Enabling Cognitive Smart Cities Using Big Data and Machine Learning: Approaches and Challenges

By M. Mohammadi, Ala Al-Fuqaha. 2018

306 citations recorded by the source corpus.

The development of smart cities and their fast-paced deployment is resulting in the generation of large quantities of data at unprecedented rates. Unfortunately, most of the generated data is wasted without extracting potentially useful information and knowledge because of the lack of established mechanisms and standards that benefit from the availability of such data. Moreover, the highly dynamic nature of smart cities calls for a new generation of machine learning approaches that are flexible and adaptable to cope with the dynamicity of data to perform analytics and learn from real-time data. In this article, we shed light on the challenge of underutilizing the big data generated by smart cities from a machine learning perspective. In particular, we present the phenomenon of wasting unlabeled data. We argue that semi-supervision is a must for smart cities to address this challenge. We also propose a three-level learning framework for smart cities that matches the hierarchical nature of big data generated by smart cities with a goal of providing different levels of knowledge abstraction. The proposed framework is scalable to meet the needs of smart city services. Fundamentally, the framework benefits from semi-supervised deep reinforcement learning where a small amount of data that has users' feedback serves as labeled data, while a larger amount without such users' feedback serves as unlabeled data. The framework utilizes a mix of labeled and unlabeled data to converge toward better control policies instead of wasting the unlabeled data. This article also explores how deep reinforcement learning and its shift toward semi-supervision can handle the cognitive side of smart city services and improve their performance by providing several use cases spanning the different domains of smart cities. We also highlight several challenges as well as promising future research directions for incorporating machine learning and high-level intelligence into smart city services.

Subfields: Computer Networks and Communications, Media Technology, Transportation

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Media Technology

In Transportation

Citations in the corpus

Deep Learning for IoT Big Data and Streaming Analytics: A Survey (2017, background): “A combination of advanced machine learning algorithms designed for semi-supervised settings fits well for smart cities systems, where a small training dataset can be used while the learning agent improves its accuracy using a large amount of unlabeled data [5].”
Quantum Machine Learning for 6G Communication Networks: State-of-the-Art and Vision for the Future (2019, background): “…sensors installed vehicles (private/public), road trafﬁc monitoring systems (visual and sensor-based), UAVs, ﬁre/earthquake/other sensors (disasters alerting and monitoring), user smartphones, etc. Leveraging intelligence to the IoT devices originates the new frontier of ‘‘IoT meets AI’’ [27].”
Federated Learning in Smart City Sensing: Challenges and Opportunities (2020, unknown): “The data is often processed with machine learning, deep learning and statistical methods to generate a trend or conclusion [26].”
Federated Learning for Internet of Things: A Comprehensive Survey (2021, background): “To realize smart cities, AI/ML techniques have been widely adopted to provide intelligence properties thanks to their ability to handle real-time big data generated from sensors, devices, and human activities [152].”
Machine Learning for Resource Management in Cellular and IoT Networks: Potentials, Current Solutions, and Open Challenges (2019, background): “learn and can adapt dynamically to the variation in the network environment [55]–[57].”
Toward Massive Machine Type Communications in Ultra-Dense Cellular IoT Networks: Current Issues and Machine Learning-Assisted Solutions (2018, methodology): “…big data analytics [19,39,40] QoS provisioning [41] Intelligence in 5G networks [16,20] Machine Learning (ML) Learning in IoT/sensor networks [26,42,43] Reinforcement learning [44,45] Ultra-Dense Networks (UDNs) [46–48] and the distinct features of mMTC trafﬁc as compared to the conventional…”
Blockchain and Deep Reinforcement Learning Empowered Intelligent 5G Beyond (2019, background): “AI exhibits great potential on in-depth feature discovery and events forecasting utilizing such data [12].”
Applications of Blockchains in the Internet of Things: A Comprehensive Survey (2019, background): “Scalable infrastructures are necessary to efficiently handle this massive growing volume of data [73].”
Artificial Intelligence for the Metaverse: A Survey (2022, background): “Currently, the actual impact and benefit of these technologies to the smart cities is limited; however, the metaverse can be an accelerant to spread the presentation of smart services in the daily life of citizens [157].”
Smart Transportation: An Overview of Technologies and Applications (2023, background): “Furthermore, historical data may contain many unclear values, missing values, outliers, and nonsensical data [81].”
Fusing Blockchain and AI With Metaverse: A Survey (2022, methodology): “Big Data and Machine Learning [53] Semi-supervised deep reinforcement”