Enabling AI Technologies for Megadata Mining in the Internet of Cognitive Things (SI-AIoCT)

Currently, there are many variations and versions of IoT (e.g., Internet of People (IoP), Internet of Sheep (IoS), Internet of Tomato (IoTo), Internet of Drones (IoD), etcetera), to name a few. The rapid advancement in related “data generation/production” tools and technologies have enabled the speedy transformation from “metadata” to “big data”, and lately given rise to the mind-blowing volume of the so-called “megadata”. Megadata necessitates burdensome resources and capabilities in order to aggregate, analyse, visualise and harvest knowledge from collected data. Besides, megadata emitted from the above-mentioned IoT-oriented applications and domains comes in different shapes, sizes, presentations, and formats – depending purely on the application domain and system features. For example, in Internet of Tomato (IoTo), data concerns about best time of implantation of tomato seeds, growth, and the harvesting season etc; whereas Internet of Sheep (IoS) alarms shepherd how much a sheep eats and drinks, or if a sheep is close-by the fence. Obviousley, there is no direct link between the two applications of IoTo and IoS, and their produced data – they work in silos.

To address the above specific limitation, the Internet of (Cognitive) Things (IoCT) has been introduced in a bid to allow various ‘sensors-empowered IoT-enabled’ objects/systems with cognitive capabilities, such as reasoning, learning, explaining and acting, to work collaboratively together. This implies that things can build up dynamic communities with their peers’ systems in a business context as/when necessary to respond to emergent situations. The on-trend IoCT works at unlocking the siloed data via exploiting data analysis and mining algorithms of Machine Learning (ML), Artificial Intelligence (AI), deep and reinforcement learning. Thereof, the IoCT promotes obtaining conversant results for consumers as and when required. Notwithstanding, there are still many overlooked issues and challenges that hinder the full realize and benefits of ML and AI for IoCT big data analysis, mining, processing, and prediction.

The aim of this Topical Collection is to seek the latest research results from academia, industry, and individuals , reporting their findings on selected areas of integrating ML and AI with various versions of IoT and megadata to share their latest ideas and research findings within the research community.

Topics of Interest: The Topical Collection aims to cover topics that include, but are not limited to, the following:

(a) Applied ML and AI technologies for IoCT;

(b) IoCT big data fusion techniques;

(c)Energy efficiency and power management of IoCT;

(d) IoCT megadata security;

(e) IoCT cyber-physical system security;

(f) AI and ML capabilities within utilities;

(g) AI inference at edge computing nodes and IoCT;

(h) Optimization techniques for extracting the full potentials of IoCT;

(i) Development and deployment of IoCT learning environment;

(j) Performance design, modeling and evaluation of IoCT;

(k) Security and Privacy of IoCT Communication;

(l) IoCT Applications and Services;

(m) Communication technologies for IoCT

The scope for the application domains for the aforementioned topics is wide ranging and can include, but is not limited to: (a) Agriculture; (b)environemntal sustainability; (c) Health and wellbeing; and (d) Smart Cities.

The Keywords are:

Big data, Megadata, IoT, AI, Machine Learning, Data mining, Sensors