Embracing the Digital Horizon: Pioneering Commerce and Management Strategies for a Transformative Future

Congestion at Electric Vehicle (EV) Charging Stations (CS) is influenced by the activity of EV owners and their choice of charging stations. With a set price plan, certain charging stations become overcrowded with EVs waiting to charge, while others have available electric plugs. This situation negatively impacts the income of Charging Station Operators (CSOs) and the welfare of EV users.

This study proposes a dynamic pricing method to move EVs from crowded to uncongested CSs by adjusting charging fees at various periods. The issue involves scenario-based stochastic optimization to maximize CSO revenue. Furthermore, an attraction functional method is created to measure EV owners' charging choices, taking into account the effective characteristics of CSs. A genuine method for CS sites and EV routes is created using Quantum Geographic Information System (QGIS) software to compute distances between EVs and CSs. The suggested framework is evaluated in three stages and compared to the fixed-price strategy. The proposed dynamic pricing method reduces congestion in CSs, increases EV charging by up to 49%, and boosts CSO income.

This research develops and implements an edge management system, optimizes the corporate HR system using IoT-first technology, and examines the demand phase. After installing the software, hardware, and edge node management platform, the edge layer's possible interactions with the sensor, boundary, and cloud layers must be analyzed. Then, to ensure reliable communication between the various layers of the system, the organization's type-driven hyperlink selection method and the northbound multi-link transitioning method are developed and executed, respectively. The edge management system may fulfill the expansion, intelligence, and safety needs of IoT applications by implementing the aforementioned functionalities. Conducting thorough research is essential for identifying corporate functional and performance needs, creating a logical framework, and designing the system's architecture and other components. Individuals can adapt the system to their needs by utilizing its three prebuilt modules: organizational change management, remuneration and benefits, and people transformation. System components include staff change management, organizational management, benefits and pay management, and personnel data management. HR management is covered in these modules. The system meets design goals utilizing advanced tools and software.

The Internet is one of the new market economic innovations of the 21st century. The Internet of Things (IoT) is a new technology that integrates processors, the net, and portable phone networks. It has large potential applications and is considered one of the five key developing sectors. The IoT sector faces tremendous competition due to the fast growth of digital technology and the evolving market economy. The strength of an enterprise's market competitiveness is, to some degree, determined by the consequences of Human Resources (HR) management. This chapter analyzes the fundamental features of the Internet industry and examines the competitiveness of Internet enterprises to study industry HR management. It directly relates to several factors of HR and uses the AHP process to develop the Internet initiative hierarchy. Finally, this chapter investigates an IoT company's human resources organization quality assessment model, strategy, and optimization, which supports their market competitiveness with theory. 

It is helpful for management to have a clear understanding of the project's scope and trajectory early on, so they can make informed estimates about the project's primary engineering quantity and cost. Furthermore, it may aid managers in avoiding risks, making appropriate construction deployments, and preventing the introduction of adverse project elements; it can also play a significant guiding role in the project's ongoing construction management. In this research, we build a system to simulate the costs of building projects using artificial intelligence. The primary use of Building Information Modelling (BIM) in this system is to model the cost of construction engineering, which is then combined with the price file for analysis and evaluation. The chapter's intelligent model has the potential to impact building project costs significantly, according to experimental studies.

Supply chain management systems, such as order management systems, impact customer satisfaction and corporate profits. Order management software helps hardware stores keep sales and purchase data. Faulty records lead to unhappy customers, less money in storerooms, and slower sales. Customers submit orders, visit hotels or cafeterias to learn about the cuisine, and then pay; this requires time and human labor. E-commerce sites make it easier to purchase all our everyday necessities thanks to technology. In this sophisticated technological age, we have no app or website to acquire recipe ingredients. This document reduces ingredient waste, time, and cost. Provide recipes and quantities within 1–2 hours. This mechanism will enable us to deliver orders quickly to our customers.

In the present world, it is challenging to study sentiment and emotion in human multimodal language. However, in specific cases, such as calls, only sound information is available. In this work, we examined sentiment analysis and feeling identity freely. Current self-supervised learning models, especially speaker-mindful pre-preparation models, were used to prepare discourse, particularly in widespread discourse portrayals. For three sentiment tasks and an emotion test, three distinct sizes of universal models were examined. The study found that two types of sentiment analysis produced the best results, based on weighted and unweighted accuracy scores of 81% and 73%, respectively. When compared to previous methods that utilized multimodal fusion, our binary classification of unimodal audio data performed competitively. The models failed to make accurate predictions during an experiment, as evidenced by their inability to make precise predictions on tasks involving sentiment analysis and emotion detection. As the number of courses available increases, the reported performance decreases in the six-class emotion, three-class sentiment, and seven-class sentiment tasks may also be attributed to the inconsistent nature of the datasets.

Dyslexia is a condition that normally occurs in children with average or above-average intelligence. Children with dyslexia show symptoms of slow reading, writing problems, lack of communication, and also suffer from short-term memory. Studies show that education can have a significant impact on a child's life and influence their future job opportunities. Literacy is affected significantly when there is no clear understanding of the structure of the language and its phonological components. Dyslexic children especially face obstacles in reading and writing proficiency due to the linguistic component of language. To nurture children affected by dyslexia, various projects and programs are introduced to increase phonological awareness in children. Game-based learning is an effective method for enhancing children with dyslexia's interest in learning. The game-based learning model is interactive, attractive, userfriendly, and easily understandable by children. A Game-based learning model must address language-based learning difficulties that are considered both cognitive and emotional. In this prototype model, a 5x5x5 LED cube has been developed and is being tested for its performance.

Popular exercises, such as yoga, have numerous positive health benefits. However, incorrectly executed yoga poses can be harmful and lessen the advantages of the practice. In recent years, yoga positions have been investigated and altered using computer vision techniques. This study presents a novel Mediapipe-located cue-located calculating fantasy method for reconstructing yoga postures. The urged method is compared to good adjustment by requesting deep knowledge algorithms to identify the ultimate main physique parts in yoga poses. The computer, therefore, instructs the consumer to adjust their posture to achieve the appropriate adjustment as soon as possible. The projected method is validated using a dataset of yoga poses, and the veracity and real-world opportunity accomplishment effects are promising. In yoga classes, especially when conducted online or in other settings, the submitted method may be employed to increase benefits, though it also poses risks of harm to practitioners.

 Inventory Management System (IMS) is software that assists firms operating hardware stores to track sales and purchases. This project streamlines the process by eliminating paperwork, human errors, and manual delays. The Inventory Management System will track sales and available inventory, alerting business owners when it's time to restock. Sales, inventory, Customer Relationship Management (CRM), Human Resource Management (HRM), and account handling are just some of the business activities that the program documents. The application mainly allows a self-checkout environment supporting a takeaway system. The software is beneficial to firms that manage retail outlets and keep track of sales and purchases. In this application, the purchase and sales transactions are employed using Code Igniter, PHP, and Web Development tools to develop an inventory system. It provides a thorough corporate perspective and helps owners manage development and sales.

Popular website and e-commerce software platforms, such as CMS, are favored due to their simplicity and versatility in practice. However, website usability is one of the most significant quality elements that is difficult to quantify since it relies on various other variables. Thus, this article aims to determine which CMS is best suited for the usability and design needs of website developers. This study examines many ecommerce CMSs and outlines their characteristics that may aid with usability and design. This study uses E-commerce Total Quality Management usability. Choose Shopify, Open Cart, with the help of Magento, or WooCommerce, the most prominent e-commerce CMS. The design idea is evaluated using website design components. CMS characteristics that aid usability and design principles were found in this investigation. 

The popularity of buying things online has skyrocketed in recent years. During this time, most features of online buying have been refined, but others are still missing, such as the ability to negotiate prices with shop owners. To negotiate product costs, this article proposes utilizing a chatbot equipped with a voice assistant. A customer may ask the chatbot for help determining a fair price for an item. Both the buyer and the seller run the risk of having their finances squeezed when they shop online. Machine learning algorithms have been developed to aid in buying by predicting future outcomes based on past data to prevent compromise. Nevertheless, the accuracy of predicting prices might be compromised if the dataset is inappropriate or if irrelevant elements or attributes of the information are used. Because even one incorrect price forecast could result in substantial financial losses, e-commerce enterprises do not rely solely on such techniques. Furthermore, when data becomes too abundant or a characteristic is no longer available after the period used to make the model's prediction, very few models do poorly. The model proposed in this work maintains its accuracy and reliability by regulating these adjustments. 

Online marketplaces offer a global platform for a diverse range of items. To make online purchasing convenient for users, our suggested system offers a variety of features. Just as when buying in a physical store, users need product and systemspecific instructions when making purchases on these sites. Our e-commerce platform offers limitless talking services, and we incorporate an artificial chatting engine to provide stuff like this online. Users have the option to ask questions in the system during initialization of the e-commerce site. To acquire answers, the e-commerce system uses a pattern-matching algorithm to transmit consumer queries to the AIML Knowledge Base System. The user is then redirected to the system after receiving this response. Also, to demonstrate the chat system's multilingual capabilities, we are working on a Bangla-supported shopping assistant.

To increase client loyalty and satisfaction via digital transformation, banks are investing more in new technologies like artificial intelligence. This research uses CAMELS to examine how AI affects bank financial performance. The biggest bank in the private sector and a pioneer in AI technologies for customer experience, HDFC Bank, was chosen for the study. The authors have taken into account the four years preceding the adoption of AI (FY2013–2017), the four years following the adoption of AI (FY2019–2022), and 2018 as a transitional year for technological deployment. By comparing mean values, the research evaluates HDFC's financial performance before and after AI banking apps. Using SPSS and Excel, we tested a pair of sample t-tests on the secondary data we gathered. The results show that seven parameters—Tier 1 Capital Ratio, Business per employee, Dividend per share, Market price, Profit per employee, Cost-income ratio, and Expenses to Interest Earned Ratio—have improved noticeably. The positive effects of AI are becoming apparent, according to the authors, although major transformation may be delayed for some years. 

Deep learning and Artificial Intelligence (AI) struggle to automatically determine gender and age from visual cues or other traits. Despite the improvements in age and gender detection algorithms resulting from large datasets and deep learning, numerous issues remain. The absence of variety in training data can lead to bias and poor performance, particularly among underrepresented groups. Deep learning for age and gender determination raises ethical considerations about discrimination and abuse. This study will evaluate precise algorithms for gender and age to address these issues. They strive to reduce prejudice and ensure fair and reliable results for everybody by stressing the diversity of training data and ethical model development. This study aims to enhance age and gender identification while promoting fairness, accuracy, and ethical practices. Our primary aim is to incorporate ethical considerations into age and gender identification model development and implementation, thereby benefiting underrepresented groups. We also seek to optimize these models for real-time marketing and security applications. This project aims to provide accurate, fair, and ethical gender and age identification systems. By emphasizing ethics, they want to design models that uphold justice, transparency, privacy, and dignity. These endeavors strive to advance technology in a manner that aligns with social ideals and benefits society. 

These days, the population in big cities is growing rapidly, along with the need for comfortable living, as more people move from rural to urban areas. The distribution interference with the supply, protection, utilization, and quality of water, as well as other water-related issues, has grown in importance as cities have expanded in tandem with their populations. Resolving water supply concerns requires an effective monitoring and control system. The system uses water pressure and ultrasonic wave sensors, a motorized electrical water valve, a GSM module, a Raspberry PI, and an Arduino UNO microcontroller. This article focused on IoT-based water distribution regulation and monitoring. The project aims to enhance the community's water infrastructure by utilizing a reliable and cost-effective approach. Prototypes mimicked water distribution. Pumping station monitoring and operation were done via a front-end web app. The prototype system makes scientific conclusions using fuzzy logic. In conclusion, the experiment met all requirements for monitoring and managing water delivery using an IoT model. 

A consumer is said to have “churned” when they no longer connect with their usual service provider. An examination of customer churn is conducted to prevent the loss of existing clients. Churn prediction is becoming increasingly important in the telecommunications industry, as it involves analyzing the behaviors of various consumers to identify those who are on the verge of terminating their membership with a service provider. It is more cost-effective to keep current customers than it is to get new ones. This study focuses on several machine learning and Neural network approaches to forecast customer churn and identify customers who are closer to leaving their service from the relevant service provider. Specifically, the methods are used to identify customers who are more likely to leave the telecommunications service. Several machine learning techniques, including Logistic Regression, Support Vector Machines, Random Forests, AdaBoost, and Artificial Neural Networks, are employed in predicting the customer turnover rate. Data is obtained from the AT&T website, and learning models are assessed using two criteria: Accuracy and Area under the Curve. The accuracy criterion is given greater weight than the Area under the Curve criterion. We assess and measure the performance of several algorithms, and the one that ends up being the most accurate throughout all of the tests wins.

 Developing a mobile app allows hawkers and peddlers to share their location in real-time. This could help customers easily locate hawkers in their area and get information about the products or services they offer. Overall, sharing the live location of hawkers can be a useful tool for supporting these small-scale businesses and helping them to reach new customers. It may also provide people with additional choices and boost a community's vitality and variety. Most mom-and-pop stores and other small enterprises are run by one person or a small team. Street corners, marketplaces, and retail districts are common places to find them. These types of companies offer a wide range of goods and services, including food stands, shops, barbershops, and more. The ability to respond quickly and effectively to changes in consumer tastes is a major strength of mom-and-pop stores and other types of small enterprises. 

A spike in email filtering due to a large number of detected emails leads to yearly losses. One way to mitigate this loss is to categorize different types of suspicious emails, such as fraudulent or promotional messages from unknown senders. The first steps in identifying message categorization were based on simple approaches, such as word filters. More complex methods, such as language modeling based on deep learning, are already being used. The text classification problem is often addressed using Recurrent Neural Networks (RNNs), with Gated Recurrent Units (GRUs) being a popular variant due to their efficiency in capturing sequential dependencies.. Since classifying phishing emails is the focus of this study, GRU techniques were used. This study's results show that, in a dropout-free environment, GRU attained a high accuracy rate. A large amount of detection mail is created worldwide from several botnets, which impacts the limited mailbox capacity. They affect the security of private mail and the loss of communication space. The time required to identify and reply to detected emails is affected by them. Identifying suspicious emails is still considered a challenging job in the modern day. Due to email detection frequency, identification can be improved. The researcher constructs a Gated Recurrent Unit- Recurrent Neural Network (GRU-RNN) to identify emails. A new method was tested with a Detecting basis dataset. The procedure is 99.8% accurate. After considerable testing, the researcher concludes that the offered technique detects emails well.

Optimizing work scheduling efficiency is crucial for reducing carbon footprint and wasteful expenditure in computational quantum mechanics-based materials and molecular designing campaigns, which need more high-performance computer resources. Quantum Machine Learning (QML) methods provide computational price estimates for typical chemistry quantum activities. QML method of wall times shows a consistent decrease in out-of-sample prediction error with training set size for 2D nonlinear toy frameworks, individual point of view, geometric optimisation, and transition state calculations. They provide numerical proof for millions of living molecule organisms, including transition states, open and closed equilibrium shell structures, and a toy system with two different roles and three regularly used optimizers. The MP2/6-311G(d), B3LYP/def2-TZVP, CASSCF/VDZF12, local CCSD(T)/VTZ-F12, as well as MRCISD+Q-F12/VDZ-F12 levels of electronic structure theory are all taken into account. After training on thousands of molecules, QML-based wall time estimates considerably increase work scheduling efficiency across the board, compared to traditional indiscriminate job treatment. Cutting CPU time overhead by 15% to 85% is the result. 

 Indian banks have used AI technology in recent years. Banks are rapidly using AI to improve customer experience by exploiting their massive datasets. Farming, AI in banking focuses on client service and interaction, such as balance inquiries, micro statements, money transfers, etc.AI can leverage the massive quantity of organized and unstructured data to eliminate inefficiencies in the banking industry. It enhances customer experience, reduces operating costs and risks, detects fraud, streamlines lending facilities, and resolves customer issues promptly. The research examined AI's effects. With a structured questionnaire, the researcher collected a substantial amount of primary data from 50 respondents and analyzed it using Logistic Regression Analysis, F-test, and Descriptive Analysis. Through this investigation, the investigator will determine the influence over time.

In today's fast-paced, technologically-driven world, managers need a way to collect massive amounts of high-quality data, and data mining is the way to go. A human resource archival information Decision Support System (DSS) is built for management and decision-making using cloud computing's scalability, stability, flexibility, and extensibility. The seven levels that make up this system are as follows: system application, data standards conversion, network support, cloud computing support, data collecting, and decision support layers. It examines the benefits of resource sharing and integration in cloud computing. The working mode is also discussed. The solution addresses shortcomings of conventional archive administration, including limited data resources, difficulty in achieving isomorphism, and insufficient standardization of processing data from diverse sources.

Production management has traditionally focused on maintenance and expenses, as unexpected breakdowns reduce system dependability and asset return. Innovative maintenance methods that gather and handle shop-floor data may save costs and boost business sustainability. A machine monitor framework with condition-based preventative maintenance is presented in this study. To assist condition-based preventative maintenance, the latter analyzes shop-floor machine tool data via information fusion. The technique is implemented as a Cloud-based software service. The service determines component life based on machine tool and equipment operating data, such as process time and time spent machining per tool. Additionally, it notifies machine tool users and maintenance departments and allows mobile communication. A machining SME case study is used to demonstrate the application of the framework.

Modern Supply Chain Management (SCM) systems use several services to use IoT and cloud technology across delivery, distribution, and manufacturing. This article performs post-processing tasks such as supply chain tracking, item location analysis, and expected delays. It does this by presenting a comprehensive microservice design and implementation that consumes and semantically annotates feeds of information from online systems. The technique is evaluated via a set of deployment scenarios, and suitable application code is also inserted using easy workflowprogramming-based structures.

The accuracy of traditional cervical cancer screening methods is low since they rely heavily on the pathologist's subjective experience. Preventing cervical cancer requires regular colposcopy screenings. Over the last half-century, colposcopy has been an important part of the effort to reduce cervical cancer incidence and death rates, in addition to pre-cancer screening and therapy. Unfortunately, visual screening leads to increased effort, which in turn promotes poor diagnostic efficiency and increases the likelihood of misdiagnosis. When it comes to deep learning for medical image processing, the Convolutional Neural Network (CNN) method is better for cervical cancer. The CYENET and VGG19-TL methods represent two deep-learning CNN architectures suggested in this research for detecting cervical cancer utilizing colposcopy pictures. For the research, VGG19 is used as a transfer learning in the CNN structural design. To automatically categorize cervical malignancies using colposcopy pictures, an original method called the Colposcopy Ensemble Network (CYENET) was created. The constructed model is evaluated for its specificity, accuracy, and sensitivity. For classification, VGG19 achieved a score of 74.4%. The findings for VGG19 (TL) are deemed satisfactory. The VGG19 model seems to fall into the intermediate categorization group based on its kappa score. The suggested CYENET demonstrated excellent performance in the experiments, with a sensitivity of 93.5%, a specificity of 97.3%, and a kappa value of 89%. The CYENET method outperforms the VGG19 (TL) model by 10% in terms of classification accuracy, achieving 93.4%.

This study investigates the effectiveness of Convolutional Neural Network (CNN) architectures in lung cancer classification. By applying deep learning techniques, it evaluates various CNN models for the accurate identification of pleural cancer using medical imaging representations of healthy tissue. Through extensive experimentation and confirmation, the effectiveness of CNNs in accurately diagnosing pleural cancer and revealing their potential as healthy textures for medical image recognition in oncology is demonstrated. Today, lung cancer remains the leading cause of cancer-related deaths globally. As urbanization and industrialization advance, air pollution continues to rise, contributing significantly to the disease’s prevalence. Early detection is crucial, as treatment for lung cancer is far more effective in its initial stages. However, early lung cancer stages generally lack symptoms and are difficult to detect. This work classified lung nodules using SPIE AAPM-Lung CT imaging data. Deep Learning (DL) has been a prominent categorization method in recent years. DL methods, such as Transfer Learning (TL), reduce training costs and enable deep learning with less data. Researchers are experimenting with deep learning approaches to enhance the efficiency of CAD in lung cancer screening using computed tomography. Neural network models, including AlexNet, ResNet18, GoogleNet, and ResNet50, are used in this research for training purposes. The networks are employed for training and categorising CT images. By utilizing Convolutional Neural Networks (CNNs) and Transfer Learning (TL), we can achieve accurate and efficient detection of lung cancer on CT images. The method used to test the model includes the confusion matrix, recall, accuracy, specificity, and F1-score.

Image denoising is an area where deep learning algorithms have shown promise, but researchers have also identified significant differences between various methods. Discriminative learning, particularly effective against Gaussian noise, and deep learning-based optimization enable real-world noise estimation. However, the diversity of methods has hindered research comparing deep learning in image denoising. By classifying photos into four categories based on the type of Convolutional Neural Network (CNN) used, this research evaluates several deep image-denoising algorithms. The CNNs in question include those trained on incremental white noise, real noise, blind wavelet transform, and composite noisy images. It compares state-of-the-art methods by analyzing common aims and principles and by doing quantitative and qualitative evaluations on denoised datasets that are publically accessible. The study concludes by outlining the obstacles identified and providing recommendations for future research.

In recent years, one of the major risks has been web-based attacks. Criminals posing a threat often trick consumers into visiting malicious websites to conduct attacks. To identify harmful URLs, various methods have been implemented, including blacklisting. Along with these untrustworthy methods arose the tedious task of keeping a database of blacklist URLs up to date. In recent years, there has been an exploration into using machine learning algorithms to identify dangerous URLs. To train a prediction model, this technique examines several aspects of URLs and utilizes a dataset comprising both normal and malicious URLs. To reliably identify harmful URLs, this study proposes a MUD (Malicious URL Detection) model that utilizes three supervised machine learning classifiers: logistic regression, support vector machine, and one additional classifier. Preliminary findings show that the algorithm exceeded the comparison group.

The condition known as Obstructive Sleep Apnea (OSA) occurs when the airway becomes frequently blocked during sleep due to the relaxation of the airway muscles and tongue. Snoring, disturbed sleep (from choking or gasping for breath), and lack of energy upon awakening are common symptoms of Obstructive Sleep Apnea (OSA). Diagnosing OSA is time-consuming and expensive. Many people go undiagnosed or misdiagnosed for this reason. The correlation between OSA and face shape has been shown in earlier studies. Using a depth map of scanned faces, this study explored the potential of deep learning for medical diagnosis. Deep maps provide a wealth of additional information on face morphology when contrasted with the more simplistic 2-D colour picture. They can obtain around 69% validation accuracy by using transfer learning, even when working with a small sample size. They are making predictions for people who have either above moderate OSA (>16) or below moderate OSA (≤ 16). In conclusion, the results of the simulations showed that the suggested VGG 19 outperformed the current method.