How the use of feature selection methods influences the efficiency and accuracy of complex network simulations

• URL: http://arxiv.org/abs/2412.01096v1
• Date: Mon, 02 Dec 2024 04:12:53 GMT
• Title: How the use of feature selection methods influences the efficiency and accuracy of complex network simulations
• Authors: Katarzyna Musial, Jiaqi Wen, Andreas Gwyther-Gouriotis,
• Abstract summary: Complex network systems' models are designed to emulate real-world networks through the use of simulation and link prediction. This study proposes feature selection methods which utilise unsupervised filtering techniques to rank real-world node features. The chosen method was coined FS-SNS which improved 8 out of 10 simulations of real-world networks.
• Score: 5.946685582099656
• License:
• Abstract: Complex network systems' models are designed to perfectly emulate real-world networks through the use of simulation and link prediction. Complex network systems are defined by nodes and their connections where both have real-world features that result in a heterogeneous network in which each of the nodes has distinct characteristics. Thus, incorporating real-world features is an important component to achieve a simulation which best represents the real-world. Currently very few complex network systems implement real-world features, thus this study proposes feature selection methods which utilise unsupervised filtering techniques to rank real-world node features alongside a wrapper function to test combinations of the ranked features. The chosen method was coined FS-SNS which improved 8 out of 10 simulations of real-world networks. A consistent threshold of included features was also discovered which saw a threshold of 4 features to achieve the most accurate simulation for all networks. Through these findings the study also proposes future work and discusses how the findings can be used to further the Digital Twin and complex network system field.

Related papers

• Leveraging advances in machine learning for the robust classification and interpretation of networks [0.0]
  Simulation approaches involve selecting a suitable network generative model such as Erd"os-R'enyi or small-world. We utilize advances in interpretable machine learning to classify simulated networks by our generative models based on various network attributes.
  arXiv  Detail & Related papers  (2024-03-20T00:24:23Z)
• Going Beyond Neural Network Feature Similarity: The Network Feature Complexity and Its Interpretation Using Category Theory [64.06519549649495]
  We provide the definition of what we call functionally equivalent features. These features produce equivalent output under certain transformations. We propose an efficient algorithm named Iterative Feature Merging.
  arXiv  Detail & Related papers  (2023-10-10T16:27:12Z)
• Geographic Location Encoding with Spherical Harmonics and Sinusoidal Representation Networks [8.765273923374982]
  Recent work embeds coordinates using sine and cosine projections based on Double Fourier Sphere (DFS) features. This work proposes a novel location encoder for globally distributed geographic data that combines spherical harmonic basis functions. We show that both spherical harmonics and sinusoidal representation networks are competitive on their own but set state-of-the-art performances across tasks when combined.
  arXiv  Detail & Related papers  (2023-10-10T16:12:17Z)
• Heterogeneous Feature Representation for Digital Twin-Oriented Complex Networked Systems [13.28255056212425]
  Building models of Complex Networked Systems that can accurately represent reality forms an important research area. This study aims to improve the expressive power of node features in Digital Twin-Oriented Complex Networked Systems.
  arXiv  Detail & Related papers  (2023-09-23T01:40:56Z)
• Provable Guarantees for Nonlinear Feature Learning in Three-Layer Neural Networks [49.808194368781095]
  We show that three-layer neural networks have provably richer feature learning capabilities than two-layer networks. This work makes progress towards understanding the provable benefit of three-layer neural networks over two-layer networks in the feature learning regime.
  arXiv  Detail & Related papers  (2023-05-11T17:19:30Z)
• ElegansNet: a brief scientific report and initial experiments [0.0]
  ElegansNet is a neural network that mimics real-world neuronal network circuitry. It generates improved deep learning systems with a topology similar to natural networks.
  arXiv  Detail & Related papers  (2023-04-06T13:51:04Z)
• Conformer: Local Features Coupling Global Representations for Visual Recognition [72.9550481476101]
  We propose a hybrid network structure, termed Conformer, to take advantage of convolutional operations and self-attention mechanisms for enhanced representation learning. Experiments show that Conformer, under the comparable parameter complexity, outperforms the visual transformer (DeiT-B) by 2.3% on ImageNet.
  arXiv  Detail & Related papers  (2021-05-09T10:00:03Z)
• When Residual Learning Meets Dense Aggregation: Rethinking the Aggregation of Deep Neural Networks [57.0502745301132]
  We propose Micro-Dense Nets, a novel architecture with global residual learning and local micro-dense aggregations. Our micro-dense block can be integrated with neural architecture search based models to boost their performance.
  arXiv  Detail & Related papers  (2020-04-19T08:34:52Z)
• Fitting the Search Space of Weight-sharing NAS with Graph Convolutional Networks [100.14670789581811]
  We train a graph convolutional network to fit the performance of sampled sub-networks. With this strategy, we achieve a higher rank correlation coefficient in the selected set of candidates.
  arXiv  Detail & Related papers  (2020-04-17T19:12:39Z)
• Dense Residual Network: Enhancing Global Dense Feature Flow for Character Recognition [75.4027660840568]
  This paper explores how to enhance the local and global dense feature flow by exploiting hierarchical features fully from all the convolution layers. Technically, we propose an efficient and effective CNN framework, i.e., Fast Dense Residual Network (FDRN) for text recognition.
  arXiv  Detail & Related papers  (2020-01-23T06:55:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.