This paper introduces a large-scale, high-resolution traffic dataset collected from over 5,000 sensors across the city of Madrid, Spain, spanning a ten-year period, from 2015 to 2024. Comprising more than 1.5 billion records, the dataset includes key traffic metrics such as intensity, occupancy, and average speed, aggregated at 15-minute intervals. It also features two distinct graph-based spatial representations of the sensor network, enabling advanced modelling of urban mobility. The dataset is designed to support a wide range of machine learning tasks, including spatio-temporal forecasting, representation learning, and transfer learning, while also serving broader applications in urban planning and smart city development. Its geographic diversity offers a valuable alternative to existing datasets predominantly sourced from California, enhancing model generalization and reducing regional bias. Additionally, traffic data from this dataset can be used as a covariate in related domains such as air quality control, supporting multi-modal approaches to urban sustainability.
David María-Arribas, J. Pantrigo, Alfredo Cuesta-Infante· Scientific Data· 0 citations
Large language models (LLMs) offer several distinctive advantages over other machine learning models. First, they are trained as general-purpose models and are readily available, which eliminates the need for task-specific training and allows them to improve rapidly over time. Second, they can be applied directly, without constructing domain-specific or signal-specific models. Third, they are easy to integrate into existing systems and can be deployed without requiring an additional training step. Finally, they are inherently interactive because users can direct them with natural language. In this paper, we investigate whether LLMs can achieve multivariate anomaly detection. To fully exploit the aforementioned benefits, we define a set of guiding principles (such as avoiding pre-learning or representation learning on the signals) to ensure the LLMs remain general-purpose models. Based on these principles, we then propose several algorithmic approaches for building multivariate anomaly detection pipelines. We compare our approaches with two alternatives: (i) classical deep learning pipelines trained specifically for anomaly detection, and (ii) a foundation-model-based approach, in which domain-specific or general purpose time-series foundation models are trained without explicit supervision for anomaly detection but are then used for this purpose. The comparison highlights trade-offs along three key dimensions: anomaly detection accuracy, computational cost, and the amount of domain knowledge required to develop the pipeline. We evaluate our methods through two case studies. The first uses a benchmarking testbed designed for anomaly detection, while the second examines real-world data from wind turbines with known anomalous events.
Allen Baranov, Sarah Alnegheimish, Alfredo Cuesta-Infante et al.· PHM Society European Confere...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
