The proliferation of recommender systems (RS) research has attracted researchers to explore new tools and techniques to address several issues in the domain. In this context, text mining has received considerable acceptance for solving cold start and data sparsity problems in recommender systems. Additionally, text mining has been exploited to design modern recommender systems through sentiment analysis of user reviews, social network posts, web data, etc. These sentiment identifications help create user profiles, identify preferences, understand the context in which a review is provided, and extract features to assess item quality, facilitating better recommendations. This paper presents a comprehensive and systematic review of text mining-based recommender systems (TMRS). We analyze how text mining techniques have been integrated into different RS paradigms and propose a novel taxonomy that classifies TMRS based on underlying text mining approaches and recommendation approaches. In addition, we examine commonly used evaluation metrics and discuss how they are applied to assess TMRS performance. Key research challenges and open issues are identified, along with promising future research directions. This review provides a structured overview of the state of the art in TMRS and serves as a useful reference for researchers and practitioners seeking to design, evaluate, and advance next-generation recommender systems.
Isomorphism identification in epicyclic gear trains (EGTs) plays a crucial role for the structural synthesis and analysis of geared mechanisms since last two decades. Although many researchers have been suggested traditional approaches based on graph theory and matrix operations often face computational inefficiencies in handling complex gear structures. This paper introduced an automated approach for detecting isomorphism in polyhedral representations of EGTs using supervised machine learning. A curated dataset comprising various classes of conventional EGT graphs is used for training and classification through Graph Neural Networks (GNNs). To validate the classification results, the Total Distance Technique (TDT) is employed, wherein the TDT value corresponds to the sum of all elements in the adjacency matrix. Two EGTs are classified as isomorphic if both their GNN embeddings and TDT values align; otherwise, they are considered distinct. All results are fully satisfied with existing literature and the proposed approach is best applicable for optimising the design of mechanical transmission systems through machine learning-driven automation.
An experimentally demonstrated open-circuit voltage of a CdTe-based solar cell is only 904.8 mV, which is -235.2 mV lower than the Shockley-Queisser voltage limit. This voltage loss can be attributed to the factors such as radiative, nonradiative, and thermodynamic recombination losses. To circumvent the voltage loss issue, this study proposes a strategy of implementation a CdSe0.2Te0.8 passivation layer in conjunction with graded bandgap absorber layers. Initially, we examined a device with a configuration of V2O5/CdTe/ZnSe in absence of passivation layer. This device resulted in a larger Shockley-Read-Hall (SRH) recombination voltage loss of 281 mV and a total voltage loss of 716 mV. We modified this device configuration by utilizing an ultrathin layer (1 nm) of CdSe0.2Te0.8, i.e., V2O5/CdTe/CdSe0.2Te0.8/ZnSe. An ultrathin layer of CdSe0.2Te0.8 works as an effective passivation layer, substantially reducing the SRH recombination voltage loss to 46 mV from 281 mV. Interestingly, when a thicker layer of CdSe0.2Te0.8 is utilized, it not only acts as a passivation layer but also functions as an absorber layer, creating a bandgap gradient. However, improving the grain boundary between CdSe0.2Te0.8 and ZnSe is necessary to further cuts down to SRH recombination voltage loss below 46 mV. To overcome this issue, a thin window layer of CdS0.102Se0.336Te0.562 has been incorporated in between CdSe0.2Te0.8 and ZnSe, close to the front contact, which reduces the SRH recombination voltage loss to 20 mV. This SRH recombination voltage loss can be further minimized to zero from 20 mV when the back interface has been optimized. All simulation data have been justified by previous reported experimental finding to validate the proposed simulation models. Additionally, two-terminal and four-terminal perovskite/CdTe tandem configurations have also been proposed and simulated, yielding power conversion efficiency of 28.64 % and 29.80 %, respectively. These findings demonstrate the efficacy of passivation layer, double absorber layer, bandgap grading, window layer, and interface engineering in mitigating SRH recombination voltage loss, offering a roadmap for future perovskite/CdTe tandem solar cells.
The Optical Phase Retrieval (OPR) technique represents an innovative encoding approach that enforces constraints in both the Fourier and spatial domains to generate a phase-only representation of data. Discarding amplitude information inherently achieves data compression in the Fourier domain while preserving essential structural details. This paper presents an efficient and secure video encryption framework integrating OPR with multimodal chaotic-map modulation for enhanced robustness and confidentiality. In the proposed method, video sequences are decomposed into individual frames, which are encoded into phase-only distributions using an iterative OPR algorithm. These encoded frames are further encrypted through multimodal chaotic maps, including Logistic and Henon systems, ensuring high key sensitivity, strong diffusion, and nonlinear randomness. The integrated framework achieves efficient data compression, secure transmission, and robustness against perturbations. The numerical results demonstrate that the proposed scheme achieves a large key space and strong sensitivity, and offers efficient computation time complexity, also effectively showing reliable performance against video processing attacks such as compression, rotation, and blurring. Comprehensive experiments under diverse degradations — such as Gaussian noise, salt-and-pepper noise, and partial occlusion — demonstrate that the proposed system provides high-quality reconstruction during decryption. Statistical evaluations using metrics such as MSE, PSNR, SSIM, and entropy confirm excellent decryption fidelity and strong resistance to standard cryptographic attacks. Furthermore, the proposed system exhibits real-time processing capabilities, making it a suitable choice for next-generation optical video encryption and secure multimedia communication applications.
The shielded metal arc welding (SMAW) process is a widely used process. A common problem with the process is the loss of material in terms of stub loss, and frequently changing electrodes that consume much time, interrupting the process's continuity. To overcome this issue, an auxiliary flux feeding method- flux metal arc welding (FMAW)- is proposed for surfacing, joining and hard-facing applications enabling continuous welding without interruption and material loss and achieving a higher deposition rate at a lower deposition current. The current work uses FMAW process for cladding operation i.e. a dissimilar welding of stainless steel layer over carbon steel. This paper investigates the effects of proposed FMAW process parameters on metal deposition rate and percentage dilution for dissimilar metal welding. Weld beads have been laid down on a mild steel plate by adjusting process parameters, such as open-circuit voltage, current, welding speed, nozzle-to-plate distance, and flux flow rate, as per the central composite design. The responses have been compared with the established conventional gas metal arc-based additive manufacturing, also called wire additive manufacturing (WAAM). The interrelations between process parameters and responses have been established using multiple linear regression. The effects of individual parameters and their interactions on responses have been studied in the proposed process. Finally, single and multi-objective optimization was performed using the grey wolf optimization (GWO). The results reveal that a modified process can achieve a higher metal deposition rate with lower dilution. The FMAW process provides 47.47% more metal deposition rate and 35.63% lower dilution than the GMAW process.
