Images play a pivotal role in documenting real-life events. With the rapid evolution of digital technology, there has been a significant increase in both the creation and dissemination of photographs. The accessibility of picture editing software has simplified the process of altering images, thereby reducing the time, costs, and expertise needed to create and manage visually manipulated content. Unfortunately, digitally altered photographs have become a primary medium for disseminating misinformation, which affects individuals and society at large. Consequently, the need for effective methods to detect and identify forgeries is more pressing than ever. One prevalent form of picture fraud, image splicing, has been thoroughly examined. In this study, we present a Depth-Wise Convolutional Neural Network (DWCNN) model specifically designed to accurately detect spliced forged images. By converting input RGB images to the HSV color space, known for its ability to withstand color and lighting variations, our model achieves high accuracy in identifying manipulated images. Furthermore, our proposed model is lightweight, based on the MobileNet architecture with seven bottleneck blocks, making it suitable for a wide range of scenarios with constrained resources. To evaluate the model's performance, we tested it on the CASIA v1.0 and CASIA v2.0 datasets. Our model accurately identified forgeries with 99.23% accuracy on the CASIA v1.0 dataset and achieved a remarkable accuracy of 99.37% on the CASIA v2.0 dataset.

The chemical response and the collective buoyancy effects of thermal and mass diffusion in magnetohydrodynamic (MHD) convection are analyzed on an infinite vertical surface with porous material that emits gas as it rises. This study investigates how the governing equations perform under a broad range of stringent conditions, incorporating subjective considerations such as determining which factor dominates—suction velocity or Hall current strength. Ancillary currents were also considered to provide a comprehensive analysis. The governing equations for liquid flow were solved using the perturbation technique, yielding results in terms of temperature, concentration, and velocity fields. Dimensionless profiles of temperature, velocity, and concentration are graphically presented for various parameter values. It is observed that an increase in the Dufour number leads to higher primary and secondary velocities, as well as increased temperature. Conversely, primary and secondary velocities decrease with an increase in the chemical reaction number and magnetic field strength.

Diesel fuel is composed of various molecules of hydrocarbons, so the properties vary from country to country. The sulfur content in Iraqi Diesel ranges from 1% to 2.5%. This abundance of sulfur contributes to a high level of emissions, including sulfur oxides, nitrogen oxides, volatile organic compounds, particulates, and carbon monoxide. To reduce emissions and improve engine performance, a diesel water mixture has been proposed as a fuel for diesel engines. This study examined the performance of a diesel engine under different operating conditions when Diesel was mixed with 10%, 20%, and 30% volume proportions of water, named W10, W20, and W30, respectively. The use of W10 and W20 caused a brake-specific fuel consumption reduction of 2.32% and 4.89%, respectively, compared to conventional Diesel, while using W30 caused an increase in brake-specific fuel consumption of about 5.75%. The brake thermal efficiency improved by 3.6% and 4.63% when using W10 and W20, respectively. While its value decreased when working with W30 by about 2.48% compared to Diesel. Working with W10 and W20 reduced engine emissions of carbon monoxide by an average of 9% and 27%, hydrocarbons by 7.8% and 20%, nitrogen oxides by 8.9% and 20.8%, and particulate matter by 4.92% and 13.1%. Operating with W10 and W20 reduced both particulate matter and nitrogen oxide emissions. The results reveal that water mixing with Iraqi Diesel is an effective means of reducing diesel engine emissions.

Heating, Ventilation and Air-Conditioning (HVAC) systems are responsible of 50-60% of energy demand of the building sector. The scientific literature highlights that HVAC units are frequently operated under faulty conditions that can significantly affect their performance. In this paper, the performance of a typical single-duct dual-fan constant air volume Air-Handling Unit (AHU) is investigated through a number of experiments performed during Italian cooling and heating seasons under both fault free and faulty scenarios. The AHU operation is analysed while artificially introducing seven typical faults: return air damper kept always closed; fresh air damper kept always closed; fresh air damper kept always open; exhaust air damper kept always closed; supply air filter clogged at 50%; fresh air filter clogged at 50%; return air filter clogged at 50%. The faulty and fault free tests are compared to assess the environmental and economic performance impacts. The experimental data highlighted that the most adverse fault is that one corresponding to the exhaust air dumper kept always closed; in particular, it increases both the daily global equivalent CO₂ emissions and the daily operating costs up to 110% in comparison with the fault free conditions.

Epilepsy seizures are complex neurological phenomena marked by recurrent and unpredictable seizures that can greatly affect an individual’s quality of life. It affects millions of people worldwide. The exact and timely detection of epileptic seizures is crucial in the management and treatment of epilepsy. Many methods have been put forth recently for the diagnosis of epileptic seizures using magnetic resonance imaging (MRI) and electroencephalography (EEG). This work focuses on using deep learning and machine learning techniques, such as Support Vector Machines (SVMs), Recurrent Neural Networks (RNNs), and Convolutional Neural Networks (CNNs), to automatically identify epileptic seizures. These techniques have shown promising results in a variety of fields, including time series data processing and medical image analysis. In this work, we present a unique method for detecting epileptic seizures using electroencephalogram (EEG) data by comparing the outcomes of three deep learning architectures: SVM, CNN, and RNN-LSTM (Long-short term memory). The experimental results demonstrate that the SVM, CNN and RNN-LSTM models exhibit promising performance in detecting epileptic seizures from EEG data.

The misleading reviews posted on shopping websites and other media platforms sway the opinions and decisions of different customers. On the other hand, dishonest reviewers will make an effort to mimic the writing style of legitimate reviews. There is no guarantee that these text-feature-based approaches will work anytime soon. In addition, the likelihood of an imbalanced category distribution in practice limits detection performance. This paper proposes a fraudulent review detection system that uses ensemble feature selection and multidimensional feature creation to overcome these limitations. Our idea builds three-dimensional characteristics, which include text, reviewer behaviour, and misleading scores. Furthermore, a data resampling approach combines Random Sampling and oversampling techniques to mitigate the effects of an imbalanced distribution of categories. In addition, we combine the outcomes of several feature selection methods that focus on information gain, XGBoost feature importance, and the Chi-square test. On various text datasets, the proposed technique demonstrates exemplary performance in fraudulent review identification according to the experimental findings using feature selection methods, resampling methods, classification, etc. Our technique outperforms existing sophisticated methods when faced with low-quality text or an imbalanced dataset.

The analyzing of friction stir welding is applied to AA6061 with AA5083 using design of experiment (DOE) in Minitab to get optimization of the tensile strength. The analyzing was achieved by varying the main parameters as rotational speed by the values 700, 1050, and 1400 R.P.M, linear velocity of 40, 60, and 80 mm/min and pin depth of 3.5, 3.6, and 3.7 mm less than thickness material weld. A total of 11 runs were included corresponding to the designated experimental design. Analysis of variance (ANOVA) software was used to procedure for full factorial design with 1 replicate and 3 center points analysis. The results clarified that the rotational speed parameter is more significant to be controlled rather than the linear velocity and pin depth of tool. Decreasing rotational speed and increasing linear velocity and pin depth led to higher tensile strength. The profiles welded at 700 RPM, 80 mm/min and 3.7 mm had achieved the optimum case to get the value of maximum tensile strength. It is concluded that the rotational speed was the key parameter that manipulate the tensile strength in friction stir welding applied to AA6061 with AA508.

This work analyzes and studies the characteristics of three enclosures on a university Campus, which present similar challenges in terms of noise pollution. To carry out an empirical and objective assessment on their acoustic performance, current regulations, and standards, are being used. Theoretical calculations are considered to calculate reverberation time parameters. In order to calculate reverberation time by using the Sabine formula, it is necessary to measure the classrooms, together with the specification of the surface occupied by each of the materials that make up the walls in the rooms under study, resulting in a T60 of between 3.6 s to 6.2 s for classrooms 11 and 12, and between 4.1 s to 7.1 s for classroom 15. To obtain the parameters that define the acoustic capacities of reverberation of the rooms, the guidelines for both measurement and calculation conditions specified in the regulations are followed. Graphical representation and mathematical calculation software are used to achieve the desired results, obtaining a T60 of between 1.8 s to 2.2 s for classroom 11, 2.0 s to 3.0 s for classroom 12, and 1.7 s to 2.7 s for classroom 15. Once the acoustic conditions of the reverberation of the room are defined, it is concluded that none of the rooms has the proper characteristics to carry out the best teaching activities in them, since they exceed the recommended 0.7 s of reverberation time, since they hinder the understanding of speech and the clarity of the word. As a conclusion, the study has served as an analysis of a challenging task in the Miguelete Campus of the Universidad Nacional de San Martin, always based on parameters commonly used in the world of acoustic pollution, both scientifically and legislatively.

In order to minimize trial-and-error costs and avoid any faults throughout the production process, sheet metal forming is commonly used in the automotive industry to fabricate body pieces. This study examines how several punch models affect thinning and wrinkling during the deep drawing of aluminum 1050, a 2.50 mm thick material, under 150 KN of pressure. The Forming Limit Diagram (FLD) was used in the simulations to study material deformation and determine safe and essential places on the blank. According to the findings, die pressure-induced material stretching caused the material to significantly rise in major-minor strain, major-minor stress, thinning, and wrinkling by the fifth step. Additionally, for punch 1 material, the safe area grew from 9.20% to 49.36%; punch 2 increased from 9.08% to 46.85%. The non-linear FLD graph analysis verified that both materials stayed in the safe zone the entire time. These results demonstrate how well deep drawing simulations work to improve punch design and aluminum component performance in the automobile industry.

Sarcasm detection is challenging in sentiment analysis, especially for morphologically complex languages like Telugu. Sarcastic statements often use positive words to convey negative sentiments, complicating automated interpretation. Existing sarcasm detection systems predominantly cater to English, leaving a gap for low-resource languages such as Hindi, Telugu, Tamil, Arabic, and others. This study fills this gap by creating and annotating a Telugu conversational dataset, which includes both standard and sarcastic responses. We employed two deep learning models—Self Attention-based Recurrent Neural Network (SA-RNN) and Gated Recurrent Unit (GRU)—to analyze this dataset. Results showed that the SA-RNN model outperformed the GRU, achieving 96% accuracy compared to 94%. The models utilized GloVe word embeddings and specific linguistic features, such as interjections and punctuation marks, to improve sarcasm detection. This research advances the field of sarcasm detection for low-resource languages, particularly Telugu.

Hand gesture recognition is a technology that enables computers to interpret and understand hand movements and gestures made by users. It has various applications across various domains, including human-computer interaction, gaming, virtual reality, sign language interpretation, and robotics. Hand recognition faces challenges such as lighting conditions, occlusions, and variations in hand shape and size. Creating reliable and precise recognition systems frequently necessitates tackling these issues. Neural Architecture Search (NAS) is a technique employed in deep learning and artificial intelligence to automate the creation and optimization of neural network topologies. The objective of NAS is to identify neural network designs that are optimally aligned with certain objectives, including image classification, natural language processing, or reinforcement learning while reducing the necessity for manual design and adjustment. YOLONAS model's integration of YOLO's speed and efficiency with NAS-driven optimization results in improved accuracy and performance in gesture recognition tasks, making it a compelling choice for real-time applications requiring accurate and efficient gesture analysis. In this research, we implement YOLO with NAS technology and training with the Oxford Hand Dataset. Performance metrics are employed for monitoring and quantifying important data, such as the number of Giga Floating Point Operations Per Second (GFLOPS), the mean average precision (mAP), and the time taken for detection. The results of our study indicate that the utilization of YOLONAS with a training time of 100 epochs produces a more reliable output when compared to other approaches.

Traffic congestion is a significant challenge in Lagos State, Nigeria. Existing methods rely on limited data sources and simplistic models that fail to capture the complexities of traffic dynamics in a congested urban environment. This study focused on traffic intensity detection in Lagos State using a Bayesian estimation model. Data was obtained from the Kaggle website which involved observing the number of vehicles intersecting junctions at various times of the day for a week. The model captured both spatial and temporal variations, providing real-time estimations of traffic congestion levels across different road segments. Comparative analysis with existing traffic estimation methods showed superior performance in terms of accuracy and reliability. The Gaussian Naive Bayes model achieved a high accuracy of 96% and balanced f1-score of 96%, precision of 0.96, and recall of approximately 0.96. On the other hand, the multinomial Naive Bayes model achieved an accuracy of 69% with a lower f1-score of 69%, precision of 0.67, and recall of 0.69. The model's capacity to provide accurate real-time site traffic facts can significantly contribute to effective traffic control and concrete making plans initiatives.

The advancement of technology within the manufacturing sector continually accelerates, frequently resulting in heightened consumer demand driven by enhancements in product quality, including consistency and quality control. This study endeavors to enhance the production process by leveraging reverse engineering alongside Indonesian anthropometry methods. The findings demonstrate that the resultant products can effectively meet customer requirements, exhibiting durability and resilience against wear and tear. By integrating reverse engineering techniques and Indonesian anthropometry methods into the production process, manufacturers can achieve greater precision and tailor products to better suit consumer needs. This approach not only enhances product quality but also contributes to increased customer satisfaction and loyalty. Furthermore, it underscores the importance of incorporating innovative methodologies to keep pace with the evolving demands of the market and to ensure continued success within the manufacturing industry. As technology continues to advance, embracing methodologies like reverse engineering and anthropometry will be crucial for manufacturers seeking to remain competitive and deliver products that surpass consumer expectations in terms of durability, performance, and overall quality.

A new technique for refining grains and creating surface composite materials was created using friction stir processing (FSP), which shows promise for improving the properties of metals. This technology can effectively overcome the limitations of methods that depend on melting. The work employed stir casting to produce in-situ composites, including an Al₃Ni reinforcing phase, Al alloy A356, and 15% weight percent pure Ni powder. The development of several phases of Al₃Ni, AlNi, and Al₃Ni₂, which are dispersed throughout the matrix of A356 alloy, was verified by XRD analytical examination. The objective was to determine how FSP impacted the stir-cast A356/Al3Ni in-situ composite's mechanical properties and microstructure. Porosity was successfully reduced, the α-Al dendrites were refined, the main Si phase and Al₃Ni were broken and fragmented, and the grain structure was improved by the FSP method. There was a consistent and equal distribution of in-situ Al₃Ni in the stir zone (SZ). There were no hazardous phases present when the particle and matrix came into contact. The application of FSP improved the tensile strength of the A356 alloy by 38.35% and the in-situ composite A356/Al₃Ni by 69.17%. It was found that the improvement in hardness was 14.47% for A356 alloy after FSP and 13.81% for in-situ composite A356/Al₃Ni compared to in-situ composite without FSP.