Grasynda: Graph-based Synthetic Time Series Generation

2026-01-27

AI Summary

This work addresses the challenge of limited time series data, which hinders the generalization of deep learning models, and the inadequacy of existing data augmentation techniques in preserving temporal dynamics. To this end, the authors propose Grasynda, a novel method that reformulates univariate time series into a graph structure, where nodes represent discrete states and directed edges encode state transitions. Temporal evolution is modeled via a transition probability matrix to generate high-fidelity synthetic data. Evaluated across six benchmark datasets with three neural network architectures, Grasynda consistently outperforms current data augmentation approaches, both in preserving the intrinsic dynamics of the original sequences and in enhancing downstream forecasting performance—surpassing the augmentation strategies employed by state-of-the-art time series foundation models.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Planning, Routing, and Scheduling: Temporal Planning
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Data augmentation is a crucial tool in time series forecasting, especially for deep learning architectures that require a large training sample size to generalize effectively. However, extensive datasets are not always available in real-world scenarios. Although many data augmentation methods exist, their limitations include the use of transformations that do not adequately preserve data properties. This paper introduces Grasynda, a novel graph-based approach for synthetic time series generation that: (1) converts univariate time series into a network structure using a graph representation, where each state is a node and each transition is represented as a directed edge; and (2) encodes their temporal dynamics in a transition probability matrix. We performed an extensive evaluation of Grasynda as a data augmentation method for time series forecasting. We use three neural network variations on six benchmark datasets. The results indicate that Grasynda consistently outperforms other time series data augmentation methods, including ones used in state-of-the-art time series foundation models. The method and all experiments are publicly available.

Problem

• time series forecasting
• data augmentation
• synthetic time series
• deep learning
• data scarcity

Innovation

• graph-based generation
• time series augmentation
• transition probability matrix
• synthetic time series
• data augmentation