# Time Series Forecasting Using LSTM Networks: A Symbolic Approach

arXiv: https://arxiv.org/abs/2003.05672 · HTML: https://mlsanity.com/paper/2003.05672/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Mar 2020.

## Built on by (most cited first, 25 of 25)

- [A comparison of LSTM and GRU networks for learning symbolic sequences](https://mlsanity.com/paper/2107.02248/) (2021)
- [FedREP: Towards Horizontal Federated Load Forecasting for Retail Energy Providers](https://mlsanity.com/paper/2203.00219/) (2022)
- [Spatio-temporal DeepKriging for Interpolation and Probabilistic Forecasting](https://mlsanity.com/paper/2306.11472/) (2023)
- [LATTE: LSTM Self-Attention based Anomaly Detection in Embedded Automotive Platforms](https://mlsanity.com/paper/2107.05561/) (2021)
- [An efficient aggregation method for the symbolic representation of temporal data](https://mlsanity.com/paper/2201.05697/) (2022)
- [Predicting Hate Intensity of Twitter Conversation Threads](https://mlsanity.com/paper/2206.08406/) (2022)
- [Boosted Embeddings for Time Series Forecasting](https://mlsanity.com/paper/2104.04781/) (2021)
- [Synthesis of standard 12-lead electrocardiograms using two dimensional generative adversarial network](https://mlsanity.com/paper/2106.03701/) (2021)
- [Mini-Batch Learning Strategies for modeling long term temporal dependencies: A study in environmental applications](https://mlsanity.com/paper/2210.08347/) (2022)
- [LLM-ABBA: Understanding time series via symbolic approximation](https://mlsanity.com/paper/2411.18506/) (2024)
- [LTSM-Bundle: A Toolbox and Benchmark on Large Language Models for Time Series Forecasting](https://mlsanity.com/paper/2406.14045/) (2024)
- [Dominant motion identification of multi-particle system using deep learning from video](https://mlsanity.com/paper/2104.12722/) (2021)
- [Message Propagation Through Time: An Algorithm for Sequence Dependency Retention in Time Series Modeling](https://mlsanity.com/paper/2309.16882/) (2023)
- [Predictive AI with External Knowledge Infusion: Datasets and Benchmarks for Stock Markets](https://mlsanity.com/paper/2504.20058/) (2025)
- [Tuning the burn-in phase in training recurrent neural networks improves their performance](https://mlsanity.com/paper/2602.10911/) (2026)
- [ASTRIDE: Adaptive Symbolization for Time Series Databases](https://mlsanity.com/paper/2302.04097/) (2023)
- [Forecasting Crude Oil Prices Using Reservoir Computing Models](https://mlsanity.com/paper/2306.03052/) (2023)
- [A Decision Making Framework for Recommended Maintenance of Road Segments](https://mlsanity.com/paper/2307.10085/) (2023)
- [Deep State-Space Model for Predicting Cryptocurrency Price](https://mlsanity.com/paper/2311.14731/) (2023)
- [Tree-based Learning for High-Fidelity Prediction of Chaos](https://mlsanity.com/paper/2403.13836/) (2024)
- [Advancing Long-Term Multi-Energy Load Forecasting with Patchformer: A Patch and Transformer-Based Approach](https://mlsanity.com/paper/2404.10458/) (2024)
- [QABBA: Symbolic Time-Series Compression via Integer-Quantized Aggregation](https://mlsanity.com/paper/2411.15209/) (2024)
- [Applied Machine Learning Methods with Long-Short Term Memory Based Recurrent Neural Networks for Multivariate Temperature Prediction](https://mlsanity.com/paper/2503.06278/) (2025)
- [Predicting Grain Growth in Polycrystalline Materials Using Deep Learning Time Series Models](https://mlsanity.com/paper/2511.11630/) (2025)
- [Multivariate LSTM-Based Forecasting for Renewable Energy: Enhancing Climate Change Mitigation](https://mlsanity.com/paper/2601.10961/) (2026)

## Abstract (arXiv)

Machine learning methods trained on raw numerical time series data exhibit fundamental limitations such as a high sensitivity to the hyper parameters and even to the initialization of random weights. A combination of a recurrent neural network with a dimension-reducing symbolic representation is proposed and applied for the purpose of time series forecasting. It is shown that the symbolic representation can help to alleviate some of the aforementioned problems and, in addition, might allow for faster training without sacrificing the forecast performance.