The Internet of Things (IoT) is one of the key components of the ICT infrastructure of smart cities due to its great potential for intelligent management of infrastructures and facilities and the enhanced delivery of services in support of sustainable cities. Smart cities typically rely on IoT, where a wide variety of devices communicate with each other and collaborate across heterogeneous and distributed computing environments to provide information and services to urban entities and urbanites. However, leveraging the IoT within software applications raises tremendous challenges, such as data acquisition, device heterogeneity, service management, security and privacy, interoperability, scalability, flexibility, data processing, and visualization. Middleware for IoT has been recognized as the system that can provide the necessary infrastructure of services and has become increasingly important for IoT over the last few years. This study aims to review and synthesize the relevant literature to identify and discuss the core challenges of existing IoT middleware. Furthermore, it augments the information landscape of IoT middleware with big data applications to achieve the required level of services supporting sustainable cities. In doing so, it proposes a novel IoT middleware for smart city applications, namely Generic Middleware for Smart City Applications (GMSCA), which brings together many studies to further capture and invigorate the application demand for sustainable solutions which IoT and big data can offer. The proposed middleware is implemented, and its feasibility is assessed by developing three applications addressing various scenarios. Finally, the GMSCA is tested by conducting load balance and performance tests. The results prove the excellent functioning and usability of the GMSCA.

Searching for a free parking space can lead to traffic congestion, increasing fuel consumption, and greenhouse gas pollution in urban areas. With an efficient parking infrastructure, the cities can reduce carbon emissions caused by additional fuel combustion, waiting time, and traffic congestion while looking for a free parking slot. A potential solution to mitigating parking search is the provision of parking-related data and prediction. Previously many external data sources have been considered in prediction models; however, the underlying impact of contextual data points and prediction has not received due attention. In this work, we integrated parking occupancy, pedestrian, weather, and traffic data to analyze the impact of external factors on on-street parking prediction. A comparative analysis of well-known Machine (ML) Learning and Deep Learning (DL) techniques, including Multilayer Perceptron (MLP), Random Forest (RF), Decision Trees (DT), K-Nearest Neighbors (KNN), Gradient Boosting (GA), Adaptive Boosting (AB), and linear SVC for the prediction of OnStreet parking space availability has been conducted. The results show that RF outperformed other techniques evaluated with an average accuracy of 81% and an AUC of 0.18. The comparative analysis shows that less complex algorithms like RF, DT, and KNN outperform complex algorithms like MLP in terms of prediction accuracy. All four data sources have positively impacted the prediction, and the proposed solution can determine the best possible parking slot based on weather conditions, traffic flow, and pedestrian volume. The experiments on live prediction showed an ingest rate of 0.1 and throughput of 0.3 events per second, demonstrating a fast and reliable prediction approach for available slots within a 5–10 min time frame. The study is scalable for larger time frames and faster predictions that can be implemented for IoT-based big data-driven environments for on-street and off-street parking.

The integrity and security of data must be protected in the framework of the Internet of Things (IoT). This article addresses the difficulties presented by possible cyber threats to IoT devices by introducing a unique feature selection technique called Pearson correlation matrix with random forest (PCM‐RF). IoT device security is greatly influenced by machine learning techniques, which mostly depend on the caliber of characteristics taken from IoT datasets. By merging the advantages of RF with PCM, PCM‐RF maximizes feature selection and fine‐tunes features to improve the efficacy of current machine learning techniques and bolster classification algorithms' detecting powers. The study focuses on addressing limitations in existing ways of training and testing classification algorithms, which frequently lack strategies for optimizing and fine‐tuning features. Thirty‐four different forms of network assaults are included in the IoTCIC2023 dataset, used to assess PCM‐RF. Results demonstrate PCM‐RF's effectiveness, with XGBoost achieving an astounding accuracy of 99.39% and an 86% detection rate. Comparative studies highlight PCM‐RF's superiority in detection and classification results, offering insights into IoT security and emphasizing its potential to improve the overall device robustness in the IoT ecosystem.

With the advent of smart computing, Internet of Things (IoT) and sensor technology, it is now possible to determine the wellness of the elderly living alone in a home equipped with miniaturized sensors. The technology has the potential to seamlessly

Aspect Based Sentiment Analysis techniques have been applied in several application domains. From the last two decades, these techniques have been developed mostly for product and service application domains. However, very few aspect-based sentiment techniques have been proposed for the movie application domain. Moreover, these techniques only mine specific aspects (Script, Director, and Actor) of a movie application domain, nevertheless, the movie application domain is more complex than the product and service application domain. Since, it contains NER (Named Entity Recognition) problem and it cannot be ignored, since there is an opinion often associated with it. Consequently, in this paper MAIM (Movie Aspect Identification Model) is proposed that can extract not only movie specific aspects, also identifies NEs (Named Entities) such as Person Name and Movie Title. The three main contributions are 1) the identification of infrequent aspects, 2) the identification of NE (named entity) in movie application domain, 3) identifying N-gram opinion words as an entity. MAIM incorporates the BiLSTM-CRF hybrid technique and is implemented on the movie application domain having precision 89.9%, recall 88.9% and f1-measure 89.4%. The experimental results show that MAIM performs better than baseline models CRF and LSTM-CRF.