High-dimensional time-series data for cybersecurity and attack detection in industrial sensor environments. The dataset contains labeled attack instances for supervised anomaly detection tasks. It is tagged for applications in industrial sensor monitoring and attack detection.
Use Cases
- Train supervised models to detect attack labels in high-dimensional industrial sensor time series.
- Develop unsupervised anomaly detection algorithms for cybersecurity monitoring of sensor data streams.
- Benchmark time-series classification methods on labeled attack detection scenarios.
- Analyze patterns in industrial sensor data preceding and during labeled attack events.
Strengths
- Data is explicitly labeled for attack detection, enabling supervised learning approaches.
- Focuses on industrial sensor cybersecurity, a specialized application domain.
- Tagged as high-dimensional time-series, indicating complex feature spaces for modeling.
Limitations
- Unknown data volume, dimensionality, and temporal coverage limit assessment of scale.
- Lack of sample data or column details prevents verification of data structure and feature relevance.
- Potential class imbalance is likely, as attack events are typically rare compared to normal operation.
Provenance
- Source
- Kaggle
- Collection Method
- null
- Time Range
- null
- Freshness
- null
- Geography
- null