IAES International Journal of Artificial Intelligence (IJ-AI)
Not a member yet
1769 research outputs found
Sort by
Heart disease approach using modified random forest and particle swarm optimization
Get PDFFor the past two decades, heart disease has been classified as one of the main causes of mortality globally. Fortunately, most researchers focused on data mining techniques, which play an important role in accurately predicting heart disease to develop their models. In this paper, by combining particle swarm optimization (PSO) and modified random forest (MRF), a new approach (PSO-MRF) is proposed to predict heart disease. The main purpose is to select the important features after the bootstrap method for each decision tree in the random forest, and then optimize the MRF by the PSO algorithm. The experiments are carried out using the publicly accessible UCI heart disease datasets. Thorough experimental analysis demonstrates that our approach has outperformed the random forest algorithm as well as many other classifiers. This model helps doctors and researchers improve the diagnosis and treatment of heart disease, resulting in more prompt, accurate patient care
A revolutionary convolutional neural network architecture for more accurate lung cancer classification
Get PDFThis research aimed to investigate a breakthrough in convolutional neural network (CNN) architecture with the potential to revolutionize lung cancer classification. The proposed method is a comparative optimization model of ResNet architecture, with accuracy rate of 99.68% in identifying and categorizing lung cancer types. The results showed that the use of innovative methods in CNN architecture, such as multi-dimensional convolutional layers and the integration of specific lung cancer features, effectively provided highly accurate and reliable outcomes. These showed a positive impact on the development of medical diagnostic technology, offering promising potential to enhance prognosis and response to treatment for lung cancer patients. With high accuracy rate, this breakthrough presents opportunities for further advancements in lung cancer management through artificial intelligence-based methods
Metaheuristic optimization for sarcasm detection in social media with embedding and padding techniques
Get PDFSarcasm is a sophisticated mode of expression that allows speakers to express their opinions subtly. Stakeholders provide unstructured messages with extended phrases, making it difficult for computers and people to understand. This research aims to develop a sarcasm detection method to identify words in phrases as sarcastic or non-sarcastic from text, utilizing natural language processing appliances. The first step is pre-processing, when the padding and embedding are performed. Zero padding and end padding are used for the padding. At the same time, different embedding techniques, such as word2vec, Glove, and BERT, are used. Following pre-processing, the features are extracted from the pre-processed data, including "information gain, chi-square, mutual information, and symmetrical uncertainty-based features." Then, a hybrid optimization technique known as clan-updated grey wolf optimization (CU-GWO) is used for optimized features and weight selection. An ensemble technique was applied to extract optimal features. The classifiers in the proposed suggested ensemble technique with deep convolution neural network (DCNN). DCNN offers fine weight tuning and detection results.The performance analysis and its impact on the proposed model for sarcasm detection are classified with good accuracy into sarcastic and non-sarcastic categories. The results are also compared with against those of the GloVe and BERT techniques
Efficient lung disease classification through luminescent feature selection using firefly algorithm
Get PDFOver the past couple of decades, there has been a substantial increase in the prevalence of lung ailments, resulting in 3.5 million fatalities each year. This necessitates the adoption of a lung disease detection technology that is effective, trustworthy, and cost-effective. In this study, we propose an optimized convolutional neural network (CNN) model, used for multiclass categorization of lung ailments based on frontal chest X-rays. The classification includes four categories: COVID-19, viral pneumonia, lung opacity, and non-infectious normal group. We implemented the firefly algorithm to optimize the global efficiency of feature selection of the lung abnormality in the X-ray images of lung disease and COVID-19 to classify the input according to the target class. The proposed algorithm was tested for accuracy, precision, recall, and F1-score. The findings were validated using the transfer learning model VGG-16; the algorithm achieved a superior accuracy of 99.3% compared to that of other cutting-edge models such as Inceptionv3 and ResNet50
Optimized fault detection in bearings of rotating machines via batch normalization-integrated bidirectional gated recurrent unit networks
Get PDFMotor is commonly used in industrial applications. Although motors are frequently found to have bearing problems, this causes a serious safety risk to industrial production. Traditionally, fault diagnostics methods often required only signal processing techniques and are ineffective. To overcome this problem, deep learning (DL) has been recently developed rapidly and achieved remarkable results in fault diagnosis. The intelligent fault diagnosis and classification of rolling bearing faults based on ensemble empirical mode decomposition (EEMD) and batch normalization (BN), principal component analysis (PCA) based stacked bidirectional-gated recurrent unit (Bi-GRU) neural network, is proposed in this paper. BN is introduced to improve the fast convergence of gated recurrent unit (GRU). EEMD is applied to eliminate the noise interference from the vibrational signal, and then important features are selected using the correlation coefficient value. Next, PCA is utilized for dimensionality reduction to retain only the essential. Finally, the BN based stacked Bi-GRU model is developed to classify faults based on extracted features. The proposed model correctly classifies the different types of faults in real operating conditions and also compared with existing techniques
Optimized data security and storage using improved blowfish and modular encryption in cloud-based internet of things
Get PDFThe increasing development of the internet of things (IoT) has made cloud-based storage systems essential for storing, processing, and sharing IoT data. Ensuring cloud security is crucial as it manages a large volume of sensitive and outsourced data vulnerable to unauthorized access. This research proposes an improved blowfish algorithm and modular encryption standard (IBA-MES) for secure and efficient data storage in cloud-based IoT systems. The block cipher structure in IBA enables scaling for different data sizes, ensuring secure data handling across a wide range of IoT devices. Additionally, IBA-MES adaptability helps maintain data integrity, enhancing both the security and efficiency of data storage in cloud-based IoT environments. Modular encryption standard (MES) reduces latency during encryption operations, ensuring quick data transactions between the cloud server and IoT devices. By combining blowfish’s speed and strength with modular encryption’s adaptability, IBA-MES provides robust data protection. Metrics such as execution time, central processing unit (CPU) usage, encryption time, decryption time, runtime, and latency are calculated for the proposed IBA-MES. For 700 blocks, the IBA-MES achieves encryption and decryption times of 270 and 415 ms, respectively, outperforming the triple data encryption standard (TDES)
MobileChiliNet: convolutional neural network for chili leaves classification
Get PDFChili pepper (Capsicum annuum) is an important crop in many countries, including Indonesia, which plays an important role in local economy and food production. To meet the high demand, effective agricultural management, especially the diagnosis and treatment of plant diseases, is essential. This study aims to improve the accuracy of chili leaf disease classification while reducing the computational cost so that it can be applied to low-cost smart farming systems. Through the development of the MobileChiliNet architecture, which is the result of pruning and fine-tuning of MobileNetV2, this model achieves the best accuracy, better than other CNNs such as ResNet50 and VGG16. Testing with various optimizers and learning rate schedulers shows that AdamW with PolynomialDecay provides the best performance by increasing the validation accuracy to 96.48%. This approach successfully reduces the computational complexity while maintaining high accuracy, so that it can be implemented in smart farming systems at a lower cost
A comprehensive impression on identifying plant diseases using machine learning and deep learning methodologies
Get PDFMaintaining healthy plants is essential for long-term agricultural production because agriculture is the backbone of many economies. Agricultural productivity is greatly endangered by plant diseases, which result in huge economic losses. Identifying plant diseases using traditional approaches can be quite laborious, time-consuming, and knowledge-intensive. Automated, precise, and quick diagnosis of plant diseases has been made possible by recent developments in artificial intelligence, mainly in deep learning, and machine learning. This study gives a thorough analysis of how machine learning and deep learning are currently being used to detect plant diseases. Methodologies, datasets, evaluation measures, and the inherent difficulties of the area are all examined. In order to better understand these technologies in practical agricultural contexts, this review will try to shed light on their advantages and disadvantages
Translation-based image steganography system utilizing autoencoder and CycleGAN
Get PDFTraditional image steganography involves embedding secret information into a cover image, a process that requires modification of the carrier and potentially leaves detectable marks. This paper proposes a novel method of coverless image steganography based on generative models. Initially, a CycleGAN model is constructed and trained to learn the features of different image domains. Subsequently, an Autoencoder model is trained using two sets of images to achieve a precise one-to-one mapping. Once the models are trained, the autoencoder is used on both the sender and receiver sides to convert the cover image (also known as the stego image) into the secret image and vice versa. The CycleGAN model is then utilized to enhance the visual quality of the images generated by the autoencoder. Experimental results demonstrate that this method not only effectively secures secret information transmission but also improves efficiency and increases the capacity for information hiding compared to similar methods
Unveiling precision: Eye cancer detection redefined with particle swarm optimization and genetic algorithms
Get PDFEye cancer detection is rare. The study introduces a holistic swarm intelligence method for the timely identification and categorization of three significant eye disorders: glaucoma, diabetic retinopathy, and cataract. Glaucoma is distinguished by elevated pressure within the eye and harm to the optic nerve, potentially leading to permanent loss of vision. Diabetic patients experience diabetic retinopathy primarily due to the presence of high blood sugar levels. The early detection and classification of cataracts can be achieved by combining swarm intelligence algorithms such as particle swarm optimization (PSO) and genetic algorithms (GA). In the case of diabetic retinopathy diagnosis, swarm intelligence is employed to optimize the parameters of deep learning models, thereby enhancing the accuracy of lesion segmentation and classification. Cataract detection used to improve the evaluation of lens opacity and cloudiness, providing a robust diagnostic mechanism. The accuracy obtained with a PSO is 85.79%, F1 score 83.45%, and recall 82.43%. The accuracy obtained with a GA is 82.10%, F1 score 81.16%, and recall 81.51%. The comparison of GA, convolution neural network, and PSO algorithms proves that the accuracy to detect the eye cancer is achieved with PSO and GA algorithm