Summary of Phlp: Sole Persistent Homology For Link Prediction – Interpretable Feature Extraction, by Junwon You et al.
PHLP: Sole Persistent Homology for Link Prediction – Interpretable Feature Extraction
by Junwon You, Eunwoo Heo, Jae-Hun Jung
First submitted to arxiv on: 23 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computational Geometry (cs.CG); Algebraic Topology (math.AT); Machine Learning (stat.ML)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a novel link prediction (LP) method called Persistent Homology Link Prediction (PHLP) that uses topological data analysis to interpret the features used in LP. The authors employ persistent homology (PH) to analyze the topological information of graphs and develop a new node labeling technique called degree double radius node labeling (Degree DRNL). PHLP outperforms state-of-the-art (SOTA) models on most benchmark datasets when used as a standalone classifier, and incorporating its outputs into existing SOTA models further improves performance across all datasets. This approach enables the identification of crucial factors for improving LP performance. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps us better understand how to predict links between nodes in graphs. Graph neural networks (GNNs) are good at this job, but it’s hard to see why they work so well because GNNs are complex. The authors use a different approach called persistent homology (PH) to analyze the graph and figure out what features are important for prediction. They develop a new way of labeling nodes called degree double radius node labeling (Degree DRNL). This method is good at predicting links and works with most datasets. When used with other methods, it even gets better! This paper shows that we can learn more about how to predict links without using GNNs. |
