 | asserted 1.00 |
| s7 | Moreover, a different similarity-based feature was prominent in economic, technological, and information networks. | result 1.00 | 0.92 | 0.09 | asserted 1.00 |
| s8 | Finally, no single feature dominated the biological and transportation networks. | result 1.00 | 0.94 | 0.11 | asserted 1.00 |
| s9 | The proposed approach improves the accuracy and interpretability of link prediction, thus facilitating the analysis of complex networks. | interpretation 0.87 | 0.63 | 0.61 | asserted 1.00 |
What was examined, and what was not
- arXiv listing abstract v2
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Link prediction has become a critical problem in network science and has thus attracted increasing research interest. Popularity and similarity are two primary mechanisms in the formation of real networks. However, the roles of popularity and similarity mechanisms in link prediction across various domain networks remain poorly understood. Accordingly, this study used orbit degrees of graphlets to construct multi-order popularity- and similarity-based network link predictors, demonstrating that traditional popularity- and similarity-based indices can be efficiently represented in terms of orbit degrees. Moreover, we designed a supervised learning model that fuses multiple orbit-degree-based features and validated its link prediction performance. We also evaluated the mean absolute Shapley additive explanations of each feature within this model across 550 real-world networks from six domains. We observed that the homophily mechanism, which is a similarity-based feature, dominated social networks, with its win rate being 91\%. Moreover, a different similarity-based feature was prominent in economic, technological, and information networks. Finally, no single feature dominated the biological and transportation networks. The proposed approach improves the accuracy and interpretability of link prediction, thus facilitating the analysis of complex networks.
Human review
References
How the literature uses it
1 arXiv paper cite this in 1 sentence
What it cites · 1 sentence without a confident label
Uses the method or tool of 1 paper
-
To mitigate the biases introduced by the highly imbalanced ratio in training models and evaluating algorithms, we employ the negative sampling method [ref] .
word2vec Explained: deriving Mikolov et al.'s negative-sampling word-embedding method · Results · 0.97
Cites as support 1 paper
-
Notice that it is strictly proved that if Precision, Recall, and F1-score adopt the same threshold to predict missing links (i.e., they all treat top-$L$ link with highest scores as missing links), they will give exactly the same ranking of evaluated algorithms [ref] .
Inconsistency of evaluation metrics in link prediction · Materials and Methods · 0.97
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.