Ostad: An Innovative Toolkit for Detecting Anomalies in Dynamic Univariate Time Series Data

Anomaly detection plays a crucial role in identifying malfunctions and preventing future issues across various technological domains. Despite significant advancements in this field, the rapid evolution of technology continues to present new challenges that necessitate the refinement of existing detection techniques. This paper introduces the ostad package, which implements seven state-of-the-art anomaly detection algorithms specifically designed to address these challenges, including online and non-stationary univariate time-series anomaly detection. This paper discusses the capabilities of each algorithm, their applicability to diverse datasets, and the package's user-friendly interface, which facilitates integration into existing workflows. By providing robust tools for detecting anomalies in dynamic environments, the ostad package aims to enhance the reliability of systems and contribute to ongoing research in anomaly detection methodologies,