Monitor your devices with smart algorithms to detect anomalies in your data
Easy to consume cloud service for identifying devices
sending implausible data or behaving abnormally
What is Bosch IoT Analytics – Anomaly Detection?
Our Bosch IoT Analytics services simplify the task of analyzing field data. In contrast to the usual, more project-oriented approach, our cloud services offer a cost-efficient way of addressing recurring analytics functions.
Our Anomaly Detection service enables you to identify devices that are sending implausible data or behaving abnormally. While the former can provide the basis for sensor data validation, the latter could be the first step toward developing a predictive maintenance solution.
This cloud service allows you to parameterize your anomaly detection jobs with a wizard – and does not even require any prior data analytics experience. Several analysis types such as Multidimensional Scaling (MDS), One-Class SVM Fit, and Elliptic Envelop Fit are available out-of-the-box. Data scientists can also import their own Python-based implementations of machine learning algorithms and data preprocessing scripts.
Key features of Bosch IoT Analytics
- Anomaly detection wizard: Parameterize anomaly detection jobs – no prior data analytics experience needed.
- Data import & export: Import/export data from/to a local file system (e.g. CSV files) or from/to the standard databases in Bosch IoT Cloud or AWS (e.g. MongoDB, MySQL), as well as from/to Bosch IoT Insights.
- Semantic enrichment & aggregation: Preprocess your data, add semantically richer features, aggregate and normalize time series.
- Visualization: Explore anomalies by visualizing the results in a predefined dashboard.
- Bosch IoT Suite integration: Bosch IoT Analytics – Anomaly Detection allows device data generated by connected devices to be analyzed using the Bosch IoT Remote Manager or Bosch IoT Things; anomaly detection models can also be deployed on the edge and interact with Bosch IoT Gateway Software.
- Open source foundation: Bosch IoT Analytics relies on various Python libraries, for instance Pandas and Scikit-learn.
- Available on: Bosch IoT Analytics – Anomaly Detection is available as a fully managed, shared cloud service in the Bosch IoT Cloud and on Amazon Web Services.
Service plans at a glance
Bosch IoT Analytics – Anomaly Detection is available as a free plan for evaluation purposes and as a starter plan.
|Ready-to-use algorithms and dashboard|
|“Start now” execution option|
|Execution window (with shared resources)||5 GB RAM/instance||5 GB RAM/instance|
|Execution time||2 hours/month||8 hours/month|
|Data upload||0.2 GB||2 GB|
|Subscribe||Subscribe via AWS Marketplace|