Browse Items (2150 total)
Sort by:
-
A Way Towards Next-Gen Networking System for the Development of 6G Communication System
In this talk, the advancements announced by sixth-generation mobile communication (6G) as compared to the earlier fifth-generation (5G) system are carefully examined. The analysis, based in existing academic works, underscores the goal of improving diverse communication aims across various services. This study finds five crucial 6G core services designed to meet distinct goal requirements. To explain these services thoroughly, the framework presents two central features and delineates eight significant performance indices (KPIs). Furthermore, a thorough study of supporting technologies is performed to meet the stated KPIs. A unified 6G design is suggested, imagined as a combination of these supporting technologies. This design plan is then explained by the lens of five prototype application situations. Subsequently, possible challenges contained in the developing track of the 6G network technology are carefully discussed, followed by suggested solutions. The debate ends in an exhaustive examination of possibilities within the 6G world, seeking to provide a strategy plan for future research efforts. 2024 IEEE. -
Cached-N-Proxy: An Intelligent Proxy Algorithm for Preventing Insider Email Threats to Mail Servers
Insider threats are serious security risks that come from people who work for or are contracted by an organization, such as partners, employees, or contractors. These people use their authorized access to commit hostile acts against the infrastructure, data, or assets of the company. Serious ramifications could result from these dangers, such as financial losses, reputational harm, data breaches, and possible threats to national security. Enterprises must strengthen their defenses with strong intrusion detection and prevention systems because of the growing attack surface for insider threats caused by the increasing adoption of digital technology and remote work habits. Organizations must use a combination of preventive strategies and detection mechanisms, such as privileged access management (PAM), role-based access control, data loss prevention (DLP) techniques, two-factor authentication, and thorough insider threat awareness training, to effectively combat insider threats. 2024 IEEE. -
Identification of Driver Drowsiness Detection using a Regularized Extreme Learning Machine
In the field of accident avoidance systems, figuring out how to keep drivers from getting sleepy is a major challenge. The only way to prevent dozing off behind the wheel is to have a system in place that can accurately detect when a driver's attention has drifted and then alert and revive them. This paper presents a method for detection that makes use of image processing software to examine video camera stills of the driver's face. Driver inattention is measured by how much the eyes are open or closed. This paper introduces Regularized Extreme Learning Machine, a novel approach based on the structural risk reduction principle and weighted least squares, which is applied following preprocessing, binarization, and noise removal. Generalization performance was significantly improved in most cases using the proposed algorithm without requiring additional training time. This approach outperforms both the CNN and ELM models, with an accuracy of around 99% being achieved. 2023 IEEE. -
Enhanced Edge Computing Model by using Data Combs for Big Data in Metaverse
The Metaverse is a huge project undertaken by Facebook in order to bring the world closer together and help people live out their dreams. Even handicapped can travel across the world. People can visit any place and would be safe in the comfort of their homes. Meta (Previously Facebook) plans to execute this by using a combination of AR and VR (Augmented Reality and Virtual Reality). Facebook aims to bring this technology to the people soon. However, a big factor in this idea that needs to be accounted for is the amount of data generation that will take place. Many Computer Science professors and scientists believe that the amount of data Meta is going to generate in one day would almost be equal to the amount of data Instagram/Facebook would have generated in their entire lifetime. This will push the entire data generation by at least 30%, if not more. Using traditional methods such as cloud computing might seem to become a shortcoming in the near future. This is because the servers might not be able to handle such large amounts of data. The solution to this problem should be a system that is designed specifically for handling data that is extremely large. A system that is not only secure, resilient and robust but also must be able to handle multiple requests and connections at once and yet not slow down when the number of requests increases gradually over time. In this model, a solution called the DHA (Data Hive Architecture) is provided. These DHAs are made up of multiple subunits called Data Combs and those are further broken down into data cells. These are small units of memory which can process big data extremely fast. When information is requested from a client (Example: A Data Warehouse) that is stored in multiple edges across the world, then these Data Combs rearrange the data cells within them on the basis of the requested criteria. This article aims to explain this concept of data combs and its usage in the Metaverse. 2023 IEEE. -
Improving Speaker Gender Detection by Combining Pitch and SDC
Gender detection is helpful in various applications, such as speaker and emotion recognition, which helps with online learning, telecom caller identification, etc. This process is also used in speech analysis and initiating human-machine interaction. Gender detection is a complex process but an essential part of the digital world dealing with voice. The proposed approach is to detect gender from a speech by combining acoustic features like shifted delta cepstral (SDC) and pitch. The first step is preprocessing the speech sample to retrieve valid speech data. The second step is to calculate the pitch and SDC for each frame. The multifeature fusion method combines the speech features, and the XGBoost model is applied to detect gender. This approach results in accuracy rates of 99.44 and 99.37% with the help of RAVDESS and TIMIT datasets compared to the pre-defined methods. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Improvement of Speech Emotion Recognition by Deep Convolutional Neural Network and Speech Features
Speech emotion recognition (SER) is a dynamic area of research which includes features extraction, classification and adaptation of speech emotion dataset. There are many applications where human emotions play a vital role for giving smart solutions. Some of these applications are vehicle communications, classification of satisfied and unsatisfied customers in call centers, in-car board system based on information on drivers mental state, human-computer interaction system and others. In this contribution, an improved emotion recognition technique has been proposed with Deep Convolutional Neural Network (DCNN) by using both speech spectral and prosodic features to classify seven human emotionsanger, disgust, fear, happiness, neutral, sadness and surprise. The proposed idea is implemented on different datasets such as RAVDESS, SAVEE, TESS and CREMA-D with accuracy of 96.54%, 92.38%, 99.42% and 87.90%, respectively, and compared with other pre-defined machine learning and deep learning methods. To test the real-time accuracy of the model, it has been implemented on the combined datasets with accuracy of 90.27%. This research can be useful for development of smart applications in mobile devices, household robots and online learning management system. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Cancer Tumor Detection Using Genetic Mutated Data and Machine Learning Models
Early detection of a disease is a crucial task because of unavailability of proper medical facilities. Cancer is one of the critical diseases that needs early detection for survival. A cancer tumor is caused due to thousands of genetic mutations. Understanding the genetic mutations of cancer tumor is a tedious and time-consuming task. A list of genetic variations is analysed manually by a molecular pathologist. The clinical strips of indication are of nine classes, but the classification is still unknown. The objective of this implementation is to suggest a multiclass classifier which classifies the genetic mutations with respect to the clinical signs. The clinical evidences are text-evidences of gene mutations and analysed by Natural Language Processing (NLP). Various machine learning concepts like Naive Bayes, Logistic Regression, Linear Support Vector Machine, Random Forest Classifier applied on the collected dataset which contain the evidence based on genetic mutations and other clinical evidences that pathology or specialists used to classify the gene mutations. The performances of the models are analysed to get the best results. The machine learning models are implemented and analyzed with the help of gene, variance and text features. Based on the variants of gene mutation, the risk of the cancer can be detected and the medications can be prescribed accordingly. 2022 IEEE. -
Regression Analysis as a Metric for Sustainability Development: Validation of Indian Territory
The 2030 Development Agenda styled' Transforming our world The 2030 Agenda for Sustainable Development' was hugged by the transnational locales of the UN General Assembly in 2015. Monitoring the progress of countries towards achieving these pretensions is pivotal for sustainable development. This exploration paper offers an innovative stance toward foretelling the SDG Index of Indian states for the near future times using machine learning ways, logical and visualization tools. The paper focuses on India's sweats towards achieving the SDGs and investigates the factors impacting the SDG performance of individual Indians states. A comprehensive dataset is collected, encompassing a wide range of socio-profitable pointers, demographic data, and environmental criteria applicable to each SDG target. Literal SDG Index scores and corresponding state-specific data are collected to assay and find some trends. The study demonstrates the eventuality of vaticination ways in vaticinating the unborn SDG Index scores of Indian states. The time series graph showcases varying degrees of delicacy across different SDGs, indicating the complexity and diversity of experimental challenges. 2024 IEEE. -
DKMI: Diversification ofWeb Image Search Using Knowledge Centric Machine Intelligence
Web Image Recommendation is quite important in the present-day owing to the large scale of the multimedia content on the World Wide Web (WWW) specifically images. Recommendation of the images that are highly pertinent to the query with diversified yet relevant query results is a challenge. In this paper the DKMI framework for web image recommendation has been proposed which is mainly focused on ontology alignment and knowledge pool derivation using standard crowd-sourced knowledge stores like Wikipedia and DBpedia. Apart from this the DKMI model encompasses differential classification of the same dataset using the GRU and SVM, which are two distinct differential classifiers at two different levels. GRU being a Deep Learning classifier and the SVM being a Machine Learning classifier, enhances the heterogeneity and diversity in the results. Semantic similarity computation using Cosine Similarity, PMI and SOC-PMI at several phases ensures strong relevance computation in the model. The DKMI model yields overall Precision of 97.62% with an accuracy of 98.36% along with the lowest FDR score of 0.03 and is much better than the other models that are considered to be the baseline models. 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG. -
An Efficient Multi-Modal Classification Approach for Disaster-related Tweets
Owing to the unanticipated and thereby treacherous nature of disasters, it is essential to gather necessary information and data regarding the same on an urgent basis; this helps to get a detailed overview of the situation and helps humanitarian organizations prioritize their tasks. In this paper, "An Efficient Multi-Modal Classification Approach for Disaster-related Tweets,"the proposed framework based on Deep Learning to classify disaster-related tweets by analyzing text and image contents. The approach is based on Gated Recurrent Unit (GRU) and GloVe Embedding for text classification and VGG-16 network for image classification. Finally, a combined model is proposed using both text and image modules by the Late Fusion Technique. This portrays that the proposed multi-modal system performs significantly well in classifying disaster-related content. 2022 IEEE. -
Categorization of artwork images based on painters using CNN
Artworks and paintings has been an integral part of human civilization since the dawn of the Stone Age. Paintings gives more insight about any subject compared to the scriptures and documents. Archiving of digital form of paintings helps to preserve the artworks of different painters. The anticipated work is aimed for the classification of painters' artworks. The artworks of Foreign & Indian painters are considered for the proposed work. The foreign painters' artworks are obtained from [14]. At present, the Indian painters' artwork dataset is not readily available. The images were downloaded from the specific genuine website [13]. Conventional Neural Network is used for Feature learning and classification. Around 20k images of artworks is used for the experiment and got an average accuracy of 85.05%. Published under licence by IOP Publishing Ltd. -
Secure Decentralization: Examining the Role of Blockchain in Network Security
Blockchain generation has emerged as a novel answer for securing decentralized networks. This technology, which was first created for use in crypto currencies, has received enormous interest in recent years because of its capability for boosting protection in various industries and community protection. The essential precept at the back of block chain technology is the decentralization of statistics garage and control. In a decentralized network, no central authority may control the statistics. Rather, the facts are shipped amongst multiple nodes, making it immune to tampering and single factors of failure. One of the most important advantages of blockchain in community protection is its capacity to offer cozy and transparent communication amongst community customers. Through cryptographic techniques, block chain can affirm the identities of network participants and ensure the authenticity of records trade. This feature is extraordinarily valuable in preventing unauthorized access and facts manipulation. 2024 IEEE. -
Antecedents of Ethical Goods and Services Tax Culture among young adults - Special Reference to Maharashtra and Karnataka
Since the implementation of the Goods and Services Tax (GST) in 2017, it has become clear that this new Indian indirect tax system is here to stay. The Indian GST Council is continuously deliberating and making efforts to improve GST revenue collection at the state and central levels. The focus is now on the young adults in the country who will play a vital role in shaping the future of GST compliance. Their tax mentality and behaviour in contributing to GST revenue as daily consumers will determine the ethical tax culture in India. They need to understand how crucial their role is in discouraging evasive practices by sellers in the unorganised retail sector at the point of sale. The study utilized structural equation modelling to test the acceptability of the model. The process was supported by a structured questionnaire, with 324 respondents between the age group of 17-30 years. Understanding GST significantly influences acceptance of GST as a tax system, however, the acceptance of the GST tax system does not significantly lead to young adults discouraging the evasive behaviour of sellers in the unorganised retail sector at the point of sale. And, finally, the discouragement of evasive behaviour by young adults does influence the possibility of an ethical GST tax culture. The respondents majorly represented young adults between 17-20 years of age. The model has not measured the existence of covariance among the variables, nor has any mediating or moderating factors been identified, as GST tax culture in the Indian context is still unexplored and GST in itself is relatively new in the country. 2024 IEEE. -
Digital Transaction Cyber-Attack Detection Using Particle Swarm Optimization
The cyber digital world is an essential variant in day-to-day life in advanced technology. There is a better change in the lifestyle as intelligent technology. In larger excite to increase the advanced technology which can be developed to humans in major dependent on network and internet users. Now, in modern times, the internet has changed the primary need in human lifestyle by giving access to everything in the world while sitting in one place knowing and updating the information and usage of online subscribers or Revolution. The world is moving in Rapid and Faster communications within a fraction of a second, at a lesser cost, and it has minimal paper-based processes and relies on the digitization document instead of a paperless environment. The data is handled by finch security practices, which are used in security worldwide to establish protected data management systems like digital lending, credits, mobile Banking, and mobile payment. Cryptocurrency and blockchain, B-trading, and banking as a service are included. At the same time, leveraging the new technologies is to resist hacking cyber-attacks. This article is also involved in artificial intelligence and machine learning (AI&ML) in different cyber-attacks. This article focuses on genetic algorithms to detect the cyber-attack. The main aim of the detection is future to prevent these cyber-attacks. The comparison will take two sample genetic algorithms. The first one is taken for Ant Colony Optimization (ACO), and the proposed model is taken for Particle Swarm Optimization. The average attack detection of ACO algorithm is 45 packets at the same time PSO algorithm will detect 50 packets. 2023 IEEE. -
Structural Health Monitoring Using Machine Learning Techniques
Environmental factors, particularly vibrations and temperature can damage the structural health of the building. To avoid heavy damage to the building and to maintain the building's structural health this paper suggests monitoring of building using machine learning algorithms. Machine learning algorithms are used to predict temperature and vibration damages in buildings. Temperature and vibration values are obtained through the grove vibration sensor and NTC thermistor attached to Raspberry Pi 3B plus. In the Raspberry pi, Machine learning algorithms are executed. The activation functions used are Relu, Sigmoid, and Tanh. The experimental results reveal that the Sigmoid activation function gives the best results in terms of metrics with accuracy 94.25, Precision 0.951, Recall 0.912, and F1 score 0.388. The sigmoid function is used in machine learning algorithms for predicting temperature and vibrations. Predicted temperature and vibrations damages are sent to the server and viewed through the user mobile. K- Nearest Neighbor algorithm produced best results with an accuracy rate of 85.50, Precision of 0.922, Sensitivity of 0.830, Specificity of 0.840 and F1 score of 0.873. 2023 IEEE. -
A Deterministic Key-Frame Indexing and Selection for Surveillance Video Summarization
Video data is voluminous and impacts the data storage devices as there are CCTV surveillance videos being created every minute and stored continuously. Due to this increase in data there is a need to create semantic information out of the frames that are being stored. Video Summarization is a process that continuously monitors changes and helps in reducing the number of frames being stored. This work enables summarization to be carried out based on selecting threshold-based system that can select key-frames ideally suit for storage and further analysis. Initially a Global threshold based on Otsus method is carried out for all frames of a surveillance video and based on the set threshold a retrospective comparison is done on each frame based on statistical methods to converge on determining the keyframes. A similarity index is generated based on the iterative comparison of frames based on global and local threshold comparison. The local threshold is indexed based on Analysing Method Patterns to Locate Errors(AMPLE), An-derbergs D(AbD), Cohens Kappa(CK), Tanimoto Similarity(TS), Tversky feature contrast model(TFCM), Pearson coefficient of mean square contingency(Pmsc). The Global threshold is updated each time a keyframe is selected based on the comparison of local and global threshold. The results are compared with five surveillance videos and six methods to identify keyframes Selection Rate is the metric used for calculating the performance. 2019 IEEE. -
A classified study on semantic analysis of video summarization
In today's world data represented in the form of a video are prolific and has increased the requisite of storage devices unconditionally. These video sets takes up a huge space for amassing data and takes a long time to ascertain the content that requires a higher cognitive process for content search and retrieval. The efficient method for storing video data is to remove high-degree redundancies and for creating an index of important events, objects and a preview video based on vital key-frames. These requirements imbibes the need to build algorithms that can concise the necessity of space and time for video and adequate approaches are to be developed to solve the needs of summarization. The three effective attributes for a semantic summarized video system are Un-supervision, efficient and dynamically scalable system that can help in reducing time and space complexities. Dimensionality reduction based on sub space analysis helps in plummeting the multidimensional data into a low-dimensional data to enable faster feature extraction and summarization. In this paper we have made a study and description related to several summarization methodologies for video's that are available. 2017 IEEE. -
Resume Ranking and Shortlisting with DistilBERT and XLM
The research presented in this paper offers a solution to the time-consuming task of manual recruitment process in the field of human resources (HR). Screening resumes is a challenging and crucial responsibility for HR personnel. A single job opening can attract hundreds of applications. HR employees invest additional time in the candidate selection process to identify the most suitable candidate for the position. Shortlisting the best candidates and selecting the appropriate individual for the job can be difficult and time-consuming. The proposed study aims to streamline the process by identifying candidates who closely match the job requirements based on the skills listed in their resumes. Since it is an automated process, the candidate's individual preferences and soft skills remain unaffected by the hiring process. We leverage advanced Natural Language Processing (NLP) models to improve the recruitment process. Specifically, our emphasis lies in the utilization of the distilBERT model and the XLM (Crosslingual Language Model). This paper explores the application of these two models in taking hundreds of resumes for the job as input and providing the ranked resumes fit for the job as output. To refine our approach further, two types of metrics for resume ranking, such as Cosine similarity score and Spatial Euclidean distance, are used, and the results are compared. Intriguingly, distilBERT and XLM result in different sets of top ten ranked resumes, highlighting the nuanced variations in their ranking approaches. 2024 IEEE. -
A review on prediction of cardiac arrest analysis in early stage
Cardiac arrest occurs as the heart muscle fails to contract properly, resulting in a sudden loss of blood supply. The ECG signal is one of the techniques for detecting cardiac electrical activity and is used to investigate heart block. In this paper different standardized work for early detection of cardiac arrest is described. Stages of ECG signal pre-processing involves denoised using digital filtering algorithms and extracting different features from clean ECG predicting cardiac arrest in early stage. Several other body parameters were also considered for this purpose. In this work denoising validation parameters were compared for showing effectiveness of the filtering algorithm and several body parameters and its implication on cardiac arrest was described. 2022 Author(s). -
Attention Based Meta-Module to Integrate Cervigrams with Clinical Data for Cervical Cancer Identification
Cervical cancer remains a significant burden on public health, particularly in developing countries, where its malignancy and mortality rates are alarmingly high. Early diagnosis stands as a pivotal factor in effectively treating and potentially curing the cervical cancer. This study introduces a novel approach of meta module based on recurrent gate architecture designed to enhance the classification of cervix images efficiently. This innovative framework incorporates a meta module capable of dynamically selecting image modalities most pertinent attributes. Furthermore, it integrates clinical data with extracted image features and employs a range of EfficientNet architectures (B0-B5) for image classification. Our results indicate that the EfficientNet B5 architecture outperforms its counterparts, achieving an AUC (Area Under the Curve) score of 55.1 and an F1-Score of 75.1. Overall, this work represents a crucial step towards improving the early detection of cervical cancer, which in turn can lead to more effective treatment strategies and, ultimately, better outcomes for patients worldwide. The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.