Demystifying Emerging Trends in the Digital Horizon

A large portion of the Gross Domestic Product (GDP) of emerging nations like India comes from agriculture, which is why the sector is so important to these countries' economies. An increase in food demand has resulted from the population's rapid growth. Crop quality, yield, and profitability can all take a hit when farmers choose their crops, fertilizers, and pesticides without taking into account factors like soil type, water needs, temperature, and crop profitability analysis for a specific area. In addition to advising on fertilizers, diseases, and pesticides, researchers are using computational tools to provide crop recommendations based on soil conditions, water needs, and market profitability. This study presents AgriRec, a crop and fertilizer recommendation method based on machine learning. The machine learning algorithm takes into account information about the soil, the size of the farm, the water table, and the minimal provision price of the crop to provide seasonal crop predictions. In addition, they provide a method that takes into account the specifics of the soil, the crop, and the fertilizer to visualize a possible mix of fertilizers for a certain soil and crop combination. Using 4500 land samples from Tamil Nadu and 25 distinct crops, the system outperforms current benchmark recommendation algorithms by a factor of five while successfully recommending crops with an accuracy of 96.86% and fertilizer with an accuracy of 93.12%.

The owners of electric vehicles are confronted with the challenge of having a restricted number of charging station alternatives. A feasible solution to this issue might be the installation of individual charging stations in close proximity to residential areas. It will be easier for owners of charging stations to have a clear picture of how much energy they need to generate if they have access to a forecasting model that is capable of accurately predicting the amount of power that charging stations will use. This paper's objective is to lay forth a plan for a charging station technology that would allow electric cars to be plugged in using the Internet of Things (IoT). The study goes on to provide an ARIMA-based model that, to improve sales prediction overall, may have its learners fit to the station's subgroups using optimum parameters. With an average MAPE, RMSE, and R2 value of 13.89%, 6.68%, and 1.80%, respectively, the suggested model also forecasted the amount of electricity that would be used by seven charging stations. 

The perception system of autonomous cars and the Advanced Drive Assistance System relies heavily on pedestrian protection systems. Perceptual systems of such intelligence systems employ several sensors to get additional information about their surrounding, hence the fusion of heterogeneous sensor data is crucial. This research offered a unique technique to integrate radar and video data to identify pedestrians and get dynamic information. The paper contributes to the following steps: The first step in obtaining usable target signals from noisy original radar data is to propose a novel inter-frame track technique and an intra-frame cluster method. Secondly, a coordinate transformation matrix is obtained using the least square approach to achieve the alignment of the radar-vision data space. The next step is to provide a versatile technique for generating Regions Of Interest (ROI) using radar point projections. In addition, a faster object estimation approach is suggested to generate a more precise possible target area using ROI, which would further speed up detection. Finally, the Histogram Of Gradient (HOG) properties of the prospective region is retrieved, and a support vector machine is utilized to determine pedestrian status. Real experiments demonstrate the feasibility and efficacy of the suggested procedure.

Recent advancements in the field of face identification have highlighted the importance of gender identification in various modern applications. The detection of gender from face photographs has become a topic of considerable interest since it has several applications in security, marketing, and human-computer interaction. The proposed work examines the techniques employed in utilizing FaceNet for gender recognition in various situations, encompassing variations in backdrop, lighting, and facial expressions. In this study, we supply an inclusive test of practical verdicts, judging the efficacy of FaceNet across many circumstances. Based on the verdicts of the projected work, it may be decided that the projected work outperformed, as distinguished by additional brand-new designs for gender detection. The FaceNet model illustrated a notable level of veracity, accompanying a 94.6% happiness rate on the CMU standard database.

 In the world, one of the most common diseases that affects women is breast cancer. When it comes to killing women, it comes in second place among all cancer kinds. Therefore, early identification is essential to lowering the fatality rate. Women's survival rates rise as a result of early detection and treatment of breast cancer. Mammography is a common radiological screening method that is used to examine women for breast cancer, even if there are no symptoms. A few machine learning techniques, like the support vector machine, are used in the literature for the identification and categorization of breast cancer. SVM, however, generated the results with incorrect categorization. As a result, the research work uses a Statistical neural net (PNN) classification mechanism that is based on advanced deep learning to solve this challenge. For effective breast cancer region detection, a segmentation method based on k-means clustering is initially applied. Last but not least, PNN was created for the classification of breast cancer with the grey level co-occurrence matrix (GLCM) based Texture features, Discrete Wavelet Transforms (DWT) based low-level features, and statistical colour features, respectively, to achieve the maximum efficiency of the system. To classify benign and malignant breast tumours, the research work can be efficiently applied. When compared to state-of-the-art methods, the suggested strategy performs better in the simulations, both quantitatively and qualitatively.

This research focuses on exploring the process of capillary retention of water in an unsaturated porous medium. To find this, we may use the non-linear parameter in the Richards equation, which is hydraulic diffusivity. When hydraulic conductivity is divided by a specific water volume, the resulting quantity is the hydraulic diffusivity. Moisture transfer and liquid transport are two forms of hydraulic diffusivity. This equation may have a linear or non-linear form, depending on the type of function. Hydraulic dispersion and non-linear moisture content equations derived from experimental data have often made use of this approach. The project incorporates ten distinct construction materials sourced from a variety of published sources. Researchers assess and analyse experimental data on how temperature influences various construction materials. The findings showed that the constant diffusivity increases with increasing temperature for all ten materials. However, due to material inhomogeneity or experimental errors, the results demonstrated that the diffusivity starts to decrease for a few of these substances, which indicates a limitation of the current model.

The human face serves as the most distinctive identifier of an individual, making facial features a viable biometric for a face recognition system. Managing attendance has often proven to be a challenging task. Conventional attendance tracking methods involve manually calling out students and meticulously recording their presence or absence, demanding substantial time and effort. This work introduces an OpenCV-based facial recognition technique comprising a camera for image capture, a face detection algorithm, facial feature encoding, identification, and attendance management within a spreadsheet. To establish a training database, the system is trained with the images of authorized students' faces. This research identifies people using Convolutional Neural Networks (CNNs). Compared to other methods, the Local Binary Patterns Histogram (LBPH) technique is accurate.

Modern customers across the globe are finding that ever-improving information technology makes their lives easier; as a result, it is crucial to apply this knowledge to the irrigation industry now. Worldwide population growth is driving up food consumption at an exponential rate, making it imperative to increase arable land. Producing food through irrigation is the most dependable method in light of the current climate crisis caused by industrialised nations' contributions to global warming. With more and more areas available for irrigation, the best management and efficient use of water are becoming more important since it is the only resource necessary for agricultural production. It offers a simple method of automated irrigation and is based on the Arduino platform. With the help of the GSM module, this project intends to build and install an automated system for controlling plant irrigation using Arduino and the module, which will allow the user to get updates on the farm's status. Two primary components make up this suggested system: hardware and software. The hardware unit, which consists of mechanical components, includes systems for instrumentation and watering with irrigation. A microprocessor, flow metre, humidity sensor, LCD, and a GSM module form the basis of the equipment system. The C++ code that makes up the programme allows for the incorporation of interdependent modules. In this setup, the microcontroller acts as the central nervous system, coordinating the operation of all the other modules; it also runs the watering system, keeps everything in sync, and sends alerts to the user via GSM about the state of the field and irrigation area. Incorporating a timer-based device for automatic watering can greatly improve efficiency, and the project's implementation will greatly contribute to water savings of 40–60% compared to conventional watering methods like sprinklers. It will also improve the growth and discourage weeds by ensuring that water is only served to the areas that need it.

This project aims to create a low-cost, automated hand dispenser. These dispensers employ infrared sensors to identify a hand without touching it. The microcontroller receives proximity signals from this system's infrared sensor. An LCD panel displays heart rate and temperatures collected from this project. This expertise helps detect viral symptoms early, especially COVID-19-related ones, allowing for rapid care.

The study introduces a compact, multiband antenna featuring a stepped Vshaped design integrated through an Artificial Magnetic Conductor (AMC). Constructed on a 25 mm × 25 mm FR4 substrate, the antenna supports multiple frequency bands, operating effectively between 3.4–5.86 GHz, 10.28–11.86 GHz & 13.69–14.28 GHz, with resonant points at 4.6, 11, and 13.9 GHz. Comprehensive analysis highlights favorable return loss, radiation patterns, and gain characteristics. Notable features of this design include simplicity, reliable multi-band operation, enhanced gain, and suitability for cost-effective wireless communication solutions.

Wireless technology has grown quickly in recent years, and vast amounts of information have been transmitted via numerous communication routes. However, a select few apps, including those used in the military, healthcare, multimedia, web, and civil applications, among others, must offer information-transmitting security. Therefore, secure transmission is a very difficult task. Here, utilizing video steganography and pixel mapping, we present a novel secure text picture transmission strategy. Pixel mapping is a relatively straightforward technique. After the video is scattered over the picture frames on the suggested page using a Matlab programme, the hidden message will remain in the video sequence. Once every frame has been saved sequentially. The original encrypted video file is created by first preserving all of the text in the photographs, then arranging the images sequentially, and then cascading all of the frames. 

The Internet ushered in the information revolution and facilitated the exchange of information among people all over the world. It is a remarkable innovation in human history that has become available to many people worldwide at very little cost. It has become a major tool for entertainment, conversations, and knowledge sharing. Educational institutions have embraced the Internet to promote and share knowledge. Teachers have created online tutorials, tests, and reading materials for students to improve their knowledge and skills. Textbook writers have begun adding web exercises; web seminars have gained prominence. Given these developments, the role of the Internet in education has become significant. All powerful tools are subject to some misuse. Much like how the use of educational television was questioned, students' use of the Internet was questioned. A spate of articles came into the media discussing the uses and misuse of the internet. The present study seeks to inquire about how university-level students use the Internet.

Artificial Intelligence (AI) has great potential as a method for advancing health innovation. Artificial intelligence in picture identification greatly increases findings beyond human perception. Medical image analysis using Artificial Intelligence (AI) aids automated diagnosis. AI is making diagnostic radiology more scientific and less subjective. Recognition of things in medical photos is a significant AI technology in medicine. Early brain tumour identification has improved using CNNs and deep learning systems. Unfortunately, these algorithms need a big-picture library for training. Time, computer resources, and expensive equipment are needed for this step. This research describes a standard automatic segmentation method for early brain cancer diagnosis using MRI data. The Harmony Search Algorithm (HSO) was modified to segment MRI brains using a multilayer thresholding approach, and parameter selection was optimised. Variance and entropy thresholds divide the histogram into several parts with distinct colours. Segmentation is followed by morphological procedures and linked component analysis to remove noise and find brain tumours. Precision, Dice Coefficient, and Jaccard’s index measure brain tumour detection ability. Professionals in the field compare the findings to those obtained manually. The BraTS challenge used the Brain Images dataset to compare the results against several CNN and DLA methods. To compare performance, they utilized the average Dice Index. In comparison to CNN and DLA approaches, the suggested strategy produced results that were comparable in their accuracy while also being much superior in terms of computing complexity, data management, and execution time. 

The city's sensors collect large amounts of spatial-temporal data that shed light on the daily routines. To help with identifying suspects, data analysts may examine and categorize the trajectories of specific targets, considering that each person's lifestyle is represented by their movement data. In practical applications, however, it still faces several obstacles; for example, trajectories are too varied, and transit behaviour traits are difficult to specify due to their breadth and expense. In addition, data-sensitive elements could lead to erroneous results if people's lifestyle patterns vary throughout the city. The unique traits of movement behaviours concerning trajectory aspects are discussed in this. The following text should be remembered: “Suggest a neural network-based model and a Trajectory Pattern Model (TPM) to enhance identification and performance. Instead of relying solely on location and timestamps to characterize user behaviour, the Trajectory Pattern Model (TPM) offers a new perspective by creating more useful and universal characteristics.” The goal of the model that is built on end-to-end neural networks is not to use human traits. The behaviour of a specific target may be better understood with the use of statistical analysis and intelligent explanations. Extensive examination has shown that the recommended solutions are more successful than peer options.

Seizures brought on by irregularities in brain activity might pose significant risks to a patient's health. It is useful to know when epileptic seizures are going to happen so that medication can be taken at the right time. Neural network analysis and computational methods are used in epileptic seizure prediction using electroencephalogram (EEG) data. The two most important problems with preprocessing EEG signals for noise reduction and feature extraction are the expectation time and the true positive rate for prediction. This is why a model is provided with reliable methods for feature extraction and preprocessing. With a greater true positive rate, the model can anticipate epileptic episodes with enough lead time. To train a prediction model, they have used time and frequency domain information collected using Empirical Mode Decomposition (EMD) as preprocessing. The suggested method accurately identifies the beginning of the preictal state, a period just before a seizure begins, on the EEG CHB-MIT scalp data sets of 23 subjects with a true positive rate of 93.24%, a maximum expectation time of 34 minutes, and a normal time for prediction of 24.7 minutes. In contrast to traditional methods, this model performs better.

Guarding The Gates: A Secure System for Cloud Data Sharing” suggests a healthy foundation designed to guarantee the data in cloud surroundings. The system leverages progressive encryption methods, approach control methods, and real-time attention to transparency and secrecy. It aims to address key security challenges, including unlawful approaches, data breaches, and compliance with solitude managing, providing a reliable answer for businesses and things to share data in the cloud. To ensure optimal data safety by reducing risks, the study delves into the specifics of data protection tactics and methods utilised throughout the world. While cloud storage has several usages, it also exposes information to apps that might have security flaws that might compromise it. Similarly, data might be in danger when using virtualization for cloud computing and running a virtual operating system on a hypervisor without understanding the reliability of the guest OS or if it has a security flaw. Aspects of data security to data-in-transit as well as data-at-rest, are also discussed in this chapter. Platform as a service, Software as a service, and infrastructure as a service are the 3 pillars around which the research rests.

 Social media strongly influences today's youth, especially their purchase decisions. This study examines how social media ads affect young people's purchases. This research examines how social media advertising affects young customers' buying patterns by evaluating the literature and conducting empirical analysis. The method involves asking a representative sample of young people about their social media advertising experiences and purchase patterns. The key findings show that social media ads influence young consumers' attitudes, likes, and purchasing behaviour. Ad content, platform involvement, and peer influence strongly affect consumer behaviour. Social media marketers targeting youth must understand these characteristics to succeed. Marketers might benefit from the study's results and suggestions for additional research on social media advertising's impact on young customers' purchase choices.

In cloud computing, information is becoming increasingly valuable; hence, it must be encrypted as soon as possible to prevent unauthorized individuals from gaining access to it. A fresh approach to key generation is presented here as a potential answer to the problem of client-side cloud computing. This solution comes in the form of a research study. This approach combines the processes of DNA encryption and decryption with the generation of ensemble keys. It was created from scratch just for the sake of this study. Following the initial DNA encryption, the method next develops an asymmetric key based on the principles of RSA and ElGamal. This step is followed by the last step of decrypting the data. At long last, the data have been deciphered. The approach that has been proposed can be applied to carry out an analysis of data linked to the traditional investigation. In this particular part of the study, in addition to focusing on key sensitivity and statistical analysis, we are going to be paying close attention to key space. The findings of the study led the researchers to the conclusion that the proposed method is robust enough to withstand an exhaustive attack and is appropriate for deployment in real-world scenarios. This conclusion was reached as a result of the research. 

Promotion, demotion, transfers, and incentive policies are all forms of internal mobility. An employee is said to be “upgraded” when their income is raised but their work grade remains the same, as opposed to being promoted. After consulting with the unions, you should get their stamp of approval on the promotion policy. Before being promoted, a thorough evaluation of one's work and performance should be conducted. A job description is useful for understanding the duties and responsibilities of the position, and a performance review may tell management whether the worker is up to snuff. Openings for promotions should be posted in several locations so interested individuals may apply, and awards should be offered to qualified applicants determined by their work level when promotions and prizes are made based on competence.

Deep Learning (DL) and Machine Learning (ML) techniques have shown promise in medical research, where globally, cardiac disease is a significant cause of fatality. To identify cardiac issues, deep learning and ML approaches are used in this study. The method makes use of a large dataset that includes clinical, demographic, and imaging features. To extract high-level representations, a variety of deep learning architectures, including CNNs, SVMs, and RNNs, are used. Utilizing stratified crossvalidation techniques, the models are trained, validated, and tested. Identification of risk factors and potential biomarkers is aided by interpretability analysis. Deep learning models outperform conventional machine learning algorithms in accuracy and AUCROC, according to preliminary results, which indicate promising performance. Additional study is required to validate and improve the strategy, ultimately improving clinical outcomes and lessening the burden on the healthcare system. 

The research examines the implementation of an IoT AI-driven Predictive Maintenance (PdM) system to be used in manufacturing premises to forecast and prevent equipment failure. This system is equipped with advanced IoT sensors that monitor critical parameters in real time, including temperature, vibration as well as pressure. All data are further analysed with edge computing and cloud computing for additional and detailed analytics and storage. Therefore, the methodology explains the system’s architecture in detail, including thorough data preprocessing and AI model elaboration. Six months of experimental validation and a couple of cases of motors and pumps substantiate the expected outcomes of the system’s functioning. Eventually, the neural network model was the most efficient, with an accuracy of 92% along with an F1-score of about 89% as compared to the remaining models, whose accuracy was barely above 80%. The major findings prove there is a possibility to forecast the failure to schedule a time for repair and hence reduce the unplanned downtime and repair costs. Thus, the motor and pump cases demonstrate that the repairs of $10,000 and $8,000 were indeed avoided, respectively. However, it is also necessary to acknowledge certain issues, including data integration and the necessity for highly qualified personnel. Thus, the study recommends possible directions for further development of AI models and expansion of potential perspectives for use across various industrial equipment. Thus, the research provides an opportunity to employ modern IoT and AI perspectives in enhancing the efficiency of maintaining equipment in a manufacturing plant.

More dispassionate research abilities enforce the protection of human means. This study provides an in-depth examination of blockchain technology, renowned for its appeal, transparency, and decentralization. It explores the experiential aspects of blockchain, consent mechanisms, the fundamental components of signaling protocols, and various application types. Additionally, the research investigates the facial features of machine intelligence techniques employed in brain tumor detection. Utilizing a comprehensive brain tumor image dataset, the study also addresses security considerations, including risk assessment and mitigation strategies. The conclusion highlights current challenges and proposes future research directions aimed at improving the decentralization and scalability of blockchain frameworks.

Learning Analytics (LA) is pioneering educational innovation via data collection and analysis. This innovative technique digs deep into learners' backgrounds to improve performance. Dynamic approaches are used to experiment across many aspects, including varied learners and motivated instructors. Through its meticulous examination, LA unlocks the true potential of education, fostering transformative advancements. Machine Learning in Learning Analytics may improve learning by providing visualisation approaches that reinforce complicated ideas and enhance engagement. This extensive article analyses and assesses a variety of visualisation techniques to improve education and establish desirable values. By investigating the potential of these technologies, educators and researchers may use data-driven insights to personalise and alter educational experiences.

 Humans have a greater tendency to interpret information via visual images rather than written language. An image can convey many searching ideas and emotions that would need a significant amount of written language to express. Within our realtime applications, visual imagery is more advantageous than written information. Image retrieval systems refer to computer systems designed for searching, browsing, and retrieving photos from huge digital databases. Most image retrieval systems primarily rely on information such as subtitles, keywords, or descriptions to search for photographs using annotation terms. This research proposes a Pattern-Based Picture Retrieval (PBIR) system that utilises multi-domain picture databases. The image retrieval system now incorporates object identification and segmentation to enhance its performance. The patterns are retrieved and then analysed at three levels to enhance the scheme's presentation searching. Ultimately, an evaluation is conducted to assess the effectiveness of multi-domain pictures in enhancing comprehension.

Researchers have explored the possibility of deep learning techniques for image denoising, revealing important differences among these approaches. Discriminative learning, particularly effective against Gaussian noise, and deep learning-based optimisation enable real-world noise estimation. However, the diversity of methods has hindered research comparing deep learning in image denoising. This study evaluates multiple deep image denoising approaches, categorising images into four collections by the type of Convolutional Neural Network (CNN) used: incremental white noise-trained CNNs, true noise-trained CNNs, blind wavelet altered CNNs trained on compound noisy images. It examines shared goals and principles, conducting qualitative and quantitative assessments on publicly accessible denoised datasets to compare state-of-the-art techniques. The study concludes with identified challenges and future investigation instructions. 

This study explores entrepreneurial opportunities in industrial innovation through the lens of Human-Machine Interfaces (HMIs). The objective is to understand the dynamics of HMI integration, enhance user experiences, and offer strategic insights for entrepreneurs. The study identifies gaps in the existing literature by emphasising the underexplored entrepreneurial dimensions of HMI utilisation in industrial settings. A comprehensive examination is conducted through qualitative discussions and document analysis, based on the Technology Acceptance Model (TAM). Analyses of entrepreneurial ventures, user experience effect, and HMI adoption across sectors show trends. We recommend user-centric design, iteration, and human factors expert cooperation. The policy consequences encourage research, standardisation, and education. The study aligns with TAM and effective reasoning, fills a significant research gap, and offers practical insights for business professionals, policymakers, and future researchers. This study illuminates the relationship between technology, business ownership, and customer service, enabling further research on this dynamic subject.

Pharmaceutical scientists' trial-and-error approaches dominate formulation development. This method is difficult and expensive. Automatic feature extraction has made deep learning popular in many fields. Deep learning is used to forecast pharmaceutical designs in this study. Two dose forms were used as model systems in this work. The models were evaluated using criteria specific to pharmacokinetics. To pick typical data for test datasets and validation, an automated system was created. Deep learning is related to six machine learning approaches. Both deep neural networks have reported accuracies of 80%, greater than previous machine learning models, and predicted pharmaceutical formulas well. First-time pharmaceutical formulation prediction was achieved using deep learning, an autonomous data splitting technique, and pharmaceutical formulation data assessment criteria. Pharmaceutics and AI may reshape pharmaceutical research from experience-based to data-driven.

A significant obstacle stands in the way of resolving concerns over privacy and information security in the cloud. This study suggests that a keyword-based system could be useful for authorizing granular searches by cloud owners. The resulting technology is a boon to data owners who have lost control of their data in the cloud; it enables them to set and enforce user access policies by Employing cutting-edge cryptography techniques, such as attribute-based encryption (ABE) and searchable encryption. Data security, query confidentiality, and policy enforcement are significant challenges that can be addressed through efficient and secure keyword-based searches. The proposed technique appears to be practical and extensible, according to our testing results, which makes it a good fit for usage in significantly large-scale cloud structures. A robust structure is established to enhance user control and data security in light of these findings.

 In an Indian context, this research seeks to quantify the influence of happiness, mood, leisure, and distinctiveness on the trade customer experience and, by extension, customer satisfaction. In this varied and ever-changing environment, businesses can develop more effective strategies to enhance the overall shopping experience and foster consumer loyalty by understanding these dynamics. The 431 respondents' data were gathered for the current study using a survey-based methodology. The conceptual framework was tested using the bootstrapping approach and the PLS algorithm. With the exception of pleasure, the results show that mood, leisure, and distinctiveness are strong predictors of retail consumer experience. Additionally, in Indian contexts, the retail customer experience has an important effect on consumer happiness. The findings discussed in this article will assist retail shop managers in communicating their approach to improving customer satisfaction and the retail customer experience. This study examines pleasure, mood, leisure, distinctiveness, retail client satisfaction, and customer experience in Indian contexts for the first time. In addition, by including a variable, namely customer happiness, the current research expanded the framework of previous works.

The E-recruitment sector has evolved due to technological improvements and the shifting expectations of the Indian generation. This research examines the impact of current developments in e-recruitment methods on the next generation. This report reviews literature and empirical research to examine the critical elements that affect the use of e-recruitment platforms by young job seekers and employers. Mobile platforms are becoming more important in e-recruitment, according to the report. The research examines how social media platforms affect the recruiting process. The research combines primary and secondary data. To enhance recruiting, the study proposes exploring the efficiency of various tactics. According to the report, organizations should prioritize authenticity, diversity, and professional development opportunities to meet the expectations of the young generation in recruiting. 

There are a substantial number of security breaches that are revealed each year in the electronic banking (e-banking) system. These breaches bring to light the need to protect consumers and inform them about the dangers of being exposed to harmful acts that are implemented by cybercriminals. Attack and financial scams are growing more complicated and are being committed by a distinct category of criminals, which is something that both customers and financial institutions are aware of. Utilising technology as a component of their strategy, this class is becoming increasingly complex. In addition, a rise in the number of internal fraud and security breaches is anticipated by professionals as a consequence of the current global recession. First, this study aims to examine the literature on the subject of possible threats to the security of online banking services. Second, it seeks to identify methods and tools that can make online banking safer for consumers. Third, it presents the findings of a pilot investigation that looked at how Romanian consumers felt about online banking security. 

This research examines the human resource (HR) implementation problems faced by small and medium enterprises (SMEs) in Tirunelveli City. SMEs struggle owing to limited funds, a small staff, and unique operating characteristics. This study will look at how these companies' hiring, training, and advancement practices are perceived by their employees. Primary data is provided by 600 workers from various industrial clusters using a descriptive study style. The use of descriptive statistics, oneway analysis of variance (ANOVA), and the Kruskal-Wallis test is prevalent. Male participation outnumbers female. Most responders are married and aged 31-40. Bachelor's degrees and certificates are the most prevalent forms of education. The analysis includes both technical and non-technical occupational cadres to achieve balance. Employees usually like recruiting and selection, especially formal interviews and recruitment and selection strategies. Individual job demands and abilities are met via learning and development opportunities. Promotion policies should be better.

A study on smart manufacturing management for cybersecurity is presented herein to fill the research gap regarding the complexities in cybersecurity of smart manufacturing systems. This paper seeks to investigate components of security culture, examine the implications of organisational structure, and suggest useful methods for workforce training and safety awareness programs. The research highlights an area in the literature that is underserved, especially regarding how these types of models are used to design SMTP curriculum. The research shows that effective security practices include security leadership, integration of security into job roles, and involvement across work areas. Security culture is largely shaped by organisational structure, among other factors, where flatness allows more flexibility within its hierarchy. As established, the gap exists only on paper, as adopting the social learning theory can be efficient when implementing training programs. Given these challenges, it emphasises the importance of contextually tailored communication and multi-channel approaches in awareness programs. These recommendations can have approaches aimed at creating a positive commitment within leaders, integrating job roles, and improving collaborative efforts across functions.

The phrase “Internet of Things” (IoT) describes a system that allows billions of linked devices to exchange data and coordinate their movements across a given area with little human involvement. While the sector of computing known as the IoT is growing at a fast pace, it is also vulnerable to a broad range of attacks because of the internet's infamously hostile climate. To solve this problem, practical countermeasures must be used to protect IoT networks, such as network anomaly detection. The ability to detect threats early is crucial for successful protection, even when complete prevention may be impossible. Due to the limited processing and storage capabilities of IoT devices, traditional, high-end security techniques cannot be used to safeguard an IoT system. Internet of Things devices can now remain online for longer periods without any type of human intervention. It is essential to create smart network-based security solutions using machine learning. Recent research has focused on using machine learning (ML) techniques for attack detection; however, limited attention has been given to detecting attacks specifically within IoT networks. This paper evaluates ML techniques for detecting IoT network attacks rapidly and efficiently, contributing to ongoing research. Bot-IoT is a new dataset-testing detection technique. Seven machine learning algorithms wereimplemented, with most achieving good performance. The Bot-IoT dataset yielded superior results, with additional characteristics retrieved during implementation compared to previous research. 

The field of digital marketing has recently emerged as a hot topic among academics in the sciences, the arts, and LIS. Papers published between 1992 and 2022 were searched for by the researchers using the Web of Science database. They discovered 1006 Indian publications. Both Biblioshiny and Vosviewer were used to examine the extracted data. This chapter examines the operation of co-authorship network analysis and its potential applications in the medical field.