IoT Analytics in Action
IoT Analytics Explained
AGT’s cloud-based IoT analytics platform’s capabilities are organized in a hierarchical structure that consists of multiple layers, each with increasing analytical power and corresponding business value. Each layer highlights both the platform’s value and its distinction from traditional business analytics.
The platform handles many different data types, corresponding to the “things” that produced the data in the first place. The IoT data types encode a richer context that AGT’s IoT analytics platform processes and analyzes.
The first layer of the platform offers analytics applied directly to raw IoT sensor data (often deployed at the edge). It collects, integrates, cleanses and filters data from a wide spectrum of IoT sensors and devices. It then applies additional processing, such as feature extraction and some low-level analytics (e.g., applying video analytics to extract the location of a moving object in a video stream, which can then be used to track that object). The output of this layer is more meaningful than the raw data, yet still requires further analytics to deliver additional context and value to the end user.
The next layer of analytics produces richer context and meaning, presenting higher value for the user. It applies data fusion and includes richer analytics, such as pattern recognition, event classification, behavior/routine learning and anomaly detection, activity recognition, object tracking clustering and more.
The third layer of analytics includes complex operational capabilities, which creates significantly more business value. It includes model-based and data-driven prediction analytics, as well as optimization and simulation.