Summary of Time Series Data Augmentation As An Imbalanced Learning Problem, by Vitor Cerqueira et al.
Time Series Data Augmentation as an Imbalanced Learning Problem
by Vitor Cerqueira, Nuno Moniz, Ricardo Inácio, Carlos Soares
First submitted to arxiv on: 29 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Machine Learning (stat.ML)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a novel method for generating synthetic samples of univariate time series data to improve forecasting models’ accuracy. The authors frame this task as an imbalanced learning problem, leveraging oversampling strategies to create synthetic observations and mitigate the issue of limited available data. They demonstrate the effectiveness of their approach on 7 databases containing 5502 univariate time series, showing that it outperforms both global and local models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper develops a new way to make fake time series data for forecasting models. The idea is to use techniques from imbalanced learning to create more training examples when there’s not enough real data available. This helps the model learn better by reducing the impact of limited data. The authors test their approach on many different datasets and find that it works better than other methods. |
Keywords
» Artificial intelligence » Time series
