Osteoporosis is a chronic bone disease that tends to make the bone fragile enough to break it, hence its early diagnosis would prove to be helpful for both the patients and the medical practitioners to personalize the treatment. Numerous works have already been done involving deep learning algorithms and pre-trained models to predict osteoporosis via the X-ray datasets of knee etc. while very few research works have explored the diagnosis of osteoporosis using the Dental Periapical Radiograph dataset which is the focus of this work. In this paper, a customized deep learning model with various layers and parameters has been implemented on the Dental Radiograph dataset having three classes— Osteoporosis, Osteopenia, and Normal. Apart from the custom model, transfer learning has also been implemented on the pre-trained models like DenseNet121, DenseNet201, VGG19, and MobileNetV2 for the overall comparison and analysis. The highest test accuracy for the custom models is 95.17% and the highest AUC is 99.59% whereas the highest test accuracy among all the fine-tuned pre-trained models mentioned above is 94.57%, and the highest AUC is 97.68%. It is evident that the custom model’s accuracy and AUC have surpassed the fine-tuned model’s performance, making it seem suitable for deployment in real-world settings and developing other applications.

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition that impacts individuals of all ages. This disorder comprises of symptoms ranging from social and communicational challenges to an exhibition of repetitive behaviors. Early diagnosis and prognosis of ASD is focal for managing its symptoms, supplementing learning outcomes, and social skill development. Conventional diagnosis of ASD based on behavioral evaluations confronts problems owing to the lack of clear-cut rules. On the other hand, Machine Learning and Deep Learning models-based techniques excel in diagnosing ASD by assessing behavioral factors, finding even the tiniest contrariety in gaze and facial attributes. This paper aims to propose an AI-based Autism Detection System leveraging two distinct models. For children aged 2 to 8, XceptionNet model is employed to assess facial dysmorphology in Autistic children's images with an accuracy of 85%, while individuals aged 9 and above undergo evaluation by Light Gradient Boosting Machine Classifier which considers diverse indicators, such as the ASD score, age, gender, ethnicity, etc. to detect Autism with 99% accuracy. Thus, the proposed models enhance the accuracy of ASD detection by addressing the distinct characteristics and requirements of people of all age groups with Autism.

Vehicular Ad hoc Networks (VANETs) are highly mobile wireless networks that play a crucial role in public safety dispatches and commercial operations. Recent advancements in VANETs have enabled the integration of data generated from these networks into smart operations, providing quality of life services. Data mining is a process that involves extracting valuable patterns and information from data. One promising area of research involves applying data mining techniques to VANETs to extract useful patterns. This paper presents an overview of basic data mining techniques, including pre-processing, outlier detection, clustering, and data ordering. Additionally, this paper describes the most commonly used classification and clustering techniques, comparing them based on their strengths and weaknesses.

In the current study, we sought to construct an integrated model to identify various elements and evaluate the impact of these identified factors on customers’ behavioral intention to use or not use specific M-wallets for payment. To this end, we proposed and validated a conceptual model. In all, 600 questionnaires were distributed, and 482 responses were deemed usable. Structural equation modeling was used to demonstrate the stability of the proposed model and to test the research hypotheses. Perceived value, trust, compatibility, and social influence were all found to have a substantial influence on behavioral intention; however, consumers are less likely to use an M-wallet on the basis of perceived enjoyment. We also found that trust, followed by compatibility, has a stronger influence on customers’ behavioral intentions in the context of M-payments. This study only included six M-wallets and was restricted to a certain age group in a single city. Understanding the many characteristics of behavioral intention can help M-wallet providers gain consumer trust and increase the frequency with which consumers use M-wallets for M-payments. The findings suggest that M-wallet service providers should consider and manage all influencing elements as proactive strategies for M-wallet intention. This strategy can be used to create an M-wallet-user behavioral intention model that will assist enterprises/companies in managing the establishment of their users’ behavioral intentions.

Synthetic Identity Threats (SIT) present one of the greatest risks that could undermine the integrity and security of digital systems. These threats adapt Deepfake Technology through generation of new faces, modification of facial attributes or reenactment of fake emotions on human faces. This study is a full documentation of how Deepfake Detection mechanisms can be modelled to neutralize these SIT's using various Deep Learning architectures. We also explored applications of Deepfake beyond malicious intent into forms that are less nefarious like entertainment purposes. We utilized Generative Adversarial Networks to create Deepfakes and experimented with conducting face swaps between two distinct photos as well to assess the quality of Deepfake Technology. For detection of Deepfakes, Convolution Neural Network has shown the highest accuracy of 89.36%, Inception ResNet attained an accuracy of 82.978% while Visual Geometry Group and EfficientNet were at a score of 82.8% and 83% respectively. It is evident that although the current detecting methods are competent in their abilities, the SIT environment continues to develop. Through analyzing the nuances of producing Deepfakes and comparing current quality detection methods, we hope to offer useful information on how scholars and administrators can better protect the internet from deceitful attacks of Deepfakes.

This study looks at how religion affects what Indian people buy. It gives a plan and ideas for more research in this area. The research looks at lots of different books and papers to understand how religion influences what people buy. It studies things like what Indian people believe, the things they do for their religion, what they value, how they are part of a community, and how all these things affect what they buy. The plan also talks about people who don't follow any religion and the problems and good things about doing research on religion and buying stuff. It also talks about things that could make it hard to do this kind of research in the future. The study uses many different books and papers to get a good understanding of this topic including Aaker, Fournier, Brasel (2004), Ahmad, Rustam, Dent (2011), Allport (1950), Allport, Ross (1967).

Vision is one of the crucial senses and is the birthright of every human being. Its impairment or loss leads to various difficulties. Blind and Visually Impaired (BVI) people find it tough to maneuver outdoors daily. Even though the market is laden with countless aids for BVI people, there is still a lot to be achieved. The idea of every new research in the market is to assist these individuals in any possible way. Individuals deprived of vision require numerous reliable methods to overcome these barriers. Also, with the advent of science and technology, there is nothing that a human being can’t do. Researchers and manufacturers are coming up with new inventions and tech gadgets now and then. In this paper, Convolutional Neural Network (CNN) models, vgg16 and vgg19, are used along with the self-created dataset involving two classes: roads and crosswalks, which underwent the ML procedures resulting in the accurate detection of the respective classes.

Story Generation through Deep Learning is a fascinating area of research in Artificial Intelligence that aims to create computer systems that can produce original and compelling narratives and is an interesting concept that has flourished in the domain of Machine Learning applications starting from 2018. Most of the research carried out in this specific area has shown advances in Modelling and efficiency of story generation. However, some of the setbacks in Artificial Story Generation include little to no coherency with human generating pattern, tokens/words limitation, missing plot twists and direction of story. In this paper, we have performed a comparative study on Automatic Story Generation as well as the proposed scheme of this paper has main focus on generating a meaningful story with the help of conditional text generation using keywords upto five hundred words by optimizing hugging face generative pre trained model Version Two catering towards the problem of coherency in the text generated. As a result each sentence is semantically coherent and the first three sentences are indeed related to the title itself. The experimental results show a BLEU score of 0.704 averaging over ten genres.

In order to make it possible for computer systems to provide a dynamic defensive layer by modifying the attack surface, a security solution that goes by the name Moving Target Defensive (MTD) is now in the process of being developed. This is one of the many security solutions that are currently in the process of being developed. The use of MTD practises is one way to alleviate the concerns of cloud computing's lack of data protection. Shuffle, diversity, and redundancy are the three primary categories that make up MTD method usage, respectively. It may be possible to reorganize the components of the system by making use of randomized MTD methods (for example, IP mutation). They put a dent in the case that the attackers are making by first making reconnaissance more difficult and then destroying the material that was acquired. As a consequence of the emphasis on diversity in Modularity, Transparency, and Reliability (MTR) methods, the versions of a system's components, such as operating systems, are modified. This makes attacks more labour- and resource-intensive to carry out. The redundancy approaches provided by MTD lead to the creation of several redundant copies of system components. The first phase of this inquiry will consist of conducting a comprehensive analysis of the relevant prior work in order to identify and emphasize the most significant omissions in the current MTD study. According to the findings of our research, MTD strategies have not been thoroughly confirmed on more realistic cloud testbeds, and their efficacy has not been adequately evaluated utilizing security analysis.

With the advancement of computer technology, multiple advanced techniques rely on machine vision, especially biometric systems, to play an essential part. The recording shows an image or video data with a face in it, then recognizes and analyzes the face region. Detection and recognition consist of a group of Artificial Intelligence (AI) techniques. It has a broad scope of uses and has become a thriving topic of study. In our proposed model, we employed AveragePooling2D with MobileNetV3Large. The algorithm is trained on the collected dataset following a few preprocessing processes. According to the findings, MobileNetV3 with AveragePooling2D has an accuracy rate of 91.12 percent on the training dataset and 99.79 percent on the verification dataset for 50 epochs.

In the generation of modern and advanced technologies, an immense proportion of information is accessible. Big Data is a huddle of huge quantities of data which keeps on growing exponentially with time. Because of the expeditious widening of day to day information and data, resolutions have to be investigated and given with the aim of managing and bringing out important valuable insights from the information databases. Moreover, it is mandatory for business strategists to obtain knowledge from huge swiftly changing data. Big data analytics are acquiring huge dominance in all the domains of business development and management. Further research illustrates that logistic networks and business activities are one of the most gigantic sources of information in the company. Hence, their business strategy building methodology would be beneficial from accumulated utilization of business data analytics technologies. However, there is still a deficiency of recognizing what influences the capability of a company to build business data analytics technologies to gain competitive insights from it. In this study, we focus on drawing helpful insights from the data stored in the company's database. In today’s scenario where we have a large database with n number of tables having millions of rows, it becomes impossible for a human to explore the information and bring out perceptions from it. Tools like MS Excel fail in analyzing such a large amount of data. As a solution to it, we as a BI Developer made a simplified tool in Power BI which acts as a coupling between the data warehouse and employees. Using that tool, the employees can pull out inner sights from the large database within seconds.

Osteosarcoma is one of the most common malignant bone tumors mostly found in children and teenagers. Manual detection of osteosarcoma requires expertise and it is a labour-intensive process. If detected on time, the mortality rate can be reduced. With the advent of new technologies, automatic detection systems are used to analyse and classify medical images, which reduces the dependency on experts and leads to faster processing. In this paper, an automatic detection system: Integrated Features-Feature Selection Model for Classification (IF-FSM-C) to detect osteosarcoma from the high-resolution whole slide images (WSIs) is proposed. The novelty of the proposed approach is the use of integrated features obtained by fusion of features extracted using traditional handcrafted (HC) feature extraction techniques and deep learning models (DLMs) namely EfficientNet-B0 and Xception. To further improve the performance of the proposed system, feature selection (FS) is performed. Here, two binary variants of recently proposed Arithmetic Optimization Algorithm (AOA) known as BAOA-S and BAOA-V are proposed to perform FS. The selected features are given to a classifier that classifies the WSIs into Viable tumor (VT), Non-viable tumor (NVT) and non-tumor (NT). Experiments are performed to compare the performance of proposed IF-FSM-C to the classifiers which use HC or deep learning features alone as well as state-of-the-art methods for osteosarcoma detection. The best overall accuracy of 96.08% is obtained when integrated features extracted using HC techniques and Xception are used. The overall accuracy is enhanced to 99.54% after applying BAOA-S for FS. Further, the application of BAOA-S for FS reduces the number of features with the best model having only 188 features compared to 2118 features if no FS is applied.

The broad and regular use of pesticides is justified by the tremendous economic relevance of plant diseases that harm field crops. In some regions, plant diseases may make it impossible to cultivate or grow food plants; alternatively, plant diseases may make it feasible but attack the plants, kill some or all of them, and reduce much of their yield, or food, before it can be harvested or consumed. Early detection of plant diseases is advocated since it is essential to ensuring that the entire population has access to food. It is bad to anticipate illnesses when crops are young, nevertheless. Agriculture is a major source of income for more than 80% of individuals. The work of identifying leaf diseases is crucial since the crop's overall production is decreased by a variety of illnesses. It can be challenging for farmers to diagnose a specific disease. We looked at different sorts of vegetable and fruit leaf diseases and how to spot them. Hence, Image Processing and Deep Learning are the two main methods for detecting leaf disease. The preferred strategy now emphasizes the use of deep learning ideas, particularly Convolutional Neural Networks.

The significant health impact of lung diseases hampers the life of an individual and his/her family. It is crucial to ensure that everyone lives a healthy life, hence early detection of lung diseases is encouraged at an early stage. As several lung illnesses reduce the life span of people, they are not able to live a healthy life. There are errors in many detection algorithms, so a better algorithm is required to detect such diseases. In this paper, we have discussed lung diseases and how to recognize them. The two primary techniques for identifying lung illness are therefore image processing and deep learning. Deep learning is increasingly emphasized as the preferable method with convolutional neural networks. We further discussed various machine learning algorithms and compared their results with the newly designed algorithm of a convolutional neural network with an autoencoder. There are several approaches described in the literature for classifying medical images. This paper aims to develop a useful tool that will assist medical practitioners in quickly determining if a patient has a lung disease or is at risk of contracting one; by analyzing lung images and examining disease development risk factors with the use of an autoencoder.

Purpose: The purpose of the study is to highlight the need for research on the relationship between consumer value, risk, and trust in the context of social cross-platform perceptions, specifically in India. The study aims to provide insights into how these factors influence consumers' cross-platform buying behavior, ultimately contributing to a better understanding of online shopping dynamics and informing strategies for building trust and enhancing consumer experiences in the online shopping environment. Methodology: The research methodology involved conducting a survey among 320 online shoppers in India. The objective was to investigate the relationship between social cross-platform perceptions of consumer value, risk, and trust in the context of online shopping. The researchers developed a questionnaire to collect data on perceived value, risk propensity, trust, and cross-platform buying behavior. Partial least squares (PLS) path modeling were used to analyze the data, revealing significant effects of perceived value, risk propensity, and trust on consumers' cross-platform buying behavior. Finding: This research focuses on the rise of social cross-platform buying behaviour in India and examines the influence of perceived risk, perceived value, trust, and marked negative reporting on consumer behaviour. The study found that consumers' perceptions of perceived risk and perceived value significantly impact their likelihood of using social e-commerce platforms like Meesho. Perceived risk refers to the level of uncertainty or potential harm associated with a purchase, while perceived value refers to the benefits or worth a consumer perceives from a product or service. Marked negative reporting refers to negative reviews or news articles that can impact a consumer's trust in a platform. Managerial implication: The study's managerial implications highlight the importance of reducing perceived risk, increasing perceived value, and building trust to improve social e-commerce marketing strategies. Organizations should be cautious of negative reporting and its impact on consumer behavior. Researchers can gain insights into consumer behavior and the significance of trust and perceived risk in driving purchase decisions in the context of social e-commerce marketing. To encourage cross-platform purchases, companies should address negative reports from other platforms, establish trust with consumers, and find ways to minimize perceived risk associated with social media purchases. Value: A significant contribution to the body of knowledge is achieved through testing and subsequent confirmation of the effects of online purchasing in the study model. This research enhances our understanding of the factors influencing online purchasing in the context of social cross-platform interactions. By investigating the relationship between Risk, Trust and various constructs within the social cross-platform framework, this study provides valuable insights into the dynamics of consumer behavior in the digital era.

Chronic Renal Disease considered as one of the serious, persistent illness which is caused by the impairment of the kidney. By the help of Machine Learning Techniques, as well as Deep Learning Techniques the illness can easily be detected at a previously stable and in a good time for the illness to be treated. Symptoms of Renal Disease cannot easily be found out. Therefore, the risk of it spreading at a higher rate is not accurately found out at early stages. It was earlier seen that the major factor resulting in renal disease was age, but from the past five years it is observed that even young individuals are being infected with this disease. So, with the help of the common symptoms, and with the images, a Deep Learning technique is proposed using image processing to identify renal disease. By the help of Transformers and Autoencoders, a model with better accuracies is proposed for Chronic Renal Diseases.

Euclidean Distance has been an important metric for the calculation of distances between points on a scaled space. Recent times have seen an increased use of this metric for the purposes of pattern recognition. Hand Landmarks are a group of twenty-one points which define the skeletal structure of a human’s hand. Calculation of distances between Hand Landmarks using Euclidean Distance can cater to the need of an efficient algorithm which could recognize patterns using mathematical calculations. The last few years have witnessed the introduction of technologies which promoted virtualization of important hardware thereby, increasing the portability and handling of the systems. The Mouse is an important aspect of the computer which helps to easily interfacing with the system. History has seen the use of trackballs, laser sensors and track pads as means of controlling the system however, each one of them consisted of costly hardware equipment be it the material or the sensors used. This paper proposes a virtual mouse system which can be controlled and navigated using gestures made using any one hand, whether left or right, without the involvement of any external hardware or sensor. All this is done by means of Euclidean distances and hand landmarks. It allows the user to control all the functioning of the mouse including its movement, mouse clicks and drag and drop without physically touching any hardware. It uses only the system’s inbuilt camera to carry-out the functioning of the mouse. The system is carried out using Python along with OpenCV, Mediapipe and the PyAutoGui libraries. All the gestures become visible on the camera screen with appropriate marking to facilitate representing the working of the system.

Objective: The present study aims at developing an integrated model to identify assorted factors and also investigates the influence of identified factors on consumers’ behavioral intentions to use or not to use one particular M-wallet for payment. Method: A conceptual model is proposed and validated. Besides this, 600 questionnaires were distributed and 482 were deemed usable. Structural equation modeling was used to demonstrate the stability of the proposed model and to test research hypotheses. Results: The results showed that behavioural intention is significantly influenced by perceived value, trust, compatibility and social influence while consumers’ is less optimistic to use M-wallet with regard to perceived enjoyment. The study also showed that trust followed by compatibility has a more powerful influence on the behavioural intention of consumers in the context of M-payment. Conclusions: This study impacts researchers and India's mobile payment sector. Financial and banking institutions, entrepreneurs, retailers, policymakers, government, and telecommunications sectors benefit from research. This research could help policymakers plan and develop tactics to help m-wallet service providers achieve a cashless society. Empirical results demonstrate the value of the TAM, TRA, and TPB model in understanding youth mobile wallet use.

Out of the various types of primary bone cancers, Osteosarcoma is one of the most common malignant bone tumors. Children and teenagers are mostly affected by Osteosarcoma, which weakens the strength of their bones and sometimes may even result in death. It is important to develop an intelligent classifier that detect osteosarcoma accurately so that proper treatment can be given to the patient timely. In this paper, we propose an intelligent classifier that classifies osteosarcoma whole slide images (WSIs) into Viable Tumor, Non-Viable Tumor and Non-Tumor. To extract the region of interest (ROI) from WSIs, a Multi-Feature Non-Seed-based Region Growing algorithm (MFNSRG) based on intra-region homogeneity and inter-region heterogeneity maximization is used. We use textural heterogeneity along with color heterogeneity as the matching criteria during region growing. Finally, the background is eliminated using thresholding based on the size of a region and the ROI is obtained. The performance of MFNSRG is further improved by using Marine Predators, a recently proposed metaheuristic algorithm, which is used to obtain optimal value of segmentation parameters. Here, we use handcrafted methods to extract relevant features from the segmented image which are given as input to the classifier. The results of experimentation prove the superiority of the proposed approach as compared to the existing state-of-the-art algorithms.