This study delves into the evaluation of daily profits within the network of Sephora cosmetics branches, employing distributed systems, the K-means algorithm, and WEKA visualization techniques. The objective is to analyze the profitability patterns across Sephora branches and uncover underlying clusters to optimize strategic decision-making. The research methods encompassed the utilization of distributed systems to collect and integrate daily profit data from multiple Sephora branches. The K-means clustering algorithm was applied to segment the dataset, discerning distinct clusters based on profit margins and performance metrics. The WEKA visualization tool was instrumental in depicting these clusters in a comprehensive and interpretable manner, offering insights into profit distributions and potential areas for enhancement. This research revealed intricate profitability structures within the Sephora branch network. The application of the K-means algorithm, distinct clusters of branches with varying profitability profiles were identified. WEKA visualization facilitated the graphical representation of these clusters, elucidating the relationships between different branches based on daily profits. Such insights can aid decision-makers in strategizing resource allocation, marketing initiatives, and operational improvements tailored to each cluster's profitability dynamics. The amalgamation of distributed systems, K-means algorithm, and WEKA visualization offered a robust framework for assessing and optimizing the financial performance of  Sephora branches. This research contributes valuable insights into the evaluation of daily profits across  Sephora cosmetics branches, showcasing the potential of distributed systems and advanced analytical tools in understanding profitability patterns. The findings provide a foundation for strategic interventions aimed at enhancing profitability and operational efficiencies within the Sephora branch network.

A new robust blind image watermarking method based on lifting wavelet transform (LWT), and Schur decomposition in combination with discrete cosine transform (DCT), and entropy analysis for copyright protection of multimedia information is introduced in this paper. At first, the 2-level LWT is applied to the original image to improve the invisibility of the watermarking method and then the high-frequency sub-band of 2-level LWT is decomposed by DCT. Then, the DCT coefficients are divided into 4×4 non-overlapping blocks. After that, Schur decomposition applies to each selected block, while the first row, the first column element of the upper triangular matrix is used to embed the watermark. For evaluation, the invisibility and robustness of the proposed watermarking method, the peak signal to noise ratio (PSNR), and normalized cross-correlation (NC) are used to measure the quality and the ability of the proposed watermarking method to be robust against signal processing operations and geometric attacks. Experimental results have demonstrated that the proposed method is achieved a very good tradeoff between invisibility and robustness. The comparisons with other scheme have shown that the proposed digital watermarking scheme has a superior performance in terms of invisibility than other. 

Abstract:

In this research study, we investigate the effectiveness of a proposed automated framework that leverages AI algorithms to improve the accuracy, scalability, and efficiency of data analysis. We compare the performance of the proposed framework to traditional data analysis methods in terms of accuracy, time and effort required for data analysis, and scalability. We conducted a questionnaire-based survey to collect data on the application of the proposed framework and traditional data analysis methods in real-world scenarios. We used statistical analyses, including t-tests and regression analyses, to analyze the data and draw conclusions. Our findings indicate that the proposed automated framework outperforms traditional data analysis methods in terms of accuracy, reduces the time and effort required for data analysis, and has better scalability for processing large datasets in real-time. Our results suggest that the proposed framework offers significant benefits for organizations seeking to improve their data analysis processes. However, we also note the importance of considering the costs and benefits of implementing the framework, as well as any ethical or legal implications. Further research is recommended to confirm the findings and to investigate the effectiveness of the proposed framework in different contexts and with different datasets.

Background and aims. Online learning has become increasingly popular in recent years due to its convenience and flexibility. However, it can be challenging for learners to stay engaged and motivated when studying online. One solution to this problem is the use of chatbots, which are computer programs that use Artificial Intelligence (AI) and Natural Language Processing (NLP) to simulate conversation with humans. ChatGPT, a language model that can generate human-like conversation in real time developed by OpenAI, has the potential to revolutionize online learning. This is by providing personalized and interactive learning experiences. This paper provides an overview of the potential impact of ChatGPT on online learning, exploring the benefits of using ChatGPT in enhancing the online learning process as well as identifying the limitations and challenges of this technology. Methods. Two experiments were conducted to evaluate ChatGPT's effectiveness in the domain of learning. Experiment 1: presents the authors’ experience with ChatGPT as Arabic speakers. Experiment 2: reviews ChatGPT’s performance in writing academic papers and retrieving references. Results. Overall, both experiments demonstrated the limitations of ChatGPT in generating reliable academic text that is sufficient for a scientific paper both in Arabic and English. Conclusion. ChatGPT is a tool that can guide learners throughout their learning process. But this tool exhibits limitations that learners need to dedicate more efforts to succeed

The network connectivity is often lost during some situations such as in war zones, battle fields or in regard to natural disasters like floods or earthquakes. Therefore, the connection becomes difficult for rescuing teams to locate people. At the same time, mobile nodes are difficult to be tracked/located. Therefore, flying ad-hoc networks (FANETs) is the solution which provides a way to tackle this situation without the need for any central infrastructure with the use of flying objects (nodes). FANTEs are frequently formed of tiny unmanned aerial vehicles (UAVs) which are flexible, affordable, and rapid to deploy. UAV as a router and without complex hardware deployment, provides an effective real-time communication solution for multiple UAV systems. FANETs can also be used to provide connectivity in such areas in which flying nodes are utilized. In this paper, we discuss the importance of the FANETs and present a routing protocol comparison that depends on the link budget to identify the optimized packet route for the data to be transmitted. The simulation design results show that a throughput enhancement of 0.62Gbps is achieved in comparison to the distance-based routing protocol. The results also confirm the efficiency of the optimized link state routing (OSLR) protocol over the other. The performance of FANETs can be optimized further by using the OLSR protocol.

In this paper, a novel robust blind grayscale image digital watermarking scheme is introduced based on lifting wavelet transform (LWT) in combination with discrete cosine transform (DCT), Hessenberg decomposition, and entropy analysis for copyright protection of multimedia information. At first, the two levels of LWT are applied to the host grayscale image to improve the imperceptibility of the watermarking scheme and then the high-frequency sub-band of the 2-level of LWT is decomposed by DCT. Next, the DCT coefficients are divided into 4× 4 non-overlapping blocks. After that, Hessenberg decomposition performs on each selected block, whereas the first row, first column element of the upper Hessenberg matrix is utilized to hide the watermark. To evaluate the imperceptibility and robustness of the proposed digital watermarking scheme, the peak signal-to-noise ratio (PSNR), and normalized cross-correlation (NC) are utilized to measure the quality and the ability of the proposed watermarking scheme to robust against signal processing operations and geometric attacks. Experimental and analysis results have demonstrated that the proposed scheme is achieved a very good tradeoff between imperceptibility and robustness. The comparison with other scheme have shown that the proposed digital watermarking schemes have a superior performance in terms of imperceptibility and robustness than other

Counting the number of people in many estimation systems, such as still images or video frames, is a buoyant research area that is challenging in the field of computer vision. It plays a considerable role in a variety of applications, such as security, management, education, and commerce. The purpose of this paper is to suggest a system to estimate the number of people in digital still images based on the Face Detection method. This system composed of two parts: face detection and counting of detected faces. In the detection step, the Skin Color Face Detection method was applied on the input of a digital still image. In the counting part, the obtained detected faces by the Skin Color Face Detection method have counted to estimate the number of people in an input color image with simple software and simple low-cost hardware. The skin color face detection algorithm was tested using 133 images from the People Image Groups dataset, which contains about 2573 color images of people, to test the proposed system. based on the obtained results, the best precision achieved of the proposed Skin Color face detection algorithm was 85%

Face recognition is a biometric technique that can be used for a variety of purposes, such as national security, access control, identity fraud, banking, and finding missing children. Faces are highly dynamic and facial features are not always easily extracted, which can lead to discarding textural information like the smoothness of faces, a hairstyle that, might contain strong identity information. In addition, brightness, scale, and facial expressions play a significant role in the face-recognizing process. Therefore, face recognition is considered as a difficult problem. To figure out this problem effective methods using databases techniques are needed. This paper describes face recognition methods and their structure. Based on Wen Yi Zhao and Rama Chellappa work the face recognition methods are divided into three groups: a holistic approach, feature-based approach, and hybrid approach, where Principal Component Analysis PCA, a holistic approach method, is presented as a mathematical technique that can assist the process of face recognition. Also, the paper shows how the PCA is used to extract facial features by removing the principal components of the available multidimensional data.

Counting the number of people in many estimation systems, such as still images or video frames, is a buoyant research area that is challenging in the field of computer vision. It plays a considerable role in a variety of applications, such as security, management, education, and commerce. The purpose of this paper is to suggest a system to estimate the number of people in digital still images based on the Face Detection method. This system composed of two parts: face detection and counting of detected faces. In the detection step, the Skin Color Face Detection method was applied on the input of a digital still image. In the counting part, the obtained detected faces by the Skin Color Face Detection method have counted to estimate the number of people in an input color image with simple software and simple low-cost hardware. The skin color face detection algorithm was tested using 133 images from the People Image Groups dataset, which contains about 2573 color images of people, to test the proposed system. based on the obtained results, the best precision achieved of the proposed Skin Color face detection algorithm was 85%

Face recognition is a biometric technique that can be used for a variety of purposes, such as national security, access control, identity fraud, banking, and finding missing children. Faces are highly dynamic and facial features are not always easily extracted, which can lead to discarding textural information like the smoothness of faces, a hairstyle that, might contain strong identity information. In addition, brightness, scale, and facial expressions play a significant role in the face-recognizing process. Therefore, face recognition is considered as a difficult problem. To figure out this problem effective methods using databases techniques are needed. This paper describes face recognition methods and their structure. Based on Wen Yi Zhao and Rama Chellappa work the face recognition methods are divided into three groups: a holistic approach, feature-based approach, and hybrid approach, where Principal Component Analysis PCA, a holistic approach method, is presented as a mathematical technique that can assist the process of face recognition. Also, the paper shows how the PCA is used to extract facial features by removing the principal components of the available multidimensional data.