![[2108.02431] AutoLL: Automatic Linear Layout of Graphs based on Deep Neural Network](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2108.02431] AutoLL: Automatic Linear Layout of Graphs based on Deep Neural Network
Summary
The paper presents AutoLL, a novel method for automatic linear layout of graphs using deep neural networks, enhancing the reordering of adjacency matrices for better visualization.
Why It Matters
AutoLL addresses limitations in existing graph layout techniques by employing a data-driven approach to feature extraction, making it applicable to a wider range of graph structures. This advancement can significantly improve graph visualization in various fields, including data science and machine learning.
Key Takeaways
- AutoLL introduces a one-mode linear layout method for graph visualization.
- The method utilizes two neural network models for directed and undirected networks.
- Experimental results demonstrate the effectiveness of AutoLL in reordering adjacency matrices.
Statistics > Machine Learning arXiv:2108.02431 (stat) [Submitted on 5 Aug 2021 (v1), last revised 13 Feb 2026 (this version, v2)] Title:AutoLL: Automatic Linear Layout of Graphs based on Deep Neural Network Authors:Chihiro Watanabe, Taiji Suzuki View a PDF of the paper titled AutoLL: Automatic Linear Layout of Graphs based on Deep Neural Network, by Chihiro Watanabe and 1 other authors View PDF HTML (experimental) Abstract:Linear layouts are a graph visualization method that can be used to capture an entry pattern in an adjacency matrix of a given graph. By reordering the node indices of the original adjacency matrix, linear layouts provide knowledge of latent graph structures. Conventional linear layout methods commonly aim to find an optimal reordering solution based on predefined features of a given matrix and loss function. However, prior knowledge of the appropriate features to use or structural patterns in a given adjacency matrix is not always available. In such a case, performing the reordering based on data-driven feature extraction without assuming a specific structure in an adjacency matrix is preferable. Recently, a neural-network-based matrix reordering method called DeepTMR has been proposed to perform this function. However, it is limited to a two-mode reordering (i.e., the rows and columns are reordered separately) and it cannot be applied in the one-mode setting (i.e., the same node order is used for reordering both rows and columns), owing to the characte...
