Time Series Forecasting Utilizing Automated Machine Learning (AutoML): A Comparative Analysis Study on Diverse Datasets

Abstract

Automated Machine Learning (AutoML) tools are revolutionizing the field of machine learning by significantly reducing the need for deep computer science expertise. Designed to make ML more accessible, they enable users to build high-performing models without extensive technical knowledge. This study delves into these tools in the context of time series analysis, which is essential for forecasting future trends from historical data. We evaluate three prominent AutoML tools—AutoGluon, Auto-Sklearn, and PyCaret—across various metrics, employing diverse datasets that include Bitcoin and COVID-19 data. The results reveal that the performance of each tool is highly dependent on the specific dataset and its ability to manage the complexities of time series data. This thorough investigation not only demonstrates the strengths and limitations of each AutoML tool but also highlights the criticality of dataset-specific considerations in time series analysis. Offering valuable insights for both practitioners and researchers, this study emphasizes the ongoing need for research and development in this specialized area. It aims to serve as a reference for organizations dealing with time series datasets and a guiding framework for future academic research in enhancing the application of AutoML tools for time series forecasting and analysis.

Power Factor Modelling and Prediction at the Hot Rolling Mills’ Power Supply Using Machine Learning Algorithms

“…This study analyzed and obtained RMSE values for power factors ranging from 0.011 to 0.33, which were comparable to those obtained by us. Article [45] discusses the authors' results on time series forecasting utilizing methods such as CNN ANN and RNN ANN on various datasets from diverse applications, similar to the methods we implemented. The datasets most similar to ours are likely those related to electricity, with reported mean squared error (MSE) values ranging from 0.129 to 0.197 and mean absolute error (MAE) values ranging from 0.222 to 0.290.…”

“…Due to RNNs and LSTM networks, they can identify and learn from complex data patterns [42]. As the dataset expands, these models become more accurate and flexible for different forecasting tasks [45]. Deep learning models can generalize well to new, unknown data, making them robust for forecasting in dynamic situations.…”

“…Time series prediction has important theoretical implications as well as various technical applications. Forecasting is a procedure that determines future values of time series data based on past data and can be utilized in applications where estimation is not feasible [36][37][38][39][40][41][42][43][44][45][46][47].…”

“…The accuracy and performance of time series forecasting models can be assessed using several indicators [43][44][45][46]. The type of time series data and forecasting goals influence the evaluation indicators.…”

Benchmarking Automated Machine Learning (AutoML) Frameworks for Object Detection

“…Lenkala et al compared three different AutoML frameworks on three different time-series datasets for epileptic seizure detection [29]. A similar study of the aforementioned was performed by Westergaard et al, with the same AutoML frameworks and three new time-series datasets [30]. Paladino et al similarly examined three AutoML tools on three datasets, but in this case, they were tabular datasets [31].…”