International Journal of Communication Networks and Information Security (IJCNIS)
Not a member yet
1021 research outputs found
Sort by
Advanced Encryption Techniques in Biometric Payment Systems: A Big Data and AI Perspective
No full textIn the rapidly evolving landscape of biometric payment systems, the integration of advanced encryption techniques is crucial for ensuring robust security and privacy. This paper explores cutting-edge encryption methodologies tailored for biometric data in the context of big data and artificial intelligence (AI) applications. We investigate how these advanced techniques address the unique challenges posed by the vast amounts of sensitive biometric information generated and processed in modern payment systems. The study provides an overview of various encryption strategies, including homomorphic encryption, secure multi-party computation, and quantum-resistant algorithms, evaluating their effectiveness in safeguarding biometric data against emerging threats. Additionally, the paper examines the role of AI in enhancing encryption mechanisms and optimizing performance, highlighting how machine learning models can predict and mitigate potential vulnerabilities. By analyzing real-world case studies and empirical data, we offer insights into the practical implementation of these technologies and their impact on the security landscape of biometric payments. This research contributes to a deeper understanding of how advanced encryption and AI can collaboratively fortify biometric payment systems, ultimately paving the way for more secure and privacy-preserving financial transactions
Optimizing Panorama Generation from Real-Time Image and Videos: A Study of Feature Detectors and Descriptors
No full textThe proposed work aims to develop an image mosaicing model for combining images and evaluate its performance based on runtime and key features used. The model employs Histogram Equalization to ensure consistency across images and uses feature descriptors like SIFT, ORB, and BRISK. Features are matched with K- Nearest Neighbor, and homography is computed using RANSAC. Image warping is achieved with a 40x40 smoothing filter to create a panorama. Testing across various datasets shows that ORB, combined with Histogram Equalization, is the most efficient feature extractor, requiring only 500 key features and producing panoramic images with minimal runtime of 0.0836 seconds. The model addresses the limitations of existing methods by optimizing feature matching and runtime, demonstrating superior performance in creating seamless panoramic images. Video mosaicing creates panoramic views from video frames by seamlessly combining them into a comprehensive scene. It also addresses an automatic algorithm that aligns and blends non-overlapping frames, addressing camera motion and content variations. The algorithm effectively produces a continuous mosaic, demonstrating strong performance in frame alignment and blending
Digital Evidence and it’s Admissibility under the Indian Legal Regime
No full textThe advent of digital technology has revolutionized every aspect of modern-day society, including the judicial landscape. Often, technological advancements lead to an imbalance of power in favor of the party with the most access to technology and the most adept use of it in legal proceedings. This imbalance of power has a severe impact on the fairness of legal proceedings.
For instance, those who have access to the most up-to-date technology are in an advantageous position to collect, analyze, and present evidence more effectively and efficiently than those who do not, giving an unfair advantage in the courtroom.
Electronic records/ digital evidence is increasingly presented and accepted in courts without scientific validation of the digital forensic methodology or tools. While classical investigative measures are subject to strict limits and fair trial guarantees, digital investigations still lack quality assurance and accountability. There are no minimum standards for digital evidence to establish and enforce scientific validation in digital forensics.
In addition, digital advancements like Chat GPT have allowed for the introduction of automated systems that can analyze and interpret legal documents. These automated systems are often able to make decisions and render judgments more quickly than human lawyers, and they can often do so with less bias. This has led to an increase in the number of cases being decided by automated systems, which can lead to more unfair outcomes.
Contemporary criminal investigation assisted by computing technology imposes challenges to the right to a fair trial and the scientific validity of digital evidence. Admissibility of an evidence is a very crucial stage in any civil/criminal trial and substantially effects its outcome. Technological advancements keep on presenting new and unique challenges before the courts and judiciary by offering the various new forms of electronic evidences.Another challenge that is faced in regards to electronic evidence is the ease with which it can be forged, fabricated, and manipulated and makes it all the more difficult to decide about the admissibility and veracity.
The varied type of electronic evidence such as email, instant chat messages, SMS/MMS,
communication made on social networking platforms; data stored on hard disk/ memory card
CD, DVD, browsing history on search engines, etc. poses unique problem and challenges for proper authentication and subject to a different set of views.
This paper seeks to trace the changing legal regime regarding admissibility of digital/e-evidence with special reference to Search and seizure of digital evidence and its interaction with right to privacy as recognised under Indian constitution
The Role of Artificial Intelligence in Cybersecurity: Enhancing
No full textThis study explores the use of AI with increased application of themachine learning approach in improving cybersecurity.Addressing the significance of the four most widespreadalgorithms, CNN, RNN, RF and SVM, this work investigates theeffectiveness of these algorithms in the context of cyber threatsidentification and counteraction. To assess the performance of themodels, the system was validated with a large quantity of datasetswith emphasis on the detection capability, false alarm rate aswell as response time. The findings also show that the proposedCNN model attained the maximum detection accuracy of 96. 5 %,while developing new features the RNN was at 94%. 2%, RF at 91.5% while Naive Bayes is at 87. 7%, Random forest is at 87. 2% andSVM at 88. 9%. The false-positive rates were reported to be atlowest for CNN at 1. 8% more than Urban, thus testifying to itsincreased reliability. Moreover, it took a considerably less amountof time to give the response for CNN which was 0. 5 seconds,compared to 0scaled up for comparison with 5 seconds of readinga text online. 89 seconds for RNN, 1. That is 6 seconds in totalwhile the time taken for RF is 2 seconds, and 1 second for TF. 5seconds for SVM. These studies reaffirm the possibilities of theartificial intelligence and machine learning in improvingcybersecurity through optimized and more precise threatidentification and mitigation means. Subsequent work will bedevoted to continuing the model enhancement works as well as integration with live dataprocessing systems for theincreased effectiveness ofcybersecurity prediction and countermeasures
A Study on the Liquidity and Profitability Performance of Bharat Petroleum Corporation Limited (BPCL)
No full textThis study focuses on the liquidity and profitability performance of Bharat Petroleum Corporation Limited (BPCL) over the period from 2018 to 2023. The analysis encompasses various financial ratios, such as the current ratio, liquid ratio, absolute liquid ratio, gross profit ratio, operating ratio, operating profit ratio, and net profit ratio, to evaluate BPCL's financial health and efficiency. In the study, we have tried our best to show BPCL's resilience in managing its gross profits, liquidity position, and profitability. We have tried to suggest the need for improved financial strategies. Enhancing operational efficiency and better cost management are critical for achieving sustainable economic performance
Impact of Venture Capital on Startup Success Rates Across Indutry: an Emperical Study
No full textThis empirical study investigates the impact of venture capital (VC) on the success rates of startups across various industries. With venture capital playing a pivotal role in the growth and scaling of new ventures, the study aims to explore how VC funding influences startup outcomes in terms of survival, growth, and market competitiveness. Using a dataset of over 1,000 startups from diverse sectors such as technology, healthcare, and consumer goods, the analysis examines key variables including the size of investment, stage of funding, and the role of VC firms in strategic decision-making. The study employs regression analysis and survival models to assess the correlation between VC involvement and startup success metrics, controlling for industry-specific factors. Results indicate that startups with VC backing demonstrate higher survival rates and accelerated growth, particularly in technology and biotech sectors, compared to self-funded or traditionally financed startups. However, the impact of VC varies significantly across industries, with some sectors showing diminishing returns or negligible effects of venture capital investment. The findings contribute to the ongoing discourse on the role of venture capital in entrepreneurial ecosystems, highlighting both its benefits and limitations. The study concludes by suggesting industry-specific strategies for maximizing the advantages of VC funding and emphasizing the importance of aligning investor expertise with startup goals
Prediction of Lecturers' Satisfaction Using M-Learning by Fast Learning Network
No full textM-learning, or mobile learning, has seen remarkable advancement in the past few years owing to the growth of mobile technology. This development has greatly improved the features and usability of mobile learning applications. This paper aims to present and analyze the accelerated learning network model, which identifies the determinants of mobile learning satisfaction of lecturers at Southern Technical University. The model employs a questionnaire given to 250 participating lecturers using multiple variables to examine the factors that affect their satisfaction levels. The study results revealed that the suggested model was more effective than others, including ANN, KNN, and MLP, in predicting factors affecting lecturers' satisfaction regarding accuracy and specificity. The model also showed good accuracy and specificity, with the latter reaching 93.55% in the prediction of the satisfaction factors of lecturers on mobile learning, with an accuracy of 92.00%. It emphasizes the need to consider the different aspects of different assessment methods and lecturers in research within an m-learning context
Assessing the importance of collaborative decision-making as part of ADR in resolving environmental harm caused by industrialization
No full textThe research is aimed at analysing the effectiveness of ADR in promoting collaborative decision-making in resolving cases of environmental harm caused by industries. The research has pointed out that industrialisation is needed for economic and technological growth so a balance between this and environmental protection is needed. Closing industrial operations is not a solution to such issues rather ADR process can be applied to come to negotiation in terms of compensation, industrialisation and environmental protection. The research has conducted a case study analysis based on Stake’s case study analysis framework for intrinsic case studies. The research outcome reveals that ADR is ideal in cases where both parties are open to negotiation and the party at fault is ready for compensation but in cases where the parties are not ready for negotiation, judicial intervention is needed
Leveraging Machine Learning and SMOTE for Diabetes Prediction: Implementation of an Application Based on Indonesian Hospital Data
No full textDiabetes mellitus is a widespread chronic condition affecting millions globally, including a substantial population in Indonesia. Accurate and early detection is critical for effective management and treatment, and machine learning offers promising solutions for enhancing predictive accuracy. This study evaluates three machine learning algorithms: Support Vector Machine (SVM), Logistic Regression, and Naive Bayes, with and without the application of Synthetic Minority Over-sampling Technique (SMOTE) to tackle data imbalance. Data were meticulously collected from an Indonesian regional hospital, including various medical parameters such as age, body mass index (BMI), blood sugar levels, blood pressure, and family history. Our findings reveal that the SVM model, without SMOTE, achieved an accuracy of 95%, precision of 95%, recall of 97%, and an AUC of 98%. With SMOTE, SVM's performance improved to an accuracy of 95.8%, precision of 97%, recall of 94.6%, and an AUC of 99.1%. Logistic Regression without SMOTE demonstrated an accuracy of 94.8%, precision of 96.2%, recall of 96.2%, and an AUC of 98.3%, while with SMOTE, it reached an accuracy of 95.6%, precision of 97.9%, recall of 93.3%, and an AUC of 98.7%. The Naive Bayes model showed an accuracy of 93.5%, precision of 98.5%, recall of 91.9%, and an AUC of 98.1%, improving with SMOTE to an accuracy of 94.3%, precision of 98.3%, recall of 90.2%, and an AUC of 98.6%. The best-performing model, SVM with SMOTE, was implemented into a desktop application. This application successfully validated the model's predictive capabilities, demonstrating effective and accurate diabetes detection in practical scenarios. Our study highlights the significant impact of SMOTE on enhancing model performance and emphasizes the importance of sophisticated machine learning techniques in advancing healthcare diagnostics. This work provides a foundation for further development and deployment of predictive models in clinical settings, contributing to improved patient care and disease management
Designing the Interior of Pars Rasht Hospital Using the Color Psychology Approach
No full textNowadays, architects and designers are obsessed with designing beautifully appropriate environments using architectural models that would provide environmental comfort. As is known, such environmental designs could increase people’s quality of life in society. Since hospitals and treatment centers are used by humans and serve as places where humans interact with each other, it is essential to provide desirable designs for such places and the relevant environment that would increase the comfort of people, including patients and clients, as well as the treatment personnel. The method of the present study was descriptive-analytical, and library sources were used. Findings showed that surface colors should be so designed that they would not cause light reflection or glare, which will harm patients’ vision and cause related problems. Also, strong and dark colors should not be used on ceilings and floors. Colors used in administrative sections should be selected so as not to cause stress, nervous pressure, and fatigue for the employees. When designing the interior of hospitals, the impact of colors on patients should be considered. For example, cold colors may be useful for patients with high blood pressure and anxiety. Red color is recommended not to be used for patients with epilepsy, whereas blue color is not good for cardiac patients. It is thus important to design an intimate environment for children’s inpatient wards, which can be made possible by using happy, albeit neutral, colors.