An Empirical Evaluation of the t-SNE Algorithm for Data Visualization in Structural Engineering
Эмпирическая оценка алгоритма t-SNE для визуализации данных в строительной инженерии
2021-12-01
SCID: 54.1/83kthfwd
Discuss with AI
SMOTEearthquake engineeringimbalanced dataneural network classifierst-SNE algorithm
Figures from the paper
Abstract (AI)
A fundamental task in machine learning involves visualizing high-dimensional data sets that arise in high-impact application domains. When considering the context of large imbalanced data, this problem becomes much more challenging. In this paper, the t-Distributed Stochastic Neighbor Embedding (t-SNE) algorithm is used to reduce the dimensions of an earthquake engineering related data set for visualization purposes. Since imbalanced data sets greatly affect the accuracy of classifiers, we employ Synthetic Minority Oversampling Technique (SMOTE) to tackle the imbalanced nature of such data set. We present the result obtained from t-SNE and SMOTE and compare it to the basic approaches with various aspects. Considering four options and six classification algorithms, we show that using t-SNE on the imbalanced data and SMOTE on the training data set, neural network classifiers have promising results without sacrificing accuracy. Hence, we can transform the studied scientific data into a two-dimensional (2D) space, enabling the visualization of the classifier and the resulting decision surface using a 2D plot.
Key Findings
1
Across four options and six classification algorithms, combining t-SNE on imbalanced data with SMOTE on training data produces promising neural-network results without sacrificing accuracy.
2
SMOTE is used to address the dataset’s class imbalance, particularly by oversampling the training data.
3
The proposed workflow enables scientific structural-engineering data and classifier decision boundaries to be represented in interpretable 2D plots.
4
The study applies t-SNE to reduce an earthquake-engineering dataset to two dimensions for visualizing classifier behavior and decision surfaces.
Research Object
an imbalanced earthquake-engineering-related dataset
Research Subject
the effectiveness of t-SNE-based two-dimensional visualization and SMOTE-assisted classification performance across classification algorithms
Publication Details
Publication Date
2021-12-01
Journal
Publisher
ISSN
Cited by
66
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest