GraphTSNE : a visualization technique for graph-structured data
We present GraphTSNE, a novel visualization technique for graph-structured data based on t-SNE. The growing interest in graph-structured data increases the importance of gaining human insight into such datasets by means of visualization. However, among the most popular visualization techniques, clas...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/77030 |
| Summary: | We present GraphTSNE, a novel visualization technique for graph-structured data based on t-SNE. The growing interest in graph-structured data increases the importance of gaining human insight into such datasets by means of visualization. However, among the most popular visualization techniques, classical t-SNE is not suitable on such datasets because it has no mechanism to make use of information from graph connectivity. Our proposed method GraphTSNE is able to produce visualizations which account for both graph connectivity and node features. It is based on unsupervised training of a graph convolutional network on a modified tSNE loss. The network is trained in a scalable fashion and can be used inductively to project unseen data points. By assembling a suite of evaluation metrics, we demonstrate that our method outperforms existing visualization techniques on graph datasets and produces desirable visualizations on benchmark datasets. |
|---|