Qualitative and Quantitative Evaluation of Multivariate Time-Series Synthetic Data Generated Using MTS-TGAN: A Novel Approach

Abstract

To obtain high performance, generalization, and accuracy in machine learning applications, such as prediction or anomaly detection, large datasets are a necessary prerequisite. Moreover, the collection of data is time-consuming, difficult, and expensive for many imbalanced or small datasets. These challenges are evident in collecting data for financial and banking services, pharmaceuticals and healthcare, manufacturing and the automobile, robotics car, sensor time-series data, and many more. To overcome the challenges of data collection, researchers in many domains are becoming more and more interested in the development or generation of synthetic data. Generating synthetic time-series data is far more complicated and expensive than generating synthetic tabular data. The primary objective of the paper is to generate multivariate time-series data (for continuous and mixed parameters) that are comparable and evaluated with real multivariate time-series synthetic data. After being trained to produce such data, a novel GAN architecture named as MTS-TGAN is proposed and then assessed using both qualitative measures namely t-SNE, PCA, discriminative and predictive scores as well as quantitative measures, for which an RNN model is implemented, which calculates MAE and MSLE scores for three training phases; Train Real Test Real, Train Real Test Synthetic and Train Synthetic Test Real. The model is able to reduce the overall error up to 13% and 10% in predictive and discriminative scores, respectively. The research’s objectives are met, and the outcomes demonstrate that MTS-TGAN is able to pick up on the distribution and underlying knowledge included in the attributes of the real data and it can serve as a starting point for additional research in the respective area.

Design of Mathematical Model and Selected Coefficient Specifications for Composite Materials Reinforced with Fabric from Waste Tyres

“…The following metrics were used to assess the individual variants obtained by changing the coefficients: Mean Absolute Error (MAE)—The mean absolute error indicates the average difference between the actual and predicted values. The closer the MAE value is to 0, the more appropriate the prediction [ 22 , 23 , 24 ]. where n is the number of values, y i is the measured value, and as a predicted value.…”

“…Mean Absolute Error (MAE)—The mean absolute error indicates the average difference between the actual and predicted values. The closer the MAE value is to 0, the more appropriate the prediction [ 22 , 23 , 24 ].…”

Exploring Innovative Approaches to Synthetic Tabular Data Generation

“…In an attempt to address the need for large datasets, a recent study introduced an approach with MTS-TGAN, a novel generative adversarial network (GAN) architecture tailored to generate multivariate time-series data closely resembling real-world datasets [26]. The results showed that MTS-TGAN was effective in capturing the distribution and characteristics of real data and potentially reduced errors in predictive and discriminative scores.…”

Implementation of a Sequence-to-Sequence Stacked Sparse Long Short-Term Memory Autoencoder for Anomaly Detection on Multivariate Timeseries Data of Industrial Blower Ball Bearing Units

“…It is worth noting that healthy production line data are comparatively easier to obtain, since fault conditions are less frequent in regard to their occurrence and documentation; thus, simulated datasets can present a way to train an ML model on exact labeled error conditions or to evaluate an already trained ML model, specifically in the case of anomaly detection where only normal operation data are available [ 16 ]. Furthermore, the acquisition of annotated datasets is especially challenging due to the expenses and time needed for the sensitive task of labeling, since experienced human labor must be involved, and the businesses that have carried through these processes understandably need to keep their data confidential [ 17 ].…”