Innovations in Electrical and Electronic Engineering

A subfield of forensic science called “digit al forensics” focuses on the examination and recovery of digital evidence from digital devices discovered at crime scenes. The gathered evidence must be securely stored to avoid manipulation or tampering. To ensure justice and make wise decisions, the integrity of the evidence is essential. Blockchain, a new technology, is used to preserve papers in a decentralized setting to address this problem. Blockchain creates a tamper-resistant ledger system by offering a safe and unchangeable chain of data, where each record is cryptographically connected to the one before it. In order to keep evidence in a dependable storage medium, this study suggests a safe chain of custody framework that makes use of blockchain. Only authorized persons can access or possess the evidence since every transmission of it is recorded on a private Ethereum blockchain from the moment it is seized. A digital evidence system that uses smart locks to physically store and secure the evidence is smoothly linked with the framework. The key to unlock the evidence is only in the possession of the Admin, an authorized entity, in order to maintain the integrity of the evidence submission and retrieval process. Multiple parties, including law enforcement organizations, attorneys, and forensic specialists, will benefit from our framework’s secure approach for maintaining the admissibility and integrity of the evidence.

As quantum computing advances, it poses a significant threat to the security of conventional cryptographic systems that rely on mathematical problems. The inherent power of quantum computing threatens to render conventional encryption techniques obsolete, thus prompting the development of post-quantum cryptography. In this work, we investigate the possible effects of quantum computing on current cryptography as well as the need for post-quantum encryption. We talk about how quantum computing is progressing right now and the many algorithms that are being created to address mathematical issues. Additionally, we give a general introduction of post-quantum cryptography and its many methods, including lattice-based, code-based, and hash-based encryption. Finally, we evaluate the strengths and limitations of post-quantum cryptography and its potential to withstand quantum attacks in the future. Our analysis reveals that post-quantum cryptography has the potential to provide robust and secure cryptographic solutions for the era of quantum computing.

5G networks use a network architecture based on SDN due to the efficiency, low cost, ease of management, and scalability provided by SDN. However, SDN has the issue of determining the routes and traffic optimization centrally. The problem of traffic optimization and routing in 5G networks has been solved through the use of deep learning algorithms such as Deep Deterministic Policy Gradient (DDPG). DDPG provides good results but suffers from the problem of overestimation bias and runs the risk of becoming unstable. These issues have been solved by an alternative deep learning algorithm called Twin-Delayed Deep Deterministic Policy Gradient (TD3). One of the changes in TD3 is training the agent with two Q value functions instead of a single Q value function and taking the minimum of two values. TD3 also uses delayed policy/target updates and smoothing of target policy. There is no mention of TD3 in literature for solving the problem of SDN routing, so this paper analyzes and compares DDPG (existing approach) and TD3 (proposed approach). A simulation environment consisting of the Omnet++ discrete event simulator was used to simulate a 5G network with SDN routing. Two different simulation runs were used—with DDPG and TD3. It was demonstrated that the TD3 approach provided a much better performance with lower latency.

The work presents a linear programming-based transportation model approach known as improved modified distribution load balancing algorithm (IMDLBA) to enhance the migration parameters. IMDLBA is a part of reactive load balancing mechanism that relies on the process of migration to deal with workload imbalances across the virtual resources. The important migration parameters considered in this work are migration cost, degree of balance, number of task migrations, and number of machines required for migration. The model has been reviewed with respect to existing meta-heuristics—improved weighted round robin (IWRR), honey bee behavior load balancing (HBB-LB), dynamic load balancing (DLB), and HDLB algorithms, in terms of above cited parameters which come under the class of quality of service (QoS) metrics. Experimental analysis and evaluations from IMDLB algorithm revealed the significant reduction in migration cost and improvement in balance factor—a metric that define the degree of balance if VMs. A balance factor of around 31% has been enhanced compared to the existing methods. The IMDLB algorithm also works by performing one time migration on a given set of tasks. The IMDLB algorithm reduces the number of task migrations by 28.5, 51.25, 58.33, 75.16, and 75.19% with reference to IWRR (time-shared), IWRR (space-shared), HBB-LB, DLB, and HDLB respectively. Further the minimum number of machine combinations required for performing load balancing is also achieved. The research article takes into consideration five UN sustainable development goals namely SDG7, SDG 8, SDG 9, SDG 11, and SDG 12.

The rapid expansion of the Internet of Things (IoT) has stressed the importance of energy-efficient communication protocols, particularly in networks operating under power constraints. This paper presents a unique approach for managing communication in energy-limited IoT networks using a deep reinforcement learning (DRL)-based communication protocol. By integrating Sparse Code Multiple Access (SCMA), Code Division Multiple Access (CDMA) techniques, and the Combined Experience Replay Deep Deterministic Policy Gradient (CER-DDPG) algorithm, we developed a novel protocol to improve the throughput of power-constrained sensors in an IoT network. Through comprehensive simulations, we compared the proposed protocol’s performance with benchmark systems like traditional DDPG and stochastic algorithms. The results reveal superior energy efficiency and throughput with the proposed protocol, establishing its potential to significantly enhance the performance of energy-constrained IoT networks.

The online social platforms witnessed enormous growth in its networked structure as users continue to connect and interact through e-social dialogues. This eventually causes the transformational emergence of online social systems into complex networks. These large-scale networks inherently allow social entities to get engaged in diverse forms of interactions, thereby resulting in dynamic and complex user behavior. This raises many distinct challenges in social networks, which requires cross-dimensional analysis of the complex information for societal benefits. Understanding these complex systems has become crucial for analyzing the behavioral characteristics for drawing essential inferences. Analysis and evaluation of social complex networks in the context of several parameters are performed with prerequisites of graph theory, computational statistics, and probabilistic models. Our research provides a review of the evolving challenges in the social domain with case studies being conducted over multiple social systems. The study covers major aspects of social network theory that reveal communities and the impact of the central behavior of users.

Nowadays, the processing of big data has become essential to extract valuable information from vast amounts of data generated by various systems. Traditional approaches to database management and data system supervision are inadequate in efficiently handling large datasets, and they often become outdated. Managing the substantial data generated by Vehicular Ad-Hoc Networks (VANETs) poses significant challenges. In this article, we present a two-step methodology that addresses these challenges by detecting anomalies and accidents, as well as predicting anomalies within road segments. This enables real-time calculation of the total time spent on road segments. Our methodology incorporates a database containing estimated real-time travel times within the network, facilitating optimal route selection for vehicles to minimize travel time and avoid or minimize traffic congestion and accidents along the way. The maintained database serves as input to machine learning algorithms that forecast the time plus location somewhere the likelihood of the accidents or higher traffic jams. Our simulation consequences demonstrate that the proposed methodology achieves improved road safety and effectively mitigates congestion by efficiently distributing traffic load across different roads.

Emotions are important components of a person's behavior. Convolutional neural networks (CNNs) for facial emotion identification is a fast developing discipline with applications in security, psychology, and HCI (Human Computer Interaction). This contrasting study investigates how CNNs can be used to identify emotions from facial expressions in pictures. A CNN model was trained and tested using a dataset of pictures that had been annotated with various emotions. This study underscores the need for more research in this field and shows the potential of CNNs for face emotion recognition.

Data centers play an important role in the modern computational virtual environment. The data center is made up of servers, storage devices, cooling equipment’s, and power delivery equipment’s to deliver general services such as platform-as-a-service (PaaS), software-as-a-service (SaaS), and infrastructure-as-a-service (IaaS). In this aspect, all the devices are generating more heat since all are electronics devices. It increases the power consumption and carbon emission in the environment also. As a result, it may lead to be a part of possibility of global warming. So, the carbon emission should be minimized in the entry level. In this paper, it is achieved this with the help of task scheduling. To minimize the same and improve the efficiency, we have tested three task scheduling algorithms: TPPC, RASA, and PALB. The experimentation results of all are these algorithms are compared, and then the differences in terms of efficiency and carbon minimization are also analyzed. The results are given in the comparative analysis.

This research paper gives a systematic and comprehensive review and meta-analysis, investing impact of antidepressants on tech industry employees with a specific focus on their medical history and inter-personal factors. Depression presents itself as a mental health disorder marked by persistent feelings of sadness, hopelessness, and a diminished interest or enjoyment in everyday activities. It is commonly accompanied by physical and cognitive symptoms. Developing predictive models for depression is essential for early intervention, reducing stigma, allocating resources efficiently, providing personalized treatment, and advancing research and understanding of the condition’s underlying mechanisms, mental health conditions. Additionally, 71% of tech workers acknowledged that their productivity suffers due to mental health issues, while 57% of employees within the tech industry reported experiencing burnout. This is why the subject of mental health in the tech industry should not be ignored, and hence, it is important for us to analyze the effect of antidepressants in this industry. And hence by examining relevant studies and research papers, we aim to provide a systematic and detailed analysis of the relationship between use of antidepressants, medical history, and their outcomes on specific parameters in tech employees.

With e-commerce companies, artificial intelligence (AI) has emerged as a crucial innovation that allows companies to streamline processes, improve customer interactions, and increase operational capabilities. To provide tailored suggestions, address client care requests, and improve inventory control, AI systems may evaluate consumer data. Moreover, AI can improve pricing methods and identify fraudulent activity. Companies can actually compete and provide better consumer interactions with the growing usage of machine learning in e-commerce. This essay examines how AI is reshaping the e-commerce sector and creating fresh chances for companies to enhance their processes and spur expansion. AI technology which enables companies to enhance their procedures and offer a more individualized customer experiences has grown into a crucial component of the e-commerce sector. Purpose of providing product suggestions and improve pricing tactics, intelligent machines may examine consumer behavior, interests, and purchase history. Customer service employees will have less work to do as a result of chatbots powered by artificial intelligence handling client queries and grievances. AI may also aid online retailers in streamlining their inventory control by anticipating demands and avoiding overstocking. The use of AI technologies can also identify suspicious transactions and stop economic losses. AI is positioned to assume a greater part in the expansion and accomplishment of the e-commerce sector as it grows in popularity.

When it comes to the provision of financial services, the integrated marketing communication (IMC) process is crucial in the creation and maintenance of client-provider bonds. This research presents a literature assessment on the theoretical basis for using marketing communication tools in the provision of financial services. This research is an attempt to bolster the little theoretical literature on the effectiveness of marketing communication techniques in the provision of financial services. Financial service providers use marketing communication as a channel for two-way exchanges with their clientele, with the ultimate goal of maximising the benefits their customers bring to the company. When it comes to providing financial services, an organisation’s success hinges on its ability to effectively manage its relationships with both current and potential consumers. As a result, it is important for practical reasons to be guided by well-defined marketing communications goals to identify the extent of usage and within the constraints of available resources. In this regard, businesses are free to establish specific communications objectives in accordance with their unique situations to direct the implementation of their IMC plan. This study aims to find out an impact of financial integration with IMC on business performance. This study is descriptive in nature. Primary data is collected with the help of questionnaire. The study finds that the financial integration in the IMC model has a statistically significant impact on business success.

With the significant rise in online examinations, the demand for proctors has grown exponentially, leading to resource constraints. Unlike offline exams with a few invigilators overseeing large groups of students, online exams require individual monitoring to uphold the code of conduct. However, the prevalence of unfair practices among examinees in online exams remains notably higher than in offline settings, resulting in an extensive, tiresome, and inefficient process. To address these challenges, we present “Eagle Eye”, a coherent and efficient system that employs eye gaze detection with machine learning and artificial intelligence. During the exam setup, examinees undergo a calibration test to establish a designated border-box area for eye movement testing. Data gathered from this detection enables classification of examinee behaviour as fraudulent or fair based on their gaze within or outside the box. When fraud is detected, alerts are sent to both the examiner and examinee, allowing timely actions as needed. To ensure accurate predictions, we have curated a bespoke dataset with the help of volunteers, providing unfiltered and authentic samples for training. The implementation of Eagle Eye seeks to enhance online exam integrity and streamline the proctoring process.

The use of conversational agents is recognized as a significant technological achievement that makes use of recent advances in machine learning and processing of natural languages. These “agents” which are considered to be computer programs enable effortless communication with users in natural language. Conversational bots have a lot of potential thanks to the recent integration of the processing of natural languages and artificial intelligence. In order to create an intelligent conversational bot, this research paper delves deeply into the incorporation of deep learning techniques. The implementation of a sequence-to-sequence simulation strengthened by a structure consisting of encoders and decoders is the main focus. A long-short cell memory recurrent neural network occupies the focal point of this architecture. The encoder facet is in charge of understanding user inquiries, and the decoder facet produces appropriate responses, resulting in an expert conversational system.

There is no life without water. All humans, plants, and animals need water to live. It is important to know if drinking water, a resource of human life, will be enough for everyone now and in the future. Access to clean water and hygiene is an important human right and part of the health safety policy. At the national, state, and local levels, clean water is a critical problem for health and development. This work's primary goal is to use various modeling techniques based on machine learning, deep learning, and ensemble learning to measure water quality using hyperparameter tuning of each algorithm. We have used SVM, RF, XGBoost, DT, and LGBM model stacking and voting ensemble for efficient and fast prediction. PH, chloramines, hardness, solids, sulfate, organic carbon, conductivity, trihalomethane, turbidity, and potability were the parameter used as a feature vector. A different machine, deep, and ensemble learning algorithm was used to evaluate water prediction, and the effects are compared on the accuracy, ROC AUC values, precision, recall, F1-score, MCC, and kappa score. In addition, the Freidman Ranking is also used to evaluate the model's efficiency. According to related studies, ensemble learning-based models are the most effective.

In past two decade’s farmers of Punjab suffering huge losses in farming due to changing climatic patterns. Irregular rainfalls, hail storms, windstorms, drought, heat waves, cold waves, etc. are the major factors. Farmers are facing challenges in adapting new farming practices to the ever-changing climate conditions, all while incorporating advanced agricultural technologies and innovative seed varieties. The Internet of Things (IoT) presents a promising solution to address future challenges. Traditionally, agricultural production heavily relied on weather patterns. However, the current climate crisis has significantly disrupted this norm. While older seed types still yield high outputs when properly cared for, newer seed varieties have adapted to specific climatic and moisture requirements. In this research, our objective was to train machinery to identify crucial developmental stages in the crop's life cycle. This way, it can provide farmers with guidance on how to effectively manage the equipment based on the crop's thriving conditions. Rather than aiming to establish predetermined thresholds, this enhanced agricultural initiative strives to preserve the environment while catering to the shifting needs of the crop throughout its entire life cycle. Depending on the crop's current stage and anticipated duration, farmers may explore new approaches. Any device equipped with Internet connectivity and sensor capabilities can be employed to carry out these procedures. We achieved 95.89% accuracy with our proposed model.

The rapid development of IoT and its associated hardware and software systems has facilitated the design, development, and implementation of smart home automation systems. But the existing systems are either operated using physical electrical/mechanical switches only or operated using mobile app only but not both or interchangeably. It means once a device is switch-on using a physical switch it cannot be switch-off using the mobile app developed for that device. This research paper is going to solve this limitation. This paper presents a smart home automation system using NodeMCU (an open source IoT platform) and Firebase (a powerful cloud service). This presents a mobile app-based home appliance control system which provide the flexibility to control even a physical switch operated appliances remotely. The proposed system is physically developed, implemented, and tested in real-time environment and proved successful.

Grading of gallbladder cancer (GBC) is pivotal for the diagnosis and treatment planning of patients suffering from this disease. Radiomics has emerged as a non-invasive, imperative, and efficient way for disease diagnosis and prediction with the use of machine learning approaches on medical data. Given the large dimensionality of the data, it is important to choose the most significant features to aid in improved classification of patients with respect to the subtypes/grades of GBC. This paper proposes a novel ensemble ranking-based approach called EnRaFS, for feature selection to grade GBC patients’ using CT scan images. It combines the results of multiple feature selection methods to improve the accuracy of the ranking. The ranked features are then used to train the machine learning model to predict the grade of the cancer. The proposed approach has been evaluated on a dataset of 105 patients diagnosed with GBC and compared with other state-of-the-art feature selection methods based on accuracy measure. Our study concludes that the proposed approach can be used as an effective tool for grading GBC, which can help clinicians to make more informed decisions about the treatment of the disease.

Deepfake technology, powered by deep learning algorithms, has rapidly evolved in recent years, enabling the creation of highly realistic synthetic media that can deceive human perception. Unmasking deepfakes: Advancements, Challenges, and Ethical Considerations is a comprehensive review that examines the advancements in deepfake technology, the associated challenges, and the ethical considerations that arise in the context of this emerging field. Deepfake algorithms have the ability to produce phoney audiovisual content that is hard to distinguish from authentic content. It now appears to be challenging to distinguish between authentic digital content and fraudulent content spread around the Internet in this era of the cyber age. Cybercriminals frequently employ this technology to trick security systems. If we are not careful, deepfake technology could pose a severe danger to identity verification in the future. Deepfake content may easily be produced by amateurs using free and open-source software, which makes it simple for them to produce technically excellent content. Give an introduction to deepfake, and a brief on deepfake creation and detection techniques.

In this paper, we developed a convolutional neural network (CNN) architecture-based deep learning method for height identification. Our model learns to identify parts that are essential to determining a person's height from an input image. A large collection of labeled photos with a variety of heights, positions, and camera angles is used to train the CNN architecture. We assess our model using a number of benchmark datasets and contrast it with the most advanced height detection techniques currently available. Our findings is the demonstrate of such datasets, our methodology outperforms traditional methods and reaches state-of-the-art performance. We also demonstrate the robustness of our model to changes in illumination, perspective of the camera, and occlusions. Conclusion Our developed deep learning method for height identification represents a considerable advancement over existing techniques and shows the power of deep technology in the solution of challenging computer vision issues. Our findings indicate that our model can be applied to a variety of situations where height assessment is necessary, such as crowd analysis, surveillance systems, and human–computer interfaces. In fields including security, marketing, and health care, gender detection is a critical duty. It has been demonstrated that deep learning is an effective method for detecting gender because of its ability to spot complex patterns in input. In this research, we developed a novel deep learning method for gender detection that uses the CNN, or convolution neural network, architecture. The made model learns to identify whether an image is masculine, or female based on its input. To evaluate our model, we use two publicly available datasets: CelebA and LFW which are two publicly accessible datasets that we use to assess our model. We use a portion of the data to train our model and the remaining data to test how it performs. Our studies show that the proposed model delivers cutting-edge results on the two sets of data, with an accuracy rating of above 95%. To assess the contributions of each element of our model, we also carried out several ablation experiments. Our findings demonstrate that the accuracy of the model is greatly enhanced using many convolutional neural networks and the addition of batch normalization. Overall, our created deep learning method for gender detection shows the effectiveness of CNNs in this job and lays a solid groundwork for future research.

The Internet of Things (IoT) has transformed healthcare by providing continuous remote patient health monitoring. Ensuring the security and privacy of patient health data in such IoT-enabled contexts, on the other hand, is a critical concern. This study proposes a unique method for improving IoT-based healthcare security via continuous authentication, utilizing deep learning, especially the Long Short-Term Memory (LSTM) model. The suggested system continually analyzes user behavior and health state, using biometric data to provide seamless and secure authentication. Multiple security credentials, including Personal Identity Number (PIN), password, and biometric identity, are used in the architecture to provide effective protection against unauthorized access attempts. Using Arduino Uno and smart devices, data from a broad array of sensors connected to patients are gathered, and a complete dataset is created for training the LSTM model. The performance of the suggested system is assessed using multiple performance measures such as accuracy, precision, recall, specificity, and the F1-score. The findings show that the model is very accurate and efficient at discriminating between legitimate and unauthorized users. The system consistently outperforms previous research efforts, demonstrating its superiority in predicting authentication answers. Furthermore, continuous authentication enables real-time monitoring and proactive identification of suspicious actions. The scalability, versatility, and open-source characteristics of the proposed technology ensure its use in a variety of healthcare contexts. This study helps improve IoT-enabled healthcare security by building confidence in users and stakeholders and increasing the state-of-the-art in safe and trustworthy healthcare data monitoring in the IoT ecosystem. The suggested paradigm sets the groundwork for future improvements in continuous authentication and healthcare security as the IoT ecosystem grows.

This research paper explores cross-project defect prediction as a means to improve software quality assurance (SQA) practices. Traditionally within-project defect prediction methods face challenges due to limited training data and project-specific characteristics. In contrast, cross-project defect prediction leverages knowledge transfer from multiple projects to develop more robust and generalizable defect prediction models. The study investigates various knowledge transfer strategies, such as instance-based, feature-based, and model-based transfer, and conducts extensive experiments on diverse software repositories. The results demonstrate that knowledge transfer techniques outperform traditional methods, offering higher accuracy and improved generalization to unseen projects. The paper also analyzes key factors influencing cross-project defect prediction success, providing practical guidelines for real-world SQA applications. By enabling effective defect prediction, this research contributes to enhancing software quality and maintenance.

The growth of social networking sites and online platforms has brought about an unprecedented surge in user-generated content. However, along with the immense benefits of increased communication and information sharing, there has been an alarming growth in toxic and offensive comments. Detecting and moderating such comments is crucial to maintain a healthy and safe online environment. In this research, we propose a multilingual toxic comment classification system that leverages the power of Bidirectional Long Short-Term Memory (BiLSTM) neural networks. We use a comprehensive dataset which contains a diverse range of toxic comments in multiple languages. We employ a BiLSTM architecture because it is effective at detecting both contextual and sequential dependencies in text data. We train our model by combining word embeddings with character level embeddings in order to capture the semantic and morphological information found in the comments. Multiple cutting-edge methods are used to compare the model’s performance, including RNN and LSTM. The experimental findings show that the suggested model performs competitively in classifying multilingual toxic comments, surpassing other approaches with an accuracy of 94.21%.

The research review explores the intersection of AI, IoT, and phishing attacks, highlighting their applications, risks, and vulnerabilities. It discusses various types of phishing attacks and existing solutions for detection and mitigation. The paper also addresses future directions, challenges, and the importance of interdisciplinary collaboration and user awareness. Real-world examples are examined to illustrate the impact of phishing attacks. The conclusion emphasizes the need for ongoing research and a resilient, secure AI and IoT ecosystem.

The increasing prevalence of manipulated videos across various domains highlights the critical need for effective video forgery detection methods. In parallel, the demand for authentic and trustworthy images grows, emphasizing the importance of detecting digital image forgery in our society. Blind tampering has emerged as a prominent trend in visual content manipulation. This paper presents a comprehensive investigation that addresses the diverse challenges faced in previous research studies. Recent advancements in neural network-based approaches have shown remarkable efficiency in detecting image forgery by uncovering concealed characteristics within images, thereby enhancing accuracy. In this work, an extensive inter-frames video forgery detection approach is used. The primary goal is identifying and detecting manipulation between frames in a video sequence. The report examines techniques for detecting forgeries in images and the challenges posed by inter-frame and intra-frame fakes in videos. Also, emphasis is placed on frequently utilized datasets in this field, which can assist new researchers exploring this study area. Experimental results demonstrate the proposed approach's efficiency and robustness, highlighting its remarkable accuracy in detecting inter-frame video forgeries. This contribution to the field of video forensics provides a valuable tool for verifying the integrity and authenticity of video content.

Couples who are having trouble becoming pregnant now have hope thanks to in vitro fertilization (IVF), a revolutionary medical advancement. However, the IVF procedure calls for a large number of stakeholders, intricate paperwork, and highly confidential management of information that frequently results in inaccuracies, mistakes, and worries about data confidentiality and confidence. In this study, the revolutionary potential of the blockchain and smart contracts enabling the treatment of IVF is investigated. The IVF procedure may be accelerated by utilizing smart contracts, resulting in improved effectiveness, openness, and confidence among everybody involved. The paper explores the primary advantages of using smart agreements in IVF, including automation, implementing obligations under contracts, doing away with middlemen, assuring confidentiality and anonymity, and enabling safe and auditable operations. The implementation of electronic agreements and blockchain-based technologies in the discipline of IVF is also investigated, along with the problems it may face and possible alternatives. This study offers insightful information about the use of intelligent agreements and blockchain technology in the field of IVF, accompanied by conducting an in-depth evaluation of the literature on the topic, research papers, and interviews with professionals. The results demonstrate the possibility of lower prices, more accessibility, higher success rates, and better patient experiences in the IVF field. In general, this study intends to illuminate how blockchain and smart contracts have revolutionized IVF technological advances, opening the door for a more effective, transparent, and reliable IVF procedure.

Door lock security is an important consideration for any homeowner or renter. There are various options available for door locks, including deadbolts, keyless entry systems, and smart locks. It is important to choose a lock that fits your specific needs and budget. Most of the current IOT-based smart door lock systems are based on third part apps like Blynk which are works using Wi-Fi within a limited area of home or office premises. The proposed system removes these both limitations. This paper presents an IoT-based smart door lock system with four core functionalities. An Android app is developed by the authors which offers a secure login and registration mechanism, facilitates capturing photos of individuals positioned in front of the door and remote locking/unlocking using the app. The Android app also displays captured photos and videos for easy accessibility and monitoring. The system incorporates face recognition technology utilizing the ESP32 CAM module to allow the door automatically unlocks upon recognition. Most importantly, in case of non-recognition, it incorporated an alert message functionality into the system. If an individual stood in front of the ESP32 CAM for a duration exceeding 30 s, an alert notification would be dispatched to the Android app, notifying that “Someone is at the door for more than 30 s”. The functional prototype of the system is fully developed, implemented, and tested in real time environment which proves successful.

Smart home automation refers to the use of advanced technology to control, monitor, and automate differences in a home, such as temperature, security, and lighting, remotely through a centralized interface. The Internet of Things (IoT) has transformed the way the authors interact with technology and our environment. Home automation systems are a very rapidly growing interest among the people of this generation, and it has gotten a considerable amount of attention after the introduction of communication technologies. In the field of smart homes, IoT has played a very significant role in helping it grow. IoT in its most basic terms is connecting things like software and sensors to the Internet, enabling them to collect and exchange data without human intervention. IoT deals with AI sensors, cloud messaging, networking, etc., and aims to deliver complete information at the right time. IoT-based systems have greater transparency, control, performance as well as efficiency. The two primary concerns with home automation are security and energy utilization. This paper highlights different methods of achieving smart home automation via several different technologies such as Bluetooth, Global System for Mobile (GSM), Zigbee, and Dual-Tone Multi-Frequency (DTMF). In this paper, the authors will delve into the details of IoT-based smart home automation and its various components, benefits, and challenges as well as its impact on society.

Diabetes is a metabolic disorder comprising high glucose level in blood over a prolonged period in the body as it is not capable of using it properly. Diabetes is a major cause of blindness, kidney failure, heart attacks, stroke, lower limb amputation, retinal damage, and foot ulcers. The condition is a result of the inter-linkage of lifestyle choices, xenogenetic, psychological, socioeconomic, medical disorders, and geographic attributes. Machine learning-based decision support systems for the prediction of chronic diseases have become immensely popular for better prognosis/diagnosis support to health professionals. Current computational methods for diabetes diagnosis have some limitations and are not tested on varied datasets or people from different countries which limits the practical use of prediction methods. This study identifies classifiers which work with optimal accuracy over three ethnicities. Three unique datasets were identified for this study which are an Indigenous population of USA, European population, and South Asian population for accurate prediction, diagnosing, and treatment of disease. Machine learning algorithms were applied on the datasets, and a comparative study was made. For South Asian ethnicity, GPC, RF, DT predicted with accuracy of 91.62% each. For European ethnicity, the same was performed with 97%, 98.2%, and 97.8%, respectively. For Indigenous Tribe of USA when GPC, RF, and DT were applied, the performance was 61%, 78.6%, 71.8%. SVM and LDA performed better with 80.2% for Indigenous Tribe of USA. Random forest performed with high accuracy on South Asian and European population and comparable accuracy for tribe of USA. Our study provides a base for reducing the gap in polygenic risk prediction accuracy.

For many years, newspapers have been excellent providers of information. To extract meaningful information, it is imperative to digitize these newspapers. In the case of Punjabi newspapers, the same is necessary. In this study, we used newspaper image segmentation to separate the various articles from the Punjabi newspapers. Different barriers in the extraction of Punjabi newspapers are discussed. Also, we have assembled a dataset of 400 newspaper images, and these images have been trained using cutting-edge object detection models Faster RCNN. The experimental results show that these models for extracting articles produced good outcomes with average accuracy of 81.8% for all classes.

The aim of this study is to explore the development and integration of a smart parking system enabled by advanced AI, which relies on Internet of things (IoT) technologies. The intent behind this system is to enhance the utilization of parking spaces while uplifting driver experience, all possible through delivering dynamic updates regarding parking availability. The technical infrastructure employed for building the system revolves around an amalgamation of smart sensors and cameras using IoT interactions to collect data associated with parked vehicles. This information gets subsequently processed by sophisticated AI algorithms that offer up-to-the-minute details about available parking spots stored in a prevalent database. Further, owing primarily to the technology’s highly functional characteristics, each occupant status predictably ascertains effective insights into plausible available space during driving, via display boards or mobile-based applications. After conducting a comprehensive review, the suggested smart parking system has been found to surpass traditional parking systems in critical areas such as accuracy, dependability, and efficiency. As a result of this new system being utilized, traffic congestion will significantly decrease, resulting in a noticeably improved parking experience and higher park management revenues. One profound advantage of the proposed system is that it has the capability of collecting real-time data on available parking spaces. The benefit of having access to this data is that car owners are guaranteed shorter searches. This study offers ideas for additional study and development in this field and demonstrates how AI with Internet of things technologies might enhance urban transit networks. By improving parking efficiency and giving drivers a better parking experience, the adoption of an “artificial intelligence-enabled smart parking system” offers an opportunity to revolutionize the parking sector.

Cloud services provide numerous advantages over private servers, but there are certain drawbacks, such as vendor lock, budget overrun due to an unexpected spike in computing demand, and migrating existing system to the cloud. In this study, we deployed, measured, and compared the performance of application hosting on virtual private servers with integrated cloud storage and cloud-based hosting. Cloud services and virtual private servers (VPSs) may be evaluated using a range of performance indicators that can be applied to various components of the service. We recommend combining a cloud service with an application running on a virtual private server to store and retrieve file objects, allowing the application to grow by utilizing cloud platforms like Amazon web services and other public cloud technology. Organizations may create safe and scalable apps while avoiding budget overruns and moving an existing system from a private server.

Artificial intelligence (AI) has revolutionized multiple sectors through its growth and diversification, notably with the concept of collaborative learning. Among these advancements, federated learning (FL) emerges as a significant decentralized learning approach; however, it is not without its issues. To address the challenges of trust and security in FL, this paper introduces a novel blockchain-based decentralized collaborative learning system and a decentralized asynchronous collaborative learning algorithm for the AI-based industrial Internet environment. We developed a chaincode middleware to bridge blockchain network and AI training for secure, trustworthy and efficient federated learning and presented a refined directed acyclic graph (DAG) consensus mechanism to reduce stale models’ impact, ensuring efficient learning. Our solution’s effectiveness was demonstrated through application on an energy conversion prediction dataset from hydroelectric power generation, validating the practical applicability of our proposed system.

Adulterating clarified butter involves adding impurities and subpar substances to pure clarified butter with the intention of increasing the quantity and maximizing profits. Such adulteration in food and other consumable products has a direct impact on human health, compromising the nutritional value of the substance. This study aims to provide a comprehensive analysis of various techniques employed in detecting adulteration in clarified butter. Leveraging the advancements in machine learning technology, the study explores the analysis of existing data collected from different products and laboratories to identify patterns indicative of clarified butter adulteration. The research focuses on quantifiable measures used to determine the level of adulteration in various products, ultimately contributing to a better understanding of machine learning algorithms suitable for detecting adulteration in clarified butter. The findings of this study serve as a foundation for enhancing the existing framework and guiding future research endeavors in the field of machine learning-based detection systems for clarified butter adulteration.

Face recognition and detection is an important research topic in computer vision, which has been widely used in various applications, such as security, biometrics, and law enforcement. Machine learning has played a crucial role in the development of accurate and efficient face recognition algorithms. In this paper, we review the literature on face recognition and detection using machine learning, with a focus on the methods, techniques, and applications. We discuss the different stages of face recognition, including face detection, feature extraction, and classification. We also highlight the challenges and future directions in face recognition using machine learning. The main focus of this research is to a proper system with artificial intelligence improved facilities using Machine Learning. The research work can overcome all the limitations of the existing research. The research presents a proper protection and it also reduces the manual work of a person. The present research has several benefits and a lot of strategies to work with. In this exploration continuous Illustrations UI Based Computerized Facial Acknowledgment is utilized. The calculation utilized in the proposed approach are Head Part Analysis (PCA) and the HAAR Outpouring Calculation. This research enabled with a cuttingedge era of Artificial Intelligence and Machine Learning. The gain of this research as it detects the face with the help of facial recognition method, iris detection, biometric detection.

In India, agronomy industry needs automation for monitoring the overall farm and plant health as due to the presence of plants’ diseases and ecological inadequacy which causes significant damage and dissipation to agriculturists. Therefore, various geographical conditions are required for plants and crops growth as it needs a humid climate with rainfall of 200 and temperature above 25 °C. Thus, various conditions required for farming are moderate temperature, rainfall, and lots of sunshine. As it requires lots of drainage for the fertile soil. Although India is the second largest manufacturer of various types of dry fruits, feedstock, and no vegetables, also they uses various methods of cultivation for the farming process like manuring, irrigation, weeding, cultivation, and sowing for better quality crops that grow in the primary step of sowing. The investment of pesticides in the Indian industry sector in 2022–23 is nearly 140 crore which is done by SP Gupta Chief Financial Officer of Indian Pesticides Limited. Various types of pesticides have been used for the betterment of farm like insecticides, bactericides, and fungicides for killing insects and various pests but the overuse of pesticides harm the fertility of soil and land for good quality crops and growth; thus due to these, farms get damaged and lands get infertile, because of these, farmers cannot do farming on that land to overcome this issue; this paper shows the solution for the farmers and land by using a prototype robot by using IoT which holds a record of plants as well as monitor the farm in any weather if any insect gets detected by the caretaker; the advanced robotic mechanism starts activating and sprays pesticides on the affected portion of plants. Due to this, land can be saved by unwanted spraying of pesticides and infertility of soil.

Segregation of images is a critical step in processing images, computer vision, and a variety of other disciplines. The technique involves decomposing an illustration into numerous components or components, every single one that consists of an ensemble of elements with identical features for example African descent, frequency, or consistency. The most important objective of appearance, the intention of fragmentation seems to reduce complexity or customize the mathematical representation of an illustration in a way that is more readily reasonable and more straightforward for assessment. It is widely employed to locate boundaries and features in photos, and this is favorable in an assortment of industries including clinical imaging, object detection, recognition, and autonomous vehicles. Several image segmentation techniques are available, and in incorporating thresholding, zone is an area-based differentiation and corner-based recognition. A threshold segment is a simple and commonly used technique that involves setting a threshold value and dividing the pixels into two classes based on their intensity values. Region-based segmentation involves grouping pixels based on their spatial proximity and similarity in characteristics, while edge-based segmentation involves detecting edges or boundaries in an image and using them to separate different regions.

The integration of cutting-edge technology, including deep learning, smartphone capabilities, and wearable devices, has sparked a transformative revolution in fall detection systems, offering real-time monitoring and swift response in the event of a fall. This research study presents a fall detection system that harnesses advanced deep learning techniques, particularly 1D convolutional neural networks (CNNs), to achieve remarkable accuracy scores of 91% and 92%. Rigorously evaluated using the Sisfall and UMA Fall datasets, which consist of 9 and 25 features, respectively, obtained through meticulous hand engineering, this system demonstrates its efficacy in detecting falls. The potential of this advanced fall detection system lies in its ability to significantly enhance the safety and well-being of individuals by enabling timely assistance after a fall. By leveraging the power of artificial intelligence and state-of-the-art technology, the system promises to amplify the efficiency of fall detection in real-world scenarios, providing reassurance and peace of mind for both individuals and their caregivers. Particularly beneficial for vulnerable populations like the elderly, this technology holds the promise of mitigating the risk of severe injuries and fatalities resulting from falls. The study’s findings underscore the substantial progress that can be achieved in fall detection by seamlessly integrating deep learning, smartphone technology, and wearable devices. This integration paves the way for a future where prompt assistance becomes standard practice, reducing the potential consequences of falls and ultimately improving the quality of life for those at risk. As this research sheds light on the immense benefits of advanced fall detection systems, it serves as a significant step forward in ensuring the safety and welfare of individuals, fostering a safer environment for everyone.

Acceptance into a graduate program must be part of a student’s academic journey. Every year, a huge number of people apply to schools and universities, and the admissions process may be tough and time-consuming. Many factors are considered while evaluating a student's application, including academic achievement, test scores, LOR, and extracurricular activities. However, selecting the best choices can still be arbitrary and prone to mistakes. As a result, it is required to develop a more efficient and objective technique of evaluating an applicant's chances of admission to a graduate course based on their application materials. The purpose of this study is to develop a ML model that can predict a student’s prospects of acceptance into a graduate school. The model will be trained using a dataset containing different characteristics, such as GRE scores, GPA, and letters of recommendation. The dataset will be preprocessed to cope with missing values, outliers, and categorical data. A variety of ML methods, including LR, DT, and SVM, will be used to build the model. The algorithm’s efficacy will be measured using a variety of measures, including accuracy, precision, recall, and F1 score. The best-performing model will then be picked and used to evaluate the admissions outcomes of fresh applicants.

Precision agriculture is a forward-thinking and smart method of cultivating crops that involves the meticulous control of vital factors like nutrients, air, temperature, water, and ongoing surveillance throughout the entire farming process. In this research work, we propose an IoT-based model focused on fertilization detection, aiming to accurately predict the optimal quantity of fertilizer required for crops. The model operates through three key phases: acquisition, transformation, and analysis. During the data acquisition phase, relevant data is collected, capturing vital parameters essential for fertilization. The collected data undergoes change, wherein it is appropriately formatted and migrated to a cloud platform to ensure compatibility and accessibility. Subsequently, a comprehensive analysis is conducted, unveiling valuable insights. Based on this analysis, an appropriate response is generated and communicated back to the farmer, providing practical guidance for implementation within their specific cultivation region. This model will contribute to sustainable agricultural practices and empowers farmers with the tools to achieve enhanced productivity and profitability.

Brain tumors are one of the most progressive diseases affecting both children and adults. Brain tumors spread quickly and, if not treated properly, limit the patient’s chances of survival. It is important to detect malignant brain tumors as early as possible. Proper treatment planning and correct diagnosis are very important to prolong the life of the patient. The most precise method for identifying brain tumors is via magnetic resonance imaging (MRI). Finding brain tumors can be difficult because tumors vary in location, shape, and size. This study describes an MRI-based brain tumor segmentation method. To detect brain tumors, we can use architectures of that combines Convolution Neural Network (CNN), also known as Neural Network (NN), with visual geometry group (VGG 16) transfer learning to identify brain cancers. This study includes a literature analysis on deep learning models in order to discriminate between binary (normal and pathological) and multi-class (meningioma, glioma, and pituitary) brain cancers.

Ubiquitous computing, also known as pervasive computing or ambient intelligence, has emerged as a prominent field of research and development in recent years. This review aims to provide a comprehensive overview of ubiquitous computing, covering its key concepts, historical development, enabling technologies, applications, challenges, and future directions. The review synthesizes a wide range of literature from academic research papers, conference proceedings, and industry publications. It highlights the evolution of ubiquitous computing, explores its various components, discusses notable applications in different domains, and examines the challenges and ethical considerations associated with its adoption. The review concludes by discussing potential future developments and emphasizing the transformative potential of ubiquitous computing in shaping our future technological landscape.

The outbreak of Covid-19 has triggered a worldwide problem, especially in Asia and America. The World Health Organization (WHO) declared the sickness a pandemic on March 20, 2020. It arrived in waves, and most countries worldwide have now experienced two waves and are on the approach of experiencing the third. The goal of this study is to build up and certify a Computer-Aided Diagnosis (CADx) system for distinguishing between COVID-19-positive patients and non-COVID Patients people. Chest X-ray (CXR) images will be used to accomplish this. From public datasets which we got on GitHub 2295 CXR images were obtained which included 712 COVID-19 positive and 1583 normal. The proposed CADx system utilized a Conventional Neural Network (CNN) model for data argumentation and CNN was built, compiled, and trained with the help of TensorFlow and Keras. For the sake of appraisal, our datasets were separated into three categories: Train/Test and Validation. The three sets’ accuracy was evaluated and the results for Training, Validation, and Test were 97.77%, 97.81%, and 97.72% respectively. In the end, this study was able to create a precise Computer-Aided Diagnosis system for the two categories of classification.

Cloud computing has revolutionized the business landscape by providing convenient access to data and applications via the Internet. However, its widespread adoption has also introduced significant security challenges. Blockchain, a decentralized and tamper-proof technology, offers promising solutions to tackle these security concerns in cloud computing. This paper aims to explore the potential of blockchain technology in bolstering security in cloud computing. It begins by examining the existing concept of security in cloud computing and introduces blockchain technology, highlighting its components, features, advantages, and limitations. Furthermore, it investigates the current blockchain-based solutions for securing cloud computing and conducts a comparative analysis to understand their benefits and limitations. By delving into these aspects, the paper seeks to provide a comprehensive understanding of how blockchain technology can enhance cloud computing security. Its primary goal is to inspire and encourage future research and development in this field, promoting the continued exploration of blockchain’s potential in ensuring the safety and integrity of cloud-based systems.

The IoT-SyringeX project presents a state-of-the-art solution for automating injection pumps through the integration of cutting-edge IoT technology. The objective of this project is to enhance the efficiency, accuracy, and safety of medical injection processes. Traditional manual injection methods are prone to human errors, inconsistent dosages, and delays in treatment administration. To address these challenges, the IoT-SyringeX system employs advanced sensors, actuators, and cloud connectivity to enable real-time monitoring and control of injection pumps. The core components of IoT SyringeX include a smart syringe pump module, a network of interconnected sensors, and a centralized cloud-based control system. The smart syringe pump module is designed to seamlessly integrate with existing injection devices, enabling automated and controlled dispensation of medication or fluids. The network of sensors, including pressure sensors, flow sensors, and position sensors, continuously monitors key parameters such as pressure levels, flow rates, and needle positioning. This real-time data is transmitted to the cloud-based control system, which employs sophisticated algorithms and machine-learning techniques to analyze and optimize the injection process. The IoT-SyringeX system offers several advantages over conventional manual injection methods. It ensures precise and consistent dosage delivery, minimizing the risk of dosage errors. Real-time monitoring and feedback enable healthcare professionals to proactively respond to any abnormalities or deviations during the injection process. Additionally, the cloud-based control system facilitates remote monitoring and management of multiple injection pumps, enhancing scalability and enabling centralized control in healthcare facilities. In conclusion, the IoT-SyringeX project represents a significant advancement in the field of automated injection pumps. By leveraging IoT technology, it revolutionizes the way injections are administered, providing an innovative, efficient, and secure solution for healthcare professionals. The system’s ability to automate and monitor injection processes in real-time contributes to improved patient safety, precision, and overall quality of care.

The main goal of the study is to find out the changes that are occurring on the land due to the change of surface cover by its use during the period 2001 to 2020 for Vijayawada city, Andhra Pradesh. This is found by doing digital image processing using two different classifiers Artificial Neural Networks (ANN) and Mahalanobis-Based Distance (MBD)-based novel supervised classification and comparing both to find which is more accurate. For digital image processing, satellite images downloaded for image classification from the United States Geological Survey (USGS) are used as the samples. The samples are downloaded for three different years 2001, 2011, and 2020 consisting of the urban study region. Images were acquired from both Landsat 7 ETM+ and Landsat 8. Two groups of classifiers and three samples for each group totaling to six samples were used to test the accuracy. With pre-test power at 80%, alpha at 0.05 and CI at 95%, a statistical examination was done. A p value of 0.13 denotes that there is no significant difference between the groups. The percentage of broadly classified six regions are found by doing novel supervised classification by both the algorithms and noted down. The analysis is done for the key outputs overall accuracy (OA) and kappa coefficient (KC). The obtained OA is 97.10 ± 1.61 as mean and SD for ANN, 92.49 ± 6.76 as mean and standard deviation for MBD. For KC, 0.93 ± 0.05 is derived as mean and standard deviation for ANN, 0.8574 ± 0.1539 as mean and SD for MBD classification, respectively. Artificial Neural Networks are the best approach to find the land cover changes using the satellite images compared to the Mahalanobis distance-based classification from the results of the research.

Due to their distinctive electrical characteristics, such as high electron mobility and low power dissipation, carbon nanotube field-effect transistors (CNTFETs) are developing as potential replacements for conventional metal-oxide-semiconductor field-effect transistors (MOSFETs). Low drain current is one issue that CNTFETs currently struggle with, which restricts the range of applications they can be used for. Using high k dielectric materials as gate insulators, such as hafnium oxide, yttrium oxide, and lanthanum oxide, is one method of boosting the drain current in CNTFETs. These substances can lessen gate leakage current, which demonstrates an improvement in drain current. The ambient temperature of the CNTFET device can also be changed in order to optimize the drain current. In this study, different high k dielectric materials are investigated for their potential to optimize drain current in CNTFETs. Different temperatures were used to measure the drain current, and the outcomes were compared to those of conventional MOSFETs. It was discovered that whereas the drain current of MOSFETs stayed constant with temperature, the drain current of CNTFETs rose. The findings demonstrated that the drain current of CNTFETs is significantly affected by temperature and may be effectively increased by using high k materials for dielectrics. This study offers a fresh method for improving the drain current in CNTFETs and creates fresh possibilities for their useful applications. When compared to conventional MOSFETs, CNTFETs with high k dielectric material exhibit improved drain current. The temperature dependence of the drain current in CNTFETs provides an additional degree of freedom for optimization, making them a promising technology for future high-performance semiconductor devices.

This article provides a machine learning-based drug categorization research effort. The public repository Kaggle is where the dataset for this study was obtained. Age, sex, blood pressure (BP), cholesterol, and the Na-to-potassium ratio are the feature sets with the medication type as the target feature. In this work, five machine learning methods were applied: CatBoost, LightGBM, extreme gradient boosting machine, and extra tree. The findings indicated that, except for extra tree, all four algorithms had 100% accuracy, with CatBoost doing the best. The training and testing performance of the models was displayed using the learning curve. The model performance and key characteristics were understood using explicable approaches like SHAP and feature permutation significance. The findings indicated that the most critical characteristics for medication categorization are age, sex, and blood pressure. This work sheds light on how to classify drugs using machine learning. The findings demonstrate that machine learning may be used to classify drugs with high accuracy. The study's usage of explicable approaches can aid in understanding the model's performance as well as the key elements that can be employed to enhance it.

Deep learning algorithms are used in this research to propose a novel approach to intrusion detection in Internet of Things (IoT) networks. The suggested intrusion detection system employs a six-layered deep neural network architecture, which is augmented with a feature extraction module to examine network packet data and identify dangerous activities. The system was evaluated against a large dataset that comprised the following five primary attack types encountered in IoT environments: Blackhole Attacks, Opportunistic Attacks, Distributed Denial of Service (DDoS) Attacks, Wormhole Attacks, and Sinkhole Attacks. Performance evaluation metrics such as precision, recall, accuracy, specificity, and F1 Score were used to evaluate the system's effectiveness in recognizing and categorizing attacks. The data demonstrate that across various forms of assault, accuracy and recall rates vary from 93 to 96.4%. The F1 Score demonstrates balanced performance, highlighting the system's ability to eliminate false positives and false negatives. The feature extraction module considerably improved the dataset by adding essential network packet attributes such as source and destination IP addresses, session duration, and transmission rates, increasing the overall accuracy of the intrusion detection system. The utility of the proposed deep learning model in dealing with diverse attack scenarios is shown by its ability to detect and neutralize various intrusion attempts. This research extends intrusion detection methodologies by presenting a robust and intelligent solution to safeguard critical data and resources in IoT networks. Continuous research and development, on the other hand, are essential in real-world IoT deployments to handle new attack vectors and provide system adaptability to dynamic network situations.