Accurately classifying medical images with multiple labels is essential for early disease detection and enhancing clinical decision-making. In contrast to singlelabel classification, multi- label approaches allow for the simultaneous identification of multiple co- existing pathologies in a single image. Deep learning approaches, including convolutional neural networks and transformer- based models, have shown promising results, but they often suffer from high computational costs and lack of explainability, making them impractical for many medical applications. To address these challenges, this study introduces a novel lightweight transformer- based neural network optimized for multi- label medical image classification, reducing computational complexity while preserving strong feature extraction capabilities. Evaluations on the ChestX- ray11 dataset show superior classification accuracy and computational efficiency compared to existing methods. Furthermore, Grad- CAM++ visualizations enhance interpretability by highlighting disease- relevant regions, fostering trust in medical AI applications.

This study explores advanced techniques in book recommendation systems, which are integral components of contemporary online retail and e-commerce platforms. Traditional recommendation models have largely utilized algorithms such as K-nearest neighbors and cosine similarity. While effective to a certain extent, these methods often fail to generate sufficiently personalized and context-aware suggestions. To address these limitations, we introduce a hybrid recommendation framework that combines singular value decomposition (SVD) with cosine similarity, incorporating both content-based filtering and collaborative filtering strategies. Cosine similarity serves to identify items with similar user rating patterns; however, it does not account for latent variables that may influence user preferences. By integrating SVD-based matrix factorization, the proposed approach captures these hidden factors, offering a more nuanced understanding of user–item interactions. The system’s effectiveness is assessed using standard evaluation metrics, including precision, recall, normalized mean absolute error, root-mean-squared error, and mean absolute error. Experimental results indicate that approximately 80% of the top-k recommendations are relevant, with a precision score of 0.80. Overall, the findings suggest that hybrid models combining SVD with cosine similarity significantly enhance recommendation accuracy compared to approaches that rely solely on similarity measures. Beyond book recommendations, this framework can be extended to domains such as movies, music, and product recommendations, thereby contributing to the advancement of personalized and user-centric recommendation systems.

The rapid spread of deepfake videos poses significant challenges to the credibility of digital media, raising concerns over pri- vacy, misinformation, and trustworthiness. This research introduces a hybrid model combining Long Short-Term Memory (LSTM) networks and Convolutional Neural Networks (CNNs) to enhance deepfake detection. By leveraging ResNeXt-50 for extracting relevant features and LSTMs for capturing frame-to-frame dependencies, the proposed architecture effectively detects altered facial features in videos. Key preprocessing techniques, including face detection, extraction, and segmentation, optimize input data by isolating relevant facial regions. Experimental results demonstrate that this approach outperforms current methods in identifying subtle deepfake artifacts, underscoring the need for robust detection mechanisms to protect the credibility of digital media. Future work will explore improved scalability and real-time applications of this technique.

Effective disease diagnosis is a critical unmet need on a global scale. The intricacies of the numerous disease mechanisms and underlying symptoms make developing a model for early diagnosis and effective treatment extremely difficult. Machine learning (ML) can help to solve some of these issues. Recently, various ensemble-based ML models have benefited clinicians in early diagnosis. However, one of the most difficult challenges in multi-level ensemble approaches is the classifier selection and their placement in the ensemble framework as it improves the overall performance. Let m classifiers have to select from n classifiers there are (Formula presented.) ways. Again, these (Formula presented.) possibilities can be arranged in (Formula presented.) ways. Finding the best m classifiers and their positions from total (Formula presented.) ways is a challenging and hard problem. To address this challenge, a dynamic three-level ensemble framework is proposed. A nested Genetic Algorithm (GA) and ensemble-based fitness function are employed to optimize the classifier selection and their placement in a three-level ensemble framework. Our approach used eleven classifiers and chose seven classifiers by maximizing the fitness function. The proposed model experiments on 12 disease datasets. The proposed model outperformed in terms of accuracy, F1, and G-measure on the Chronic Kidney Disease (CKD) dataset is 0.987, 0.988, and 0.989, respectively. In terms of AUC on the Heart disease dataset (HDD) is 0.998 and in terms of recall on the Hypothyroid disease dataset (HyDD) is 0.988. In addition, the proposed model superiority is statically evaluated by Wilcoxon-Signed-Rank (WSR) test compared with other ensemble models, such as random forest (RF), bagging classifier (BC), XGBoost (XGB), and gradient boost classifier (GBC) with probability value p < 0.05 results shows all the traditional ensemble model differs with proposed model and also effective size evaluated with using the matched-pairs rank biserial correlation coefficient wc and statistical results shows effective size is large with RF and BC and effective size is medium with XGB and GBC. Proposed model has outperformed comparing with State-Of-The-Art (SOTA) ensemble and non-ensemble models. Further, the proposed model outperformed in terms of the ROC curve in the majority of the disease datasets. The results suggest the usage of the proposed model for disease diagnosis applications.

The accurate classification of vegetables based on image data is a critical task with significant implications for agricultural au- tomation, supply chain management, and consumer applications. However, this task is fraught with challenges due to the inherent variability in vegetable size, shape, color, and texture, which complicates the development of robust classification models. To ad- dress these challenges, this study proposes a Convolutional Neural Network (CNN) tailored for vegetable classification across 15 categories. The model leverages a dataset of 21,000 images, incorporating advanced techniques to enhance feature extraction and generalization. The proposed CNN is evaluated using metrics such as accuracy, precision, F1-score and recall. Experimental re- sults indicate that the model achieves high performance across all metrics, demonstrating its potential for integration into automated sorting systems and mobile applications for farmers. This work not only advances the state-of-the-art in vegetable classification but also highlights the societal benefits of improving accuracy in agricultural technologies.

The precise categorization of medical cases is crucial in the field of disease diagnosis. Traditional machine learning techniques, such as ensemble learning with bagging, have shown promising results in this domain. However, the performance of these methods heavily relies on the quality of the bags, i.e., the instances selected for training the base classifiers. In order to overcome this drawback, we provide a brand-new hybrid strategy that optimizes the bag composition by fusing a genetic algorithm (GA) with teaching-learning-based optimization (TLBO). The TLBO algorithm then optimizes the bags’ composition by iteratively selecting the best bags based on fitness and in the learning phase of TLBO it improves the worst performing bag through hybrid optimization. In this study, dynamic bag size has been used for varied subset creation, which minimized overfitting and enhanced adaptability. A distinctive fitness function that balances accuracy and diversity has also been proposed. In this process, a set of base classifiers are trained on the instances within the bags. The ensemble accuracy is evaluated using a voting scheme. The proposed hybrid approach was evaluated on a real-world dataset from UCI repository for disease diagnosis and its performance was compared to the traditional bagging method. The proposed approach outperforms the most advanced ensemble model, and statistical evidence indicates that it differs significantly from baseline models.

Ensemble learning leverages the diversity among models to mitigate overfitting and underfitting issues. Insufficient diversity can lead to misclassification due to overfitting, while excessive diversity may yield random predictions from inaccurately performing models. Conventionally, ensembles train distinct models on the same dataset and aggregate their predictions. In our approach, we introduce a partitioning of the training dataset into k clusters, each containing related data. Through iterations, we randomly sample data from each cluster, combining them to create a new dataset called "ns,"which is used to train a model. After N iterations, an ensemble is constructed by combining the N trained models. Our proposed approach emphasizes the importance of diversity while addressing overfitting and underfitting concerns in ensemble learning. Experimental results validate the effectiveness of this methodology, highlighting its potential for improving ensemble performance.

Counterfeit detection in banknotes remains a significant challenge, given the advanced techniques employed by counterfeits. Many existing solutions are either in accessible to the general public or lack the robustness required for reliable authentication. To overcome these limitations, this study proposes a web-based system for bank note verification, integrating machine learning and image processing. The system allows users to upload images of banknotes through a user-friendly interface designed with responsive web technologies, while backend operations are managed using Django. Image preprocessing methods, including Gaussian blurring, normalization, and Sobel edge detection, are applied to enhance visual quality and extract essential statistical features such as entropy, variance, skewness, and kurtosis. These features serve as inputs to a logistic regression model that classifies banknotes as authentic or counterfeit. Experimental results reveal that the proposed system achieves high accuracy on a balanced dataset. Additionally, comparative analysis with other machine learning classifiers shows that the system out performs existing state-of-the-art models, offering are liable solution for practical use.

Ensemble learning becomes a backbone in disease diagnosis using several classifiers to ensure improved prediction accuracy and also model reliability. However, conventional ensemble techniques often suffer some critical challenges, like poor diversity among base models, less efficient convergence, and sometimes high computational costs. That is why addressing these matters is essential to make further strides in ensemble-based diagnostic frameworks. This study introduces the Adaptive Ensemble Optimization with Hybrid Learning (AE HL) as an Novel Bagging Approach with Teaching-Learning-Based Optimization (BA-TLBO). The AE-HL framework encompasses a new fitness function that uses a new diversity metric with the Hamming distance to optimize both accuracy and classifier diversity effectively. To counteract inefficiencies in convergence, AE-HL uses adaptive optimization strategy that learns to balance exploration and exploitation during the learning phase. A multi-phase An optimization technique is employed, that limits the amount of computation by successively refining the best promising configurations; dynamic bag size adaptations improve the trade-off between variance and bias and, hence generalization over different datasets. Furthermore, the approach is integrated with a lightweight Explainable AI (XAI) module in order to support interpretability without an increase in complexity. The method is tested on several benchmark datasets for disease diagnosis where it is shown that AE-HL outperformed best among several ensemble optimization techniques. In summary, the proposed method obtained the highest accuracy with explainability and diversity in comparison with advanced metrics and statistical analysis. These results confirm the robustness, efficiency, and transparency of the AE-HL as a solution for enhancing systems for disease diagnosis.

Accurate and efficient disease diagnosis plays a crucial role in healthcare. This study proposes a novel approach to enhance disease diagnosis performance by combining bagging and Teaching-Learning-Based Optimization (TLBO). The objective is to develop an optimized ensemble model that leverages the strengths of multiple base classifiers to improve diagnostic accuracy. The proposed methodology involves several key steps. TLBO optimization process is employed to dynamically select the most informative instances (bags) from the training data. The optimization process iteratively refines the bags by considering the fitness of each ensemble model constructed using different base classifiers. The soundness of an ensemble model is evaluated based on its accuracy in predicting the target variable. To further enhance the performance of the base classifiers, hyperparameter tuning using grid search is incorporated into the model training process. This ensures that each base classifier is optimized with the best set of hyperparameters, leading to more accurate predictions. The optimized bags are then used to train the base classifiers, which include Decision Tree (DT), K-Nearest Neighbors (KNN), Support Vector Machines (SVM), and Gaussian Naive Bayes (GNB). The base classifiers are combined into ensemble models using a Voting Classifier, allowing them to complement each other's strengths and improve overall prediction performance. The results indicate that the TLBO-optimized ensemble models outperform individual base classifiers and traditional ensemble methods. The diversity among classifiers plays a crucial role in influencing the performance of ensemble models, especially in the context of disease diagnosis. Evaluating dissimilarity measures between classifiers becomes a key strategy in disease diagnosis. The dynamically selected bags contribute to improved accuracy, as they contain the most relevant instances for disease diagnosis and also explore computation time and diversity proposed ensemble approach on disease datasets.

Ensemble learning has emerged as a powerful technique for improving classification accuracy by combining multiple base models. This study presents an innovative approach to enhance ensemble learning through diversification. The proposed method integrates bagging, a resampling technique, with teaching–learning-based optimization (TLBO), and incorporates a pairwise dissimilarity measure to promote diversity within the ensemble. The TLBO algorithm optimizes the composition of the ensemble by iteratively selecting optimal bags of instances from the training data. The diversity measure quantifies the dissimilarity between bags, ensuring that the ensemble consists of diverse and complementary models. Our proposed model experimented on four benchmarked disease datasets and experimental results demonstrate that the proposed approach achieves superior performance compared to traditional ensemble methods. The ensemble models generated through this approach exhibit improved performance. The proposed model is statistically evaluated using the statistically paired T-test, and the results show our proposed model differs from base models.

Ensemble learning has emerged as a powerful approach for disease diagnosis, combining multiple classifiers to enhance predictive accuracy and robustness. Nevertheless, the challenge lies in selecting an optimal ensemble configuration while balancing accuracy and diversity. This study introduces a Bagging Approach with Teaching-Learning-Based Optimization (BA-TLBO) algorithm for ensemble optimization in disease diagnosis. To strike a balance between accuracy and diversity, a novel fitness function is proposed. This function incorporates ensemble mean accuracy and mean diversity, utilizing the Hamming distance as a measure of diversity. Additionally, dynamic weight updating is suggested to optimize weights over iterations in the BA-TLBO optimization process, thereby balancing exploration and exploitation. The use of a dynamic bag size over iterations aims to balance bias and variance, thereby enhancing generalization. The BA-TLBO explores different classifier combinations iteratively by selecting and replacing bags within the ensemble. This process aims to achieve high accuracy while also maintaining diversity. The effectiveness of the proposed approach is tested on four benchmark disease diagnosis datasets using multiple classifiers, including Logistic Regression (LR), K-Nearest Neighbors (KNN), Decision Trees (DT), and Support Vector Machines (SVM). The model's performance is compared using diversity metrics, including Entropy, Bhattacharya distance, and Q statistics. Results indicate the superiority of the proposed model over alternative approaches. Furthermore, the robustness of the proposed model is compared with other meta-heuristic optimization algorithms, such as Artificial Bee Colony (ABC), Ant Colony Optimization (ACO), Firefly Optimization (FO), and Particle Swarm Optimization (PSO). The evidence suggests that the proposed model performs better in the majority of cases, particularly in 5-bag and 10-bag configurations. The proposed approach is evaluated using both 5-bag and 10-bag configurations, considering both worst-case and best-case bag optimization strategies. Experimental results demonstrate that the BA-TLBO-based model outperforms both state-of-the-art (SOTA) ensemble and non-ensemble models.

In this study, we present an novel approach to enhance the predictive performance of ensemble-based machine learning models for early disease diagnosis. We introduce a novel ensemble model incorporating Grey Wolf Optimization (GWO) based feature selection and a newly designed fitness function emphasizing specificity and sensitivity. The effectiveness of our proposed model is validated using five disease datasets from the UCI machine learning repository: Chronic Kidney Disease (CKD), Statlog Heart Disease (SHD), Cleveland Heart Disease (CHD), Pima Indian Diabetes (PID), and Wisconsin Breast Cancer (WBC). Our proposed model surpasses State-of-the-Art (SOTA) ensemble and non-ensemble models in terms of Accuracy, Sensitivity, Specificity, and AUC. Additionally, a Paired T-Test with 95% confidence confirms the significant superiority of our model over previous base and ensemble models. This research showcases a promising step forward in leveraging machine learning for accurate and early disease diagnosis.

Disease diagnosis is very important in the medical field. It is essential to diagnose chronic diseases such as diabetes, heart disease, cancer, and kidney diseases in the early stage. In recent times, ensembled-based approaches giving effective predictive performance than individual classifiers and gained attention in assisting doctors with early diagnosis. But one of the challenges in these approaches is dealing with class-imbalanced data and improper configuration of ensemble classifiers with optimized parameters. In this paper, a novel 3-level stacking approach with ADASYN oversampling technique with PSO Optimized SVM meta-model (Stacked-ADASYN-PSO) is proposed. Our proposed Stacked-ADASYN-PSO model uses base models such as Logistic regression(LR), K-Nearest neighbor (KNN), Support Vector Machine (SVM), Decision Tree (DT), and Multi-Layer Perceptron (MLP) in layer-0. In layer-1 three meta classifiers namely LR, KNN, and Bagging DT are used. In layer-2 PSO optimized SVM used as the final meta-model to combine the previous layer predictions. To evaluate the robustness of the proposed model It is tested on five benchmark disease datasets from the UCI machine learning repository. These results are compared with state-of-the-art ensemble models and non-ensemble models. Results demonstrated that the proposed model performance is superior in terms of AUC, accuracy, specificity, and precision. We have performed statistical analysis using paired T-tests with a 95% confidence level and our proposed stacking model is significantly differs when compared to base classifiers.

Cancer is deadly diseases still exist with a lot of subtypes which makes lot of challenges in a biomedical research. The data available of gene expression with relevant gene selection with eliminating redundant genes is challenging for role of classifiers. The availability of multiple scopes of gene expression data is curse, the selection of gene is play vital role for refining gene expression data classification performance. The major role of this article is to derive a heuristic approach to pick the highly relevant genes in gene expression data for the cancer therapy. This article demonstrates a modified bio-inspired algorithm namely cuckoo search with crossover (CSC) for choosing genes from technology of micro array that are able to classify numerous cancer sub-types with extraordinary accuracy. The experiment results are done with five benchmark cancer gene expression datasets. The results depict that CSC is outperforms than CS and other well-known approaches. It returns 99% accuracy in a classification for the dataset namely prostate, lung and lymphoma for top 200 genes. Leukemia and colon dataset CSC is 96.98% and 98.54% respectively.

Lung cancer is one of the most prevalent cancer-related diseases with a high mortality rate, and this is largely due to the lateness in detecting the presence of malignancy. Again, the conventional methods used in the diagnosis of lung cancer have had their shortfalls. While the effectiveness of computerized tomography in detecting this malignancy, the large volumes of data that radiologists have to process not only present an arduous task but may also slow down the process of detecting lung cancer early enough for treatment to take its course. It is against this backdrop that computer-Aided diagnostic (CAD) systems have been designed. One of such is the convolutional neural network, a method that best describes a group of deep learning models featuring filters that can be trained with local pooling operations being incorporated on input CT images in an alternating manner to create an array of hierarchical complex features. The need to have this type of data-driven technique is further informed by the attempt to ensure successful segmentation of lung nodules, a step that cannot be overruled when striving for a good model of detection or diagnosis. There are variations and models of the convolutional neural networks that have been effectively put to use in the lung nodule detection. The 2D CNN model has been utilized in the medical field for quite a while now, and as it has displayed its many strengths, so could the limitations not be hidden. It is in addressing these limitations and improving on the detection prowess of the convolutional neural network that the 3D model is now fast gaining traction. The 3D models have been reported to return pronounced sensitivity and specificity in detection of lung nodules, but the issues of time-consumption, training complexities and hardware memory usage could make it difficult to implement the 3D model in the medical field. In this paper, review the advances that have been made in the area of adopting 3D CNN model in the diagnosis of lung cancer.