Explore the vast range of titles available.

The use of intelligent crop recommendation systems has become crucial in the era of smart agriculture to increase yield and enhance resource utilization. In this study, we compared different machine learning (ML), and deep learning (DL) models utilizing structured tabular data for crop recommendation. During our experimentation, both ML and DL models achieved decent performance. However, their architectures are not suited for setting up conversational systems. To overcome this limitation, we converted the structured tabular data to descriptive textual data and utilized it to fine-tune Large Language Models (LLMs), including BERT and GPT-2. In comprehensive experiments, we demonstrated that GPT-2 achieved a higher accuracy of 99.55% than the best-performing ML and DL models, while maintaining precision of 99.58% and recall of 99.55%. We also demonstrated that GPT-2 not only keeps up competitive accuracy but also offers natural language interaction capabilities. Due to this capability, it is a viable option to be used for real-time agricultural decision support systems.

With the proliferation of mobile crowdsensing (MCS) and crowdsourcing, new challenges are emerging every day. Although crowdsensing has become a popular sensing paradigm to aggregate sensor readings from a variety of sources, data inconsistency has arisen as a serious challenge. Truth discovery (TD) has been developed as an effective method for reducing data inconsistency and as a validity assessment for conflicting data from various sources. In addition, MCS applications and services are moving beyond a single individual participant to community groups and are influenced by group behavior. To address these challenges in this paper, we propose a novel Fog-assisted Group-based Truth Discovery Framework over MCS Data Streams, an efficient TD system for real-time applications. Specifically, we first initialized the weights for the weight update process in TD with the participants’ credibility level. Then, we developed a novel Two-layer Group-based Truth Discovery (TGTD) mechanism in which the first layer estimates the truth of the group’s members and the second layer estimates the aggregated truth for the groups. We have conducted extensive experiments over synthetic and real-world datasets to prove the effectiveness and efficiency of our framework. The results indicate that TGTD achieves superior truth discovery accuracy compared to current streaming truth discovery approaches, while maintaining a reasonable running time. The organization of the streaming process within the fog architecture simulation is identified as an area for further investigation and future work.

In this paper, a novel Multi-Objective Hypergraph Particle Swarm Optimization (MOHGPSO) algorithm for structural health monitoring (SHM) systems is considered. This algorithm autonomously identifies the most relevant sensor placements in a combined fitness function without artificial intervention. The approach utilizes six established Optimal Sensor Placement (OSP) methods to generate a Pareto front, which is systematically analyzed and archived through Grey Relational Analysis (GRA) and Fuzzy Decision Making (FDM). This comprehensive analysis demonstrates the proposed approach’s superior performance in determining sensor placements, showcasing its adaptability to structural changes, enhancement of durability, and effective management of the life cycle of structures. Overall, this paper makes a significant contribution to engineering by leveraging advancements in sensor and information technologies to ensure essential infrastructure safety through SHM systems.

Stock market forecasting is one of the most challenging problems in today’s financial markets. According to the efficient market hypothesis, it is almost impossible to predict the stock market with 100% accuracy. However, Machine Learning (ML) methods can improve stock market predictions to some extent. In this paper, a novel strategy is proposed to improve the prediction efficiency of ML models for financial markets. Nine ML models are used to predict the direction of the stock market. First, these models are trained and validated using the traditional methodology on a historic data captured over a 1-day time frame. Then, the models are trained using the proposed methodology. Following the traditional methodology, Logistic Regression achieved the highest accuracy of 85.51% followed by XG Boost and Random Forest. With the proposed strategy, the Random Forest model achieved the highest accuracy of 91.27% followed by XG Boost, ADA Boost and ANN. In the later part of the paper, it is shown that only classification report is not sufficient to validate the performance of ML model for stock market prediction. A simulation model of the financial market is used in order to evaluate the risk, maximum draw down and returns associate with each ML model. The overall results demonstrated that the proposed strategy not only improves the stock market returns but also reduces the risks associated with each ML model.

Unreliable networks often use excess bandwidth for data integration in smart cities. For this purpose, Messaging Queuing Telemetry Transport (MQTT) with a certain quality of service (QoS) is employed. Data integrity and data security are frequently compromised for reducing bandwidth usage while designing integrated applications. Thus, for a reliable and secure integrated Internet of Everything (IoE) service, a range of network parameters are conditioned to achieve the required quality of a deliverable service. In this work, a QoS-0-based MQTT is developed in such a manner that the transparent MQTT protocol uses Transmission Control Protocol (TCP)-based connectivity with various rules for the retransmission of contents if the requests are not entertained for a fixed duration. The work explores the ways to improve the overall content delivery probability. The parameters are examined over a transparent gateway-based TCP network after developing a mathematical model for the proposed retransmission-based mutant QoS-0. The probability model is then verified by an actual physical network where the repeated content delivery is explored at VM-based MQTT, local network-based broker and a remote server. The results show that the repeated transmission of contents from the sender improves the content delivery probability over the unreliable MQTT-based Internet of Things (IoT) for developing smart cities’ applications.

Fractional order algorithms demonstrate superior efficacy in signal processing while retaining the same level of implementation simplicity as traditional algorithms. The self-adjusting dual-stage fractional order least mean square algorithm, denoted as LFLMS, is developed to expedite convergence, improve precision, and incurring only a slight increase in computational complexity. The initial segment employs the least mean square (LMS), succeeded by the fractional LMS (FLMS) approach in the subsequent stage. The latter multiplies the LMS output, with a replica of the steering vector ( Ŕ ) of the intended signal. Mathematical convergence analysis and the mathematical derivation of the proposed approach are provided. Its weight adjustment integrates the conventional integer ordered gradient with a fractional-ordered. Its effectiveness is gauged through the minimization of mean square error (MSE), and thorough comparisons with alternative methods are conducted across various parameters in simulations. Simulation results underscore the superior performance of LFLMS. Notably, the convergence rate of LFLMS surpasses that of LMS by 59%, accompanied by a 49% improvement in MSE relative to LMS. So it is concluded that the LFLMS approach is a suitable choice for next generation wireless networks, including Internet of Things, 6G, radars and satellite communication.

Stock market forecasting is one of the most challenging problems in today's financial markets. According to the efficient market hypothesis, it is almost impossible to predict the stock market with 100% accuracy. However, Machine Learning (ML) methods can improve stock market predictions to some extent. In this paper, a novel strategy is proposed to improve the prediction efficiency of ML models for financial markets. Nine ML models are used to predict the direction of the stock market. First, these models are trained and validated using the traditional methodology on a historic data captured over a 1-day time frame. Then, the models are trained using the proposed methodology. Following the traditional methodology, Logistic Regression achieved the highest accuracy of 85.51% followed by XG Boost and Random Forest. With the proposed strategy, the Random Forest model achieved the highest accuracy of 91.27% followed by XG Boost, ADA Boost and ANN. In the later part of the paper, it is shown that only classification report is not sufficient to validate the performance of ML model for stock market prediction. A simulation model of the financial market is used in order to evaluate the risk, maximum draw down and returns associate with each ML model. The overall results demonstrated that the proposed strategy not only improves the stock market returns but also reduces the risks associated with each ML model.

ABSTRACT This paper introduces a novel ensemble Deep learning (DL)‐based Multi‐Label Retinal Disease Classification (MLRDC) system, known for its high accuracy and efficiency. Utilising a stacking ensemble approach, and integrating DenseNet201, EfficientNetB4, EfficientNetB3 and EfficientNetV2S models, exceptional performance in retinal disease classification is achieved. The proposed MLRDC model, leveraging DL as the meta‐model, outperforms individual base detectors, with DenseNet201 and EfficientNetV2S achieving an accuracy of 96.5%, precision of 98.6%, recall of 97.1%, and F1 score of 97.8%. Weighted multilabel classifiers in the ensemble exhibit an average accuracy of 90.6%, precision of 98.3%, recall of 91.2%, and F1 score of 94.6%, whereas unweighted models achieve an average accuracy of 90%, precision of 98.6%, recall of 93.1%, and F1 score of 95.7%. Employing Logistic Regression (LR) as the meta‐model, the proposed MLRDC system achieves an accuracy of 93.5%, precision of 98.2%, recall of 93.9%, and F1 score of 96%, with a minimal loss of 0.029. These results highlight the superiority of the proposed model over benchmark state‐of‐the‐art ensembles, emphasising its practical applicability in medical image classification.

ABSTRACT This work presents a sub‐THz planar antenna designed for THz communication and sensing, featuring advanced beam steering capabilities. The antenna uses a silicon dioxide (SiO2) substrate with a copper radiating zone incorporating V‐shaped slots, optimized for 0.3 THz. Electromagnetic simulations demonstrate a wide bandwidth of 29 GHz (0.286–0.315 THz) and a return loss of 37.59 dB at 0.3 THz. The beam steering system employs parasitic stubs as reflectors and directors, inspired by Yagi–Uda architecture, and utilizes graphene with adjustable chemical potentials for dynamic beam control. Simulation results show effective beam steering with gains of 6.79 and 6.75 dBi for left and right directions, respectively. A coaxial feed on a circular substrate allows 180° beam steering, enhancing radiation in three directions with gains of 6.47, 6.51, and 6.21 dBi for left, right, and top, respectively, and a front‐to‐back ratio of 14.1 dB. The integration of graphene for dynamic beam steering underscores the antenna's versatility and effectiveness in THz applications. This study introduces a sub‐THz planar antenna for THz communication and sensing with beam steering capabilities. Using a SiO2 substrate, copper radiating elements, and graphene slits, the design achieves a 29 GHz bandwidth at a 0.3 THz center frequency. Graphene‐enhanced Yagi–Uda‐inspired beam steering achieves gains of 6.79 and 6.75 dBi, ideal for versatile THz applications.