This paper examines the relationships among artificial intelligence, leadership, sense of belonging, and psychological need fulfilment among employees working in hybrid mode. It draws on self-determination theory to examine psychological need fulfilment and sense of belonging in relation to leadership perceptions, with artificial intelligence examined as a mediating variable. The study is limited to IT employees working in hybrid mode in Bengaluru, India. Using purposive sampling, 500 questionnaires were distributed, 410 responses were received, and 400 completed and valid responses were used for analysis. The proposed research model was analyzed using SPSS and AMOS software programs with structural equation modelling and mediation testing based on 5000 bootstrap resamples and bias-corrected 95% bootstrap confidence intervals. The findings show that sense of belonging (β = .381, p < .001), psychological need fulfilment (β = .318, p < .001), and artificial intelligence (β = .113, p = .013) are significantly and positively related to leadership. Psychological need fulfilment is positively associated with AI (β = .551, p < .001), whereas sense of belonging is significantly negatively associated with AI (β = −.097, p = .036). Artificial intelligence significantly mediates the relationship between psychological need fulfilment and leadership through a positive indirect effect (β = .062, 95% BC CI [.011, .120], p = .018). Artificial intelligence also mediates the relationship between sense of belonging and leadership (β = −.011, 95% BC CI [−.031, −.001], p = .034). The study concludes that artificial intelligence mediates the relationships of psychological need fulfilment and sense of belonging with leadership among employees working in hybrid mode. It highlights the importance of maintaining meaningful human interaction alongside the use of artificial intelligence. The findings also contribute to applying self-determination theory to understanding leadership perceptions among employees working in hybrid mode.
This paper presents a hybrid adaptive wavelet thresholding method for denoising underwater multipath Binary Phase Shift Keying (BPSK) communication signals corrupted by real ambient noise. A raised-cosine pulse-shaped BPSK signal is transmitted through simulated underwater acoustic channels with varying delay spreads and channel conditions to model realistic underwater propagation. Real ambient underwater noise is added to the transmitted signal at input SNRs of -15, −10, −5, 0, and 5 dB to evaluate the proposed method's robustness under different noise levels. The noisy signal is decomposed using the Continuous Wavelet Transform (CWT) with the Analytic Morlet wavelet. The proposed hybrid method combines adaptive covariance-based weighting with soft thresholding to preserve the transient burst features. Delay-dependent correlation parameters are also incorporated to adapt the thresholding to the severity of multipath propagation. The denoised signal is reconstructed using the inverse Continuous Wavelet Transform, and the performance is evaluated using output Signal-to-Noise Ratio (SNR), Root Mean Square Error (RMSE), Correlation Coefficient (CC), Signal Preservation Index (SPI), computational runtime, and statistical significance analysis. The proposed method is compared with Universal, SURE (Stein's Unbiased Risk Estimate), Minimax, BayesShrink, Empirical Mode Decomposition (EMD)-based, and Variational Mode Decomposition (VMD)-based denoising methods. Monte Carlo simulations show that the proposed hybrid method provides improved denoising performance and better preservation of the transmitted BPSK signal under varying underwater channel conditions.
Federated learning (FL), which allows distributed training across several industrial machines without exchanging raw vibration data, has emerged as a promising paradigm for privacy-preserving bearing fault diagnosis in Industrial Internet of Things (IIoT) scenarios. Nevertheless, current FL approaches for fault diagnosis face two significant limitations. First, uniform client participation in each communication round results in needless and prohibitive communication overhead. Second, static or random client selection strategies are indifferent to the informativeness of individual client updates under highly non-IID data distributions. This paper proposes a novel Multi-Head Deep Q-Network (Multi-Head DQN)-based client selection framework for communication-efficient federated learning applied to bearing fault diagnosis to overcome these limitations. In the proposed framework, the Multi-Head DQN agent jointly determines the optimal number of clients, k, to select in each round from a predefined range, and the identities of the most informative clients. To accurately replicate heterogeneous IIoT deployments in the real world, a stringent non-IID partitioning approach is implemented across 12 federated clients, each of which is allocated precisely two of four failure classes (Normal, Inner Race, Outer Race, and Ball). The Flower federated learning framework is used to compare six FL configurations: FedAvg, FedProx, FedAvg+FedRandom, FedProx+FedRandom, FedAvg+FedDQN, and FedProx+FedDQN. All experiments were conducted on the CWRU bearing fault dataset. Experimental results show that the proposed FedProx+FedDQN approach reduces cumulative communication costs by 19.15% compared with the all-client FedAvg baseline while achieving a global fault classification accuracy of 94.05%, which is comparable to the full-participation baselines (FedAvg and FedProx). The DQN-based selection prioritizes informative client updates, leading to faster convergence and lower communication overhead. These findings confirm that intelligent client selection is an effective strategy for improving communication efficiency in federated learning for industrial fault diagnosis applications.
In an increasingly digital era, technological advancements have transformed the way people shop and interact with e-commerce platforms. E-commerce offers a wide range of products and services online, but consumers often feel overwhelmed by the sheer number of choices. The importance of efficient and relevant services for consumers is a significant consideration in developing e-commerce systems. One effective solution is implementing a recommendation system that helps consumers find products that match their needs and preferences. This study develops an e-commerce recommendation system using neural collaborative filtering and content-based filtering methods, integrated with an interactive dashboard to improve user experience. With a learning rate of 0.01 and a batch size of 32, the developed neural collaborative filtering model shows a low testing error of 0.0705 and precision, recall, and F-measure of 0.9583, 0.8251, and 0.8867, respectively. The content-based filtering method uses the Cosine Similarity technique, achieving an Average Precision of 0.804 and an accuracy of 80%. The interactive dashboard was successfully developed to support system evaluation and user engagement. This study provides benefits for consumers who can find products according to their needs more quickly and for e-commerce business owners who can reach consumers more efficiently.
Alzheimer’s disease (AD) is one of the most prevalent forms of dementia, typically initiating with mild memory loss and progressively leading to severe cognitive decline. Early diagnosis is critical for effective intervention and improving patient outcomes. However, existing diagnostic approaches face challenges such as long prediction times, limited feature extraction capability, and the requirement for extensive training data, particularly in Machine Learning (ML)-based systems. To address these issues, this study proposes a novel computer-aided diagnostic framework called Adaptive Spatial Attention-based Twin Neural Network (ASTNet) for early AD detection using MRI data. Unlike traditional methods, ASTNet integrates several components: a biased fuzzy clustering-based segmentation technique that enhances tissue segmentation by emphasizing brain-relevant regions while suppressing skull and irrelevant tissues; Possibilistic C-Means (PCM) to improve segmentation robustness; a Convolutional Bidirectional LSTM (CBi-LSTM) network for context-aware feature extraction that captures temporal dependencies and reduces high-dimensional noise; and an adaptive spatial attention-based Twin Neural Network (TNN) for disease classification that improves spatial feature discrimination across cognitively normal (CN), mild cognitive impairment (MCI), and AD categories. The proposed approach was evaluated on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset and achieved a classification accuracy of 0.9909, outperforming the compared Deep Learning (DL) models. Furthermore, the model demonstrates computational efficiency under the evaluated experimental setting and shows potential for computer-aided clinical decision support.
