Detecção de padrões / anomalias em séries temporais

Abstract

Within the Maia City Council Data Lake activities, multiple time series are collected, including energy consumption, water consumption, weather conditions, traffic, and urban mobility data. This work aims to study pattern and anomaly detection methods for time series, evaluating different statistical and machine learning approaches. The goal is to test Python libraries in different application scenarios, design benchmarking mechanisms based on execution time and false positive/negative rates, and assess the effectiveness of the proposed solutions on real-world data.