This research discusses the use of machine learning and deep learning techniques for cancer detection and screening. The use of machine learning and deep learning algorithms enables survey specialists, including healthcare professionals, to increase the accuracy of risk assessments for diseases such as cancer, and hence identify the best time and techniques for screening. The use of these technologies enhances the accuracy of diagnosis while simultaneously aiding in the formulation of treatment regimens based on patients’ profiles. The Conventional approaches in cancer prevention and early detection sometimes act poorly as they sometimes do not cover many aspects of changes and variations of risks in people. Skin cancer is a prevailing issue, and the worldwide effects have amplified the importance of early and appropriate diagnosing. In previous years, the field of medical research has expanded through the use of machine learning and deep learning approaches, particularly in the diagnosis and classification of skin cancers. The importance of using machine learning and deep learning techniques in the early detection and prevention of cancer. The study emphasizes the importance of machine learning and deep learning approaches in the future struggle against cancer, which will necessitate the incorporation of early detection into normal operations. Due to the machine learning approach, multifaceted connections and relationships are found in high-dimensional data and hence sharper and individualized prophylactic measures. However, issues like quality of gathered data, interpretability of models, and of course, ethical issues are still there to hold back the progress of these technologies. In this regard, the present review seeks to provide detailed information regarding the current landscape of skin cancer research by showcasing the power of ML and DL models to the scholars, clinicians, and other healthcare professionals. This paper underscores the need for care and detection of cancer at an early stage as well as preventing the cancer illness. In particular, highlighting achievements along with unexplored opportunities for further development for this kind of research, this venture aims to introduce new advances in the field of skin cancer identification.

Purpose: This paper seeks to determine how AI can be used in the furtherment and management of distributed generation of renewable power, and how it can address challenges arising from the use of distributed generation of renewable energy sources like solar and wind. Thus, this research addresses a gap in the knowledge base related to AI technologies to plan and optimize the performance of decentralized energy systems, based on the assessment of opportunities, risks, and applications of the technology. Methodology: Convenience sampling was used and to incorporate mixed methods, the Research Onion of Saunders was applied to attain methodological credibility. Self-administered questionnaires were distributed to 210 professionals drawn from renewable energy and AI industries. The data was gathered by an online survey instrument and closed was done by employing basic descriptive statistics, chi-square tests, one-way ANOVA, sentiment analysis, clustering method, and multiple linear regression. Fortunately, the analysis also incorporated liberal graphical displays so that there was a better feel of the data distribution and association patterns. Results: The study reveals a generally positive perception of AI’s benefits in managing decentralized energy grids among respondents, who average 13.41 years of experience in renewable energy and 14.59 years in AI applications, though this experience varies significantly (SD = 8.84 and 7.91, respectively). Statistical analyses showed no significant correlation between respondents' familiarity with decentralized grids and their perceptions of AI's role. The Chi-Square Test for Independence (Chi-Square = 21.8, p = 0.15) and ANOVA (F = 1.17, p = 0.324) indicate no significant association between grid familiarity and AI perceptions, nor differences in AI experience across varying levels of grid familiarity. Sentiment analysis also confirmed these patterns when AI was only rated neutrally to slightly positively; the average score was 1. 0 illustrating people’s relative agreement about the potential of AI, even if they may not personally have a lot of experience with it. Primary data analysis by clustering showed no significant dissimilarities of response patterns indicative of the homogeneity of the sample. Furthermore, the result obtained from multiple linear regression (R² = 0. 01, p = 0. 329) indicates that both experience in the renewable and energy and experience in the use of AI have no significant impact towards the perception of the role of AI in distributed grids. Conclusion: AI has a high potential and applicability for decentralized energy grids but to achieve successful implementation certain technical, operational and policy issues need to be addresses. Hence, the study points out the required future research and stakeholder cooperation in order to explore all the possibilities of AI application in this field with the specific suggestions of the directions of future studies and interdisciplinary cooperation. Practical Implications: The study focuses on the shift in the use of artificial intelligence, in the operation of decentralized re-new able energy grids and offers guidance for grid operators, policymakers as well as technology developers. By drawing awareness of the advantage and drawbacks of AI implementation, the experts will be able to solve the issues related to the improvement of grid reliability and the stable integration of renewable sources into the global energy infrastructure. Originality/Value: Following this introduction, this paper aims to present a quantitative and qualitative analysis of decentralized renewable energy grids with consideration of AI with the following organization of the work. By applying Research Onion framework of Saunders, this research enhances the methodological robustness and can be helpful for major stake holders, including academics, business and or policymakers in the domain of AI as well as renewable energy.

This paper explores a framework for an ML model for solar yield estimation using random forest, gradient boosting, support vector machines, and neural networks There are various factors affecting solar energy production, some of which are weather conditions, solar intensity, temperature, among others, and characteristics of the panel. The trend for development and usage of renewable energy sources continues to gain momentum. The estimation of production rates for solar energy is crucial for the efficiency of energy generation, transmission, and supply. Hypothesis-driven and list-based models of traditional statistics are sometimes rather low in predictive validity because the relationships between these factors may be non-linear and highly interactive. The complex structure of these dependencies indicates that advanced ML algorithms are an appropriate approach to modeling the yield and increasing the utility of the resulting predictions. This work predictively analyzes historical solar generation data and climatic factors and preprocesses the data through normalization and feature engineering. The models are then adjusted and tested using a cross-validation approach in order to minimize the risk of obtaining high accuracy values only on the training data set. Evaluations of performance with mean absolute error, root mean square error, and coefficient of determination are used to measure the efficiency of each model. Machine learning algorithms widely outperform other models when applied to predict the yield of solar energy in the same or related areas. The random forests seem to perform better for prediction because they operate in an ensemble, which allows for modeling interactions between them. Through enhanced prediction precision, the approach proves to be a helpful resource for solar energy managers to attain better correlation between supply and demand, enhance solar energy production, and encourage the use of sustainable energy resources. Extension of this study in future work would involve the use of real-time data coupled with adaptive learning models to improve the prediction resilience.

Accurate classification of peripheral blood cell types from microscopy images can support automated hematology workflows by reducing manual review burden and improving consistency. However, fine-grained recognition remains challenging due to strong inter-class similarity, intra-class morphological variability, and confounding cytoplasmic texture patterns that blur decision boundaries among related cell types. We propose SwinDiNO-AFNet for eight-class classification of basophil, eosinophil, erythroblast, immature granulocyte (IG), lymphocyte, monocyte, neutrophil, and platelet. SwinDiNO-AFNet is a dual-path hybrid that fuses hierarchical window-based features from a Swin Transformer with global representations from a pretrained DINOv2 vision transformer used strictly as a feature extractor. The flattened Swin and pooled DINOv2 features are concatenated and refined through a channel attention module to obtain attention-weighted fusion before the final classifier. To ensure rigorous evaluation, the training pipeline enforces leakage control via perceptual-hash based near-duplicate removal before fold creation, with consistent RGB enforcement, 224 × 224 resizing, ImageNet normalization, and train-only augmentation and robustness noise applied exclusively within training folds. Under leakage-safe stratified fivefold cross-validation, SwinDiNO-AFNet achieves 98.61 ± 0.21% validation accuracy with 98.46% macro-F1, and statistical testing (Welch with Holm adjustment) supports improvements over multiple baseline backbones under the same protocol. Precision-recall analysis further indicates near-ceiling separability (micro-AUPRC 0.9979), while confusion matrix and Grad-CAM based error analysis show that remaining failures concentrate in morphologically ambiguous cases, including IG confusions toward granulocytic patterns. Overall, the results show that attention-guided fusion of complementary Swin and DINOv2 representations, combined with leakage-safe cross-validation and XAI-backed diagnostics, provides a strong solution for eight-class peripheral blood cell classification on the studied dataset.

Phishing remains a widespread cybersecurity attack, leveraging social engineering techniques to trick users and acquire sensitive information. Conventional detection mechanisms tend to fail to detect advanced phishing attacks, especially those that are built into misleading textual content. This paper offers a detailed investigation on the use of Natural Language Processing (NLP) combined with state-of-the-art machine learning and deep learning models for improved accuracy in detecting phishing. Phishing attack vectors are covered, and the essential significance of NLP in extracting contextual and semantic features from phishing content is emphasized. The study compares the performance of various classification models such as Naive Bayes, Support Vector Machines, and Random Forests with deep learning models such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Long Short-Term Memory (LSTM) networks. Experimental results indicate that deep learning models, namely CNNs and RNNs, have better performance, with accuracy of over 97% and always high recall and F1 values. The findings demonstrate the capability of NLP-driven deep learning structures to improve phishing detection systems and towards more secure cybersecurity infrastructure.

This work overviews how machine learning and deep learning have revolutionized the present era of computer science. Artificial intelligence has evolved greatly in modern computer science through machine learning and deep learning. It avails inducements in several fields, such as natural language processing, computer vision, and robotics. It is enhancing the effectiveness of data analysis itself. These technologies also presented new opportunities for creativity. It includes privacy of data, the bias inherent in algorithms, and the limited, or obscure, interpretability of most models. They included the development of algorithms that are more efficient, the issue of interpretability, and the consideration of ethical concerns in the future. It is useful to gain insights on the part that these technologies play in modern computing paradigm approaches It is implemented in business and industries and a glimpse of what these technologies are likely to become in the future. It includes supervised, unsupervised, and reinforcement learning methodologies that enable the model to be used in various investigations ranging from predictive modeling to intelligent task management. The process requires vast data, second, computation power, and third, the expertise of the right algorithm selection. The obscuration of some of the algorithms can make it challenging to explain results as questions about accountability and transparency emerge. It is expected that machine learning and deep learning will witness a significant leap, mostly owing to data availability as well as enhancements in computational capabilities. These technologies, as they become increasingly implemented within organizations, will enable better organizational decision-making while also pushing the boundaries of possibility within the field of computers. The continued advancement of machine learning and deep learning precipitate further development for more applications.