Entities in Internet of Things (IoT) need intelligent associations to allow a flexible and dynamic system reaction towards varying user situations. Purely data-oriented association methods (e.g., data-mining, machine-learning, etc.) are limited in deduction from existing associations. Also such data-oriented methods are limited in accuracy for working with small-scale datasets (e.g., working with patterns retrieved from historical data), outputting associations based on statistics rather than logic. Moreover, existing semantic technologies (ontological-oriented or rule-oriented) are facing with either flexibility or dynamicity challenges to discover and maintain the associations. This paper proposes an alternative technique of semantically constructing context-centric data objects based on service logics for logical associations, which enables an event net based on association nets adapting for the changing situations (called context-centric). A proof-of-concept implementation is carried out based on a vehicle planning scenario to validate the data construction technique. Comparing to previous work, this technique possesses advantages of flexibility and dynamicity for entity associations based on service logics.

In Ambient Assisted Living (AAL), as an important applied area of Internet of Things (IoT) technology, logical entity interaction establishes an intelligent virtual world in which heterogeneous entities are emulated by object-oriented mirrors through tightly cooperating sensors and actuators from different spaces in the real world; this enables the system to generate intelligent services to support human behaviour. However, until now, few researchers have focused on enabling interactions between heterogeneous IoT entities for AAL services. Further, in relevant studies regarding semantic entity interactions, researchers have mainly focused on enabling semantic interactions between sensors (rather than integral entities) for the purpose of coordinating and managing data resources, where tight collaborations between sensors and actuators are not emphasised. Hence, the system capability of initialising intelligent service to enhance human experience has not been properly emphasised, to date. This paper discusses enabling logical object-oriented IoT entity interactions to enrich AAL services for humans through the creation of mirrors in the virtual world emulating the attributes and behaviours of heterogeneous entities via cooperating sensors and actuators in the real world. This paper introduces the Entity Device Collocating (EDC) Platform to globally locate and retrieve sensors and actuators respectively adhering to the attributes and behaviours of integral mirrors in the virtual world; this is intended to enable the interoperability between virtual and real worlds. Interactions are created upon entity mirrors mapping entities from the physical world to the virtual world, during which interactions continue to evolve based on the service logics. The proposed technique is applied to a typical AAL case of smart pill-ingestion, where a demo is implemented for verification. Comparing with relevant research, the proposed technique contributes by introducing a flexible and scalable IoT entity interaction method targeting AAL to address human needs.

The IoT paradigm proposes to connect entities intelligently with massive heterogeneous nature, which forms an ocean of devices and data whose complexity and volume are incremental with time. Different from the general big data or IoT, the data-intensive feature of IoT introduces several specific challenges, such as circumstance dynamicity and uncertainties. Hence，intelligence techniques are needed in solving the problems brought by the data intensity. Until recent, there are many different views to handle IoT data and different intelligence enablers for IoT, with different contributions and different targets. However, there are still some issues have not been considered. This paper will provide a fresh survey study on the data-intensive IoT issue. Besides that, we conclude some shadow issues that have not been emphasized, which are interesting for the future. We propose also an extended big data model for intelligent data-intensive IoT to tackle the challenges.

Internet of Things (IoT) is facing with the interoperability issue due to the massive amount of heterogeneous entities (both physical and virtual entities) constantly generating heterogeneous data objects; semantic formalization has been widely recognized as a basis for the IoT interoperability, by which IoT can acquire the ability to comprehend data and further recognize the logic relations among heterogeneous IoT entities and heterogeneous data objects, thus to establish mutual understanding between each other to support with interoperability. Even semantic-driven track has emphasizes a lot on the logic relations in connection to the service rules and policies for interoperability, it is important that the quantity-driven relations should be also explored with adhering to the framework of semantic formalization. This paper explores a Deep Recursive Auto-encoders formed data relation learner in line with the semantic framework, which supports the data interoperability enhancement in a quantity-driven way based on the logic-driven framework. The learner starts with representing the virtual IoT entities via feature extraction; based on that, learner is trained in a manner of considering the surrounding relations of the targeted entity. As a baseline, a contrast learner with "regular" structure has been proposed which cannot functionally support semantic framework, even though the semantic formalization is indispensable; regardless the limitations in lab environment, the conducted experiments show that the proposed learner performs a bit better than the contrast learner under the same conditions. So that, the proposed method can synergistically enhances the interoperability within a semantic formalization framework.