IAES International Journal of Artificial Intelligence (IJ-AI)
Not a member yet
1769 research outputs found
Sort by
Flame analysis and combustion estimation using large language and vision assistant and reinforcement learning
Get PDFIn this study, we present an advanced approach for flame analysis and combustion quality estimation in carbonization furnaces utilizing large language and vision assistant (LLaVA) and reinforcement learning from human feedback (RLHF). The traditional methods of estimating combustion quality in carbonization processes rely heavily on visual inspection and manual control, which can be subjective and imprecise. Our proposed methodology leverages multimodal AI techniques to enhance the accuracy and reliability of flame similarity measures. By integrating LLaVA’s high-resolution image processing capabilities with RLHF, we create a robust system that iteratively improves its predictive accuracy through human feedback. The system analyzes real-time video frames of the flame, employing sophisticated similarity metrics and reinforcement learning algorithms to optimize combustion parameters dynamically. Experimental results demonstrate significant improvements in estimating oxygen levels and overall combustion quality compared to conventional methods. This approach not only automates and refines the combustion monitoring process but also provides a scalable solution for various industrial applications. The findings underscore the potential of AI-driven techniques in advancing the precision and efficiency of combustion systems
How ambidextrous entrepreneurial leaders react to the artificial intelligence boom
Get PDFArtificial intelligence (AI) now plays a central role in enhancing business competitiveness by transforming systems, frameworks, and managerial strategies. This study employs a systematic literature review (SLR) approach, utilizing the 'Consensus AI' search platform to explore the characteristics and roles of ambidextrous entrepreneurial leaders in the AI era. Consensus AI is an AI-powered search engine that automates the processes of reviews, literature searches, screening, and data extraction. It also utilizes 'research question searches' within SLRs to avoid the challenges of ambiguity and irrelevant information associated with 'keyword searches,' delivering more directly relevant results and finding featured snippets that answer specific questions. A research gap exists concerning how ambidextrous leadership adapts to the AI boom, highlighting leadership dynamics in the digital age. The findings emphasize the critical role of ambidextrous entrepreneurial leadership (AEL) in guiding organizations through the AI boom, enabling them to leverage AI for innovation, agility, and competitiveness. Organizations that effectively implement AEL by integrating AI technologies can position themselves for long-term success. Key insights show the importance of AEL approaches, and future research may explore challenges that arise for ambidextrous entrepreneurial leaders in the era of AI, such as ethical considerations and organizational culture
A new wrapper feature selection approach for binary ransomware detection
Get PDFConcerns about ransomware attacks have heightened in recent years for both individuals and organizations. Detecting these malicious attacks poses considerable challenges for cybersecurity professionals, particularly due to their ever-evolving nature. Although behavior-based detection methods show promise in recognizing new ransomware variants, they face significant hurdles, especially in managing the massive volumes of data generated from real-time malware behavior monitoring, leading to high dimensionality. This paper introduces a new feature selection approach specifically for binary ransomware detection. Our method emphasizes assessing the impact of feature categories on the effectiveness and speed of detection algorithms. It involves two stages: the first stage selects the most relevant groups (categories) of features, while the second ranks and identifies the important features within those categories. Experimental results indicate that our approach surpasses similar studies regarding accuracy and ability to minimize the original features set. Moreover, both computation speed and accuracy are notably enhanced when using the selected subset compared to the original features
Blockchain and machine learning driven agricultural transformation framework to enhance efficiency, transparency, and sustainability
Get PDFThe agricultural sector is undergoing a transformative journey empowered by technological innovations. In this context, this research work endeavors to revolutionize the agricultural supply chain (ASC) by developing a comprehensive online platform that connects sellers, farmers, and customers. Through meticulous planning, design, and implementation, the system aims to streamline the process of buying and selling agricultural products, thereby fostering efficiency, transparency and accessibility. The key features include user registration, product management, order tracking, and blockchain-machine learning (ML) based transaction security. The proposed research work's success hinges on thorough testing and validation, ensuring its reliability and usability. By leveraging technology to bridge gaps in the agricultural ecosystem, this proposed work seeks to empower stakeholders and contribute to the sustainable growth of the agricultural industry. In the current agricultural landscape in India, traceability has been a significant challenge. The industry lacks a comprehensive system that provides visibility into the source and quality of produce. Our proposed system aims to address the shortcomings of the existing agricultural ecosystem by introducing a comprehensive solution powered by blockchain technology and advanced data processing techniques
A symptom-driven medical diagnosis support model based on machine learning techniques
Get PDFMedicine is a human science that is constantly evolving, and this evolution generates a large mass of data that needs to be exploited with the multitude of IT resources available to guarantee and maintain this scientific progress. Some diseases share most symptoms, whereas others could have a low probability of being identified in an early stage. Thus, when facing a such situation, an inexperienced doctor may have difficulty making the right diagnosis or may test different cases, which will be a big waste of time. In this paper, we are going to make this embarrassing situation less complex by giving practitioners every probable disease, and even the least probable ones according to the given symptoms. Indeed, this work will push the diagnosis deeper to reveal hidden symptoms and pathogenesis, to help practitioners make the right decisions. To develop such a solution, the data is organized by matching each disease with its known symptoms, then we used naive Bayes as a classification model, and different metrics to evaluate the performance of this experiment. This work proves that machine learning has become very effective in the medical sector, especially when we notice that the accuracy exceeds 90% in the detection of diseases
Skin cancer diagnosis using hybrid deep pre-trained convolutional neural networks
Get PDFAs a variant of skin cancer, melanoma represents a substantial menace to the health and overall well-being of individuals. Statistics reveal that 55% of skin cancer patients succumb to this particular disease. However, early detection plays a crucial role in reducing mortality rates and saving lives. In the past several decades, there has been a rise in the adoption of deep learning algorithms, capturing the interest of researchers working in this field. One popular method involves utilizing pre-trained deep neural networks. In this study, a hybrid approach is employed to extract features from melanoma images. This approach integrates the utilization of pre-trained architectures, including AlexNet, ResNet-50, and GoogleNet. During the transfer training phase, these networks are fine-tuned to detect skin cancer by adjusting the learning rate. Subsequently, the maximum relevance minimum redundancy (MRMR) algorithm is employed to select optimal features based on the concepts of minimum redundancy and maximum relevance in order to minimize feature redundancy and enhance classification accuracy. The bagging technique is employed for the classification of various skin cancer types. The experimental results demonstrate the success of the suggested approach, yielding 98.9% accuracy. Furthermore, the results indicate the superiority of this method according to precision, recall, and F1-score in comparison with existing algorithms
Autism spectrum disorder classification using machine learning with factor analysis
Get PDFDue to the complexity and heterogeneity of autism spectrum disorder (ASD), diagnosis and categorization have attracted a lot of interest. To improve the robustness of ASD classification across the toddler age group, this work proposes an integrated strategy that integrates machine learning approaches with factor analysis and correlation validation. Benchmark dataset representing toddlers used to test this strategy’s efficiency. To first find the latent variables behind the ASD features in each dataset, factor analysis is used. We intend to capture the shared variance between variables and lower the dimensionality of the initial feature space by identifying these latent components. The subsequent machine-learning classification models used the retrieved components as input features. To validate the categorization results, correlation analyses were carried out in addition to factor analysis. The associations between the latent components discovered by factor analysis and the diagnostic labels were examined using Pearson correlation, a measure of linear association. The results highlight the method’s potential to improve diagnostic precision and shed light on the intricate connections between characteristics and diagnostic labels on the autism spectrum for toddlers
Advancing precision in air quality forecasting through machine learning integration
Get PDFIn an era where environmental concerns are escalating, air quality forecasting emerges. Forecasting is a crucial tool for addressing the adverse impacts of pollution on public health and ecosystems. In urban centers like Bandar Lampung, economic activities intensify pollution levels. This condition leveraging advanced machine learning forecasting methods can significantly mitigate these effects. This study evaluates the precision of long short-term memory (LSTM) and Prophet methods in predicting air quality. This study utilizes data from January 12, 2022 to November 9, 2023. The results reveal a distinct advantage of the LSTM method over the Prophet. The LSTM method showcases superior accuracy across all evaluation metrics. Specifically, the LSTM method achieved an average root mean squared error (RMSE) of 5.38, mean absolute error (MAE) of 3.94, and mean absolute percentage error (MAPE) of 0.07. In contrast, the Prophet method recorded higher error rates, with an average RMSE of 18.48, MAE of 15.61, and MAPE of 0.25. These numbers underscore the LSTM method's robustness and reliability in forecasting air quality. The result highlights its potential as a pivotal resource for environmental monitoring and policymaking to safeguard public health and promote sustainable urban development
Ledger on internet of things: a blockchain framework for resource-constrained devices
Get PDFThe increasing use of resource-constrained devices such as the internet of things (IoT) in various applications has led to the need for an optimized blockchain framework for these devices. Blockchain-based IoT networks allow businesses to access and share IoT data within their organization without centralized authority. However, existing frameworks are not designed for IoT applications and lack features like decentralization, scalability, and network overhead. To overcome these limitations, a new blockchain framework is proposed: ledger on internet of things (LIoT), which has a new consensus-based leader election algorithm to address the challenges of existing algorithms with high block creation time and communication overhead. Moreover, a novel data structure has been developed to reduce the storage size of the ledger effectively. The proposed framework also employs a docker for deployment, which provides an efficient and easy setup of blockchain nodes without requiring the individual configuration of each machine, increases the efficiency of the consensus process, and enables convenient deployment and management of the blockchain framework on resource-constrained devices. Furthermore, the performance of the proposed consensus method is analyzed using various performance parameters, including CPU usage, memory usage, transaction execution time, and block generation time
Effectiveness of artificial intelligence-driven chatbot responses in diabetes knowledge: a readability and reliability assessment
Get PDFPatient education is vital in diabetes management, empowering patients with necessary knowledge and skills to manage condition effectively. However, traditional educational methods often face challenges such as limited access to healthcare professionals and variability in information quality. This study aimed to assess the reliability and readability of artificial intelligence (AI)-driven chatbot responses in disseminating diabetes knowledge. Technically, the diabetes knowledge questionnaire (DKQ-24) was administered to evaluate the effectiveness of AI-driven chatbot in disseminating diabetes-related information. Responses were evaluated for reliability and quality applying the modified DISCERN (mDISCERN) scale and global quality scale (GQS), and readability was assessed using the Flesch reading ease (FRE) score, Flesch-Kincaid grade level (FKGL), gunning fog index (GFI), Coleman-Liau index (CLI), and simple measure of gobbledygook (SMOG). The mean mDISCERN score was 31.50±2.89, indicating generally reliable responses. The median GQS score was 4, reflecting the high overall quality. The readability assessment revealed a mean FRE score of 66.30, indicating that the text was fairly easy to read. FKGL mean score was 6.54±3.19, suggesting the text was suitable for readers at a sixth-grade level. In conclusion, AI-driven chatbot provides reliable and high-quality information on the diabetes self-management, but it requires improvements to enhance accessibility