Recently, various Internet of Things (IoT) based algorithms and applications have been developed by making use of a substantial amount of sensor data. For example, they have been used in mobile data reception for wireless sensor networks and have been widely applied in urban sustainable development. To optimize the utilization of data from multiple sources for decision-making, it is vital to properly interpret and reuse sensor data from different domains. Since most IoT devices operate in real-world environments, the quality of information and services in the IoT domain can vary over time. The heterogeneity of underlying devices and networks also makes it difficult to provide one-fits-all solutions to represent data and services that emerge from the IoT networks. Building sensor ontology and mapping sensor data to domain ontology is a feasible way to address these issues. Currently, several sensor ontologies have been developed to define the capabilities of the sensors and sensor networks (e.g., Commonwealth Scientific and Industrial Research Organisation (CSIRO) sensor ontology, OntoSensor, sensor webs for mission operations agent (SWAMO), marine metadata interoperability (MMI) device ontology, sensor model language (SensorML) processes, coastal environment sensor network (CESN), wireless sensor networks ontology (WISNO), agent-based middleware approach for mixed mode environments (A3ME) and Ontonym-Sensor.

Different sensor ontologies are developed and maintained independently by different ontology engineers. The same sensor concept might be represented with different terminologies, granularities, or contexts, which raises the heterogeneity problem to a higher level. The sensor ontology heterogeneity problem brings significant challenges to data integration, data fusion, and discovery mechanisms that require interoperable and machine-interpretable data and quality descriptions. Thus, there is an urgent need to provide mechanisms to integrate and exchange knowledge from heterogeneous sensor ontologies. In particular, we need to provide techniques to enable the processing, interpretation and sharing of sensor data from IoT which use different data models, or whose information is organized into different ontological schemes.

The aim of this Special Issue is to explore recent advancements in using Machine Learning techniques (e.g., support vector machine (SVM), decision tree (DT), random forest (RF), etc.). Submissions also discussing deep learning techniques (e.g., convolutional neural network (CNN), recurrent neural network (RNN), long-short term memory (LSTM) are also welcome. Finally, this Special Issue encourages submission including optimization techniques (e.g., evolutionary algorithms, swarm intelligence, etc.) for integrating heterogeneous sensor information and knowledge.

Potential topics include but are not limited to the following:

• Machine Learning, deep learning and optimization-based sensor knowledge modelling and representation

• Machine learning, deep learning and optimization-based sensors

• Ontology engineering and sensor data annotation

• Machine learning, deep learning and optimization-based sensor ontology alignment and linked sensor data integration

• Machine learning, deep learning and optimization-based applications of semantic sensors data annotation and integration

• Machine learning, deep learning and optimization-based applications of sensor data storage and management