International Journal of Communication Networks and Information Security (IJCNIS)
Not a member yet
1021 research outputs found
Sort by
Optimizing ESG Management in Enterprises Using 'Internet +' and Big Data Technologies
No full textThe in-depth development of "Internet +" has made the application of data information technology in ESG (Environmental, Social and Governance) management more extensive and has made the optimization of ESG management the focus of listed companies. The original statistical evaluation method cannot solve the problem of improving the level of ESG management, and the ability to improve the quality of the environment, society and the company is weak, increasing ESG management's difficulty. Therefore, this paper proposes a method based on big data technology to optimize ESG management. First, the "Internet +" technology is used to analyze environmental, social and company quality data, and ESG management classification is carried out according to the requirements of listed companies. Then, according to the classification results, the "Internet+" technology forms different ESG management sets and iteratively mines different sets to obtain the optimal optimization strategy. After MATLAB testing, the effect of ESG management can be improved based on Internet + technology. The effect improvement rate reaches 92.6%, the ESG management process is improved, the improvement is 25%, and the ESG management time is reduced to 15 seconds, which meets the ESG management needs of listed companies
Robust Email Phishing Detection using Machine Learning and Deep Learning Approach
No full textPhishing Email is a type of attack that can have serious consequences, despite the advances research on the field of security. Recently the number of Phishing attacks have been increased because of Internet widespread in various daily life applications. Phishing attack have various types, which make the task of detection, using the classical solutions of security no longer practical. According to many researches, approaches based on artificial intelligence, specifically on learning modeles have proven to be an effecient solution of cyber Security. This study paper’s objective is to propose an efficient and accurate method for enhancing phishing emails detection, based on learning model and features selection technique such as Term Frequency-Inverse Document Frequency (TF-IDF) vectorization to extract only the significant features. For that reason, different computational models were developed: LSTM, Random Forest, SVM and three Naive Bayes variants (ComplementNB, BernoulliNB, and MultinomialNB) and tested rigorously using a specialized dataset. The results obtained from our experiment, revealed that the SVM model achieved the highest scores in all validation tests, with an accuracy of 0,971208, followed by the BernoulliNB with an accuracy of 0.967218. This demonstrates that our approach can enhance the accuracy rate with a minimum time of execution and that our approach is robust and reliable in terms of email phishing detection
Proactive Memory Efficient DDOS Attack Detection at IoT Gateways through Microneural Networks with Jellyfish Optimizer
No full textIn fact, the modernization of IoT technology has led to the emergence of cyber hazards with DDoS attacks as particularly risky for IoT service availability. This occurs by overwhelming the targeted services with traffic from multiple sources hence disrupting it. In the context of IoT and DDoS attack detection, Identifying and mitigating these attacks across connecting nodes is essential to ensure the smooth functioning of IoT services. IoT network is a collection of diverse elements namely 6LOWPAN, LoRAWAN etc. Effective mitigation techniques should be designed which is applicable to diverse sets of connecting IoT nodes including gateways, fully connected nodes, and edge devices for early detection of attacks. These nodes are of both categories namely constrained and non- constrained. So, mitigation policies designed must be with low memory consumption and inference time. This study presents a fresh DDoS attack detection model, a Jellyfish (JF) Jaro optimizer for feature selection with Light weight Micro Neural Network with string interning (MNNSI) on the IoT platform. The main goal of the MNNSI-JF approach is in the efficient & automated detection of DDoS attacks with minimal memory usage. The assessment of the simulation value of the MNNSI- JF methodology was evaluated on the benchmark dataset and based on the result it shows that the improvements over other prior models were evaluated in terms of distinct measures such as memory consumption & inference time
Hybrid Deep Neural Network for Predicting Online Social Network Student Performance
No full textIn educational data mining, one of the most crucial problems is predicting student success. In this work, a deep neural network is used to build an early prediction model of students' performance together with a method for characterizing students’ incomplete learning activity sequence.Educational data mining is more successful than statistical techniques for data exploration and student performance analysis in educational contexts.The purpose of this research is to examine how different variables affect students’ performance by using deep neural networks, machine learning, and data mining techniques.In order to develop a hybrid model, the paper presents the Hybrid Convolution Neural Network with Long Short-Term Memory (HCNN-LSTM) approach, which can extract spatial information and the LSTM method can extract temporal characteristics. In the meantime, the model's performance was improved by the proposed approach by adding dropout layers and batch normalization. Using the real-time high dimensional OSN user dataset, the hybrid CNN-LSTM technique that was suggested to predict students' success in Excel and Vivekanandha classes using neural network methods was tested. The proposed hybrid classification performance strategy yielded an accuracy of 99.12 percent and an F1-score of 98.42 percent, according to the experimental findings obtained from the OSN User datasaet
Influencing Responsibly: Harnessing the Power of Social Media Influencers to Promote Responsible Tourism
No full textThis study delves into the pivotal role played by social media travel influencers on advocating for responsible tourism practices. We begin by examining the emergence of travel influencers and their effect on moulding travel trends and tourist behaviour. We also explore responsible tourism, emphasising its significance in reducing adverse social and environmental effects while maximising advantages for local communities. We also analyse how influencers may promote responsible tourism by showcasing sustainable travel choices, bolstering local economies, and advocating for cultural conservation. Apart from the merits of influencers' responsible tourism promotion, possible drawbacks of influencer marketing, such as overtourism and cultural commercialization, and provide methods to reduce these dangers have been discussed. We also investigate the consequences of influencer-government partnerships in advancing responsible tourism efforts. In conclusion, influencers, governments, and tourism stakeholders need to collaborate in order to encourage responsible tourism. Influencers may responsibly use their power to contribute to the development of a more responsible and ethical tourism sector that is advantageous for both tourists and local communities
Enhancing Distance Learning For Non-Sighted and Visually Impaired Design Students: A Study on the Usability and Effectiveness of Haptic Tools
No full textIn the realm of design education, catering to the needs of non-sighted and visually impaired students poses unique challenges, particularly in distance learning environments. This study investigates the usability and effectiveness of a specially designed haptic tool tailored to the needs of such students. The participants, comprising non-sighted and visually impaired design students enrolled in distance learning programs, were recruited through purposive sampling techniques, ensuring diverse perspectives and experiences. Data were collected through focus group discussions, in-depth interviews, and structured surveys, providing both qualitative and quantitative insights. Results indicate that the haptic tool received positive usability ratings, with participants acknowledging its effectiveness in facilitating understanding of visual concepts commonly encountered in design education. Additionally, participants reported perceived benefits including enhanced understanding of visual concepts, improved engagement in design tasks, facilitated collaboration with peers, and increased confidence in design abilities. However, challenges such as technical issues, learning curve, limited customization options, and accessibility barriers were also noted, suggesting areas for improvement in haptic tool design and implementation. Furthermore, correlations between participants' prior experience with haptic technology and usability ratings indicate the potential benefits of familiarity with tactile feedback systems in enhancing user interaction and satisfaction. Overall, this study underscores the importance of integrating haptic technology into distance learning environments to better support the learning needs of non-sighted and visually impaired design students, while also highlighting avenues for further research and development in this domain
Artificial Intelligence Driven Customer Relationship Management: Harnessing the power of technology to improve business efficiency
No full textThis paper investigates how the Artificial Intelligence (AI) has significantly affected Customer Relationship Management (CRM), with focus on the transformative potential of AI tools like chatbots and predictive analytics in transforming customer-business interactions. Companies that integrate chatbots can personalize their assistance 24/7, thus improving client involvement and satisfaction. Additionally, another benefit from predictive analytics, is, the successful interpretation of customers’ behaviour patterns and their future requirements to enable early or precise tailoring of the experience. It also strengthens the existing business relationship with the customers and makes business efficient and effective in terms of increased sales turnover and revenues. This study by applying the mixed method approach underlines the crucial role of management and clients’ centric approach in conditions of the intensified competition of the nowadays high-stake market environment. The research results show that the use of AI in CRM systems can be critically beneficial for a business since such a system can enhance customer experience and provide decision-makers with tools to enhance their understanding of consumer conduct and behaviours. Such competencies help companies increase their long-term performance on the market since they uncover the potential of AI in CRM. Altogether, the findings are highly beneficial as it reveals the opportunities of leveraging AI in CRM systems to deliver the clear perspectives for improving interactions with clients and organisation’s performance.The research findings demonstrate that AI-powered CRM systems offer a significant competitive advantage by enriching customer interactions, uncovering deep insights into consumer behaviours and supporting better strategic decisions making. These competencies enable companies to strengthen their market position for long-term performance by revealing the true potential of artificial intelligence in CRM. The findings are highly beneficial as it brings forth insights into AI-powered CRM systems provide a clear roadmap for enhancing customer engagement and operational efficiency
Air Pollution Attribute-Based Lung Disease Detection with the RESNET Deep Learning Algorithm
No full textAir pollution causes respiratory illnesses. Pollution harms respiratory health, thus accurate disease detection is essential. Deep learning is used to diagnose lung issues by examining pollutant attributes. We compared RESNET deep learning sickness categorization with CNN and RNN. Pollution and lung disease labels were collected for our study. PM, NO2, O3, and CO indicate air pollution. We verify data compatibility with deep learning models. Lung diseases are classified using RESNET, CNN, and RNN. Train algorithms utilizing pollutant characteristics and sickness diagnosis data. Loss functions and performance criteria improve models during training. We evaluate deep learning methods using accuracy, precision, recall, and F1-score. The algorithms' environmental parameter lung disease diagnosis accuracy is shown by the measurements. CNN and RNN are less accurate in diagnosing illness than RESNET. RESNET detects pollution-related lung diseases with higher precision, recall, and F1-score. This study shows that deep learning systems like RESNET can reliably diagnose lung illnesses using pollutant attributes. The findings improve environmental well-being by providing a reliable and efficient method for disease detection and risk assessment in contaminated environments
Integration of Artificial Intelligence in Activity-Based Project Costing: Enhancing Accuracy and Efficiency in Project Cost Management
No full textActivity-Based Project Costing (ABPC) has long been recognized as an effective method for managing project costs. However, the increasing complexity of modern projects demands more sophisticated approaches. This study explores the integration of Artificial Intelligence (AI) into ABPC to enhance cost estimation accuracy and project management efficiency. By utilizing machine learning algorithms and big data analysis, it has been developed an AI-ABPC model capable of predicting project activity costs with higher precision, identifying hidden patterns in historical data, and providing real-time cost optimization recommendations. A case study of 50 large-scale construction projects showed that the AI-ABPC model improved cost estimation accuracy by 30% and reduced cost analysis time by 40% compared to traditional ABPC methods. These findings pave the way for a revolution in project cost management, enabling faster and more accurate decision-making in dynamic project environments. The implementation of AI in ABPC not only enhances project financial performance but also fosters innovation in overall project management practices
ENHANCING TRAFFIC MANAGEMENT AND BROADCASTING EFFICIENCY IN VANETs THROUGH AODV ROUTING
No full textOne of the main challenges today is detecting congestion on urban highways. By adopting judicious choices Vehicular Ad-Hoc Networks (VANETs) offer a helpful means of reducing overcrowding. The inquiry proposes a system for vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication-based roadblock detection to avoid delays. It provides drivers with the most up-to-date details on destinations and intensity of traffic jams. The broadcast notifications help drivers choose the best routes by emphasizing locations with heavy traffic and slow-moving vehicles. Several approaches have been put forth to deal with congestion detection. In order to improve skills for avoiding congestion, this project will include a centralized server and analyze data from vehicular communication. By exploiting data gathered and distributed through V2V interactions, the system efficiently and quickly detects congestion. The results are simulated and presented with AODV