Fun-TSG: A Function-Driven Multivariate Time Series Generator with Variable-Level Anomaly Labeling

TL;DR

Fun-TSG is a customizable multivariate time series generator that provides detailed anomaly labels and dependency structures to improve evaluation and comparison of anomaly detection methods.

Contribution

It introduces a fully transparent, flexible tool for generating diverse, interpretable, and reproducible time series datasets with fine-grained anomaly annotations.

Findings

Supports automated and manual data generation with explicit dependency and anomaly configurations.

Provides ground-truth labels at variable and timestamp levels for precise evaluation.

Enables comprehensive benchmarking of anomaly detection models.

Abstract

Reliable evaluation of anomaly detection methods in multivariate time series remains an open challenge, largely due to the limitations of existing benchmark datasets. Current resources often lack fine-grained anomaly annotations, do not provide explicit intervariable and temporal dependencies, and offer little insight into the underlying generative mechanisms. These shortcomings hinder the development and rigorous comparison of detection models, especially those targeting interpretable and variable-specific outputs. To address this gap, we introduce Fun-TSG, a fully customizable time series generator designed to support high-quality evaluation of anomaly detection systems. Our tool enables both fully automated generation, based on randomly sampled dependency structures and anomaly types, and manual generation through user-defined equations and anomaly configurations. In both cases, it provides full transparency over the data generation process, including access to ground-truth anomaly labels at the variable and timestamp levels. Fun-TSG supports the creation of diverse, interpretable, and reproducible benchmarking scenarios, enabling fine-grained performance analysis for both classical and modern anomaly detection models.