International Business Machines Corporation
INTERNET OF THINGS SENSOR EQUIVALENCE ONTOLOGY

Abstract:

First and second sets of sensor data from first and second IoT device sensors are collected. A first set of significant sensor data representing a first event is extracted from the first set of sensor data. A set of terms comprising portions of the first set of significant sensor data is weighted. By analyzing the weighted set of terms, a first set of critical variables describing the first event is identified and added to an ontology. Similarly, a second set of significant sensor data is extracted and a second set of critical variables describing the second event is identified and added to the ontology. Using the ontology, it is determined that the first event is of an event type of the second event. The first sensor and the second sensor are classified to be different variants of a class of sensors that is configurable to sense the event type.