Digital misinformation presents a dual challenge: achieving high detection accuracy while ensuring interpretability. This paper introduces X-FRAME (Explainable FRAMing Engine), a hybrid framework that combines semantic representations from XLM-RoBERTa with theory-informed features related to psycholinguistic framing, source credibility, and social context. Unlike fact-checking systems that verify claims directly, X-FRAME detects linguistic, contextual, and stylistic indicators statistically associated with misinformation. Evaluated across eight publicly available datasets totaling $\mathrm{286,260}$ samples, X-FRAME achieves $86\%$ accuracy and $81\%$ recall on the minority Fake class, significantly outperforming text-only and features-only baselines. The model demonstrates cross-domain adaptability potential, attaining $97\%$ accuracy on formal news articles and $72\%$ on social media content. Importantly, X-FRAME provides transparent, human-understandable rationales via Local Interpretable Model-agnostic Explanations (LIME) and Permutation Importance, anchoring predictions in interpretable features such as sensationalism and speaker credibility. This work advances misinformation detection by unifying high performance with explainability and cross-domain adaptability.

Over the past few decades, research on object detection has developed rapidly, one of which can be seen in the fashion industry. Fast and accurate detection of an E-commerce fashion product is crucial to choosing the appropriate category. Nowadays, both new and second-hand clothing is provided by E-commerce sites for purchase. Therefore, when categorizing fashion clothing, it is essential to categorize it precisely, regardless of the cluttered background. We present recently acquired tiny product images with various resolutions, sizes, and positions datasets from the Shopee E-commerce (Thailand) website. This paper also proposes the Fashion Category—You Only Look Once version 4 model called FC-YOLOv4 for detecting multiclass fashion products. We used the semi-supervised learning approach to reduce image labeling time, and the number of resulting images is then increased through image augmentation. This approach results in reasonable Average Precision (AP), Mean Average Precision (mAP), True or False Positive (TP/FP), Recall, Intersection over Union (IoU), and reliable object detection. According to experimental findings, our model increases the mAP by 0.07 percent and 40.2 percent increment compared to the original YOLOv4 and YOLOv3. Experimental findings from our FC-YOLOv4 model demonstrate that it can effectively provide accurate fashion category detection for properly captured and clutter images compared to the YOLOv4 and YOLOv3 models.

Detection and classification have an essential role in the world of e-commerce applications. The recommendation method that is commonly used is based on information text attached to a product. This results in several recommendation errors caused by invalid text information. In this study, we propose the development of a fashion category (FC-YOLOv4) model in providing category recommendations to sellers based on fashion accessory images. The resulting model was then compared to YOLOv3 and YOLOv4 on mobile devices. The dataset we use is a collection of 13,689, which consists of five fashion categories and five accessories' categories. Accuracy and speed analysis were performed by looking at mean average precision (mAP) values, intersection over union (IoU), model size, loading time, average RAM usage, and maximum RAM usage. From the experimental results, an increase in mAP was obtained by 99.84% and an IoU of 88.49 when compared to YOLOv3 and YOLOv4. Based on these results, it can be seen that the models we propose can accurately identify fashion and accessories categories. The main advantage of this paper lies in i) providing a model with a high level of accuracy and ii) the experimental results presented on a smartphone.

Hate speech detection research has mainly focused on monolingual contexts, with limited exploration of multilingual and code-switched languages that introduce distinct linguistic complexities. This study examines hate speech detection in code-switched Spanglish content from social media, comparing transformer-based models—XLM-RoBERTa, DistilBERT, Multilingual BERT, and mT5—with traditional machine learning approaches, including support vector machines, logistic regression, and multinomial naïve Bayes using TF-IDF features. The results indicate that XLM-RoBERTa achieves the highest performance, with 96.14 percent accuracy, 96.16 percent precision, 96.14 percent recall, and a 96.12 percent F1-score, demonstrating its superiority in detecting code-switched hate speech. Although traditional models, particularly SVM (94.03 percent accuracy), perform well, transformer-based approaches offer clear advantages in multilingual contexts.

Accurate power-system fault diagnosis is difficult when different fault types exhibit similar electrical signatures and deterministic classifiers cannot adequately represent diagnostic uncertainty. This study proposes an uncertainty-aware framework integrating machine learning (ML) and Bayesian-network (BN) reasoning using 1,021 synthetic events across six fault classes. Random Forest, XGBoost, SVM, and ANN were evaluated, with RootCause excluded from ML inputs and inferred only by the BN to avoid information leakage. The selected RF model achieved 53.66% accuracy and 58.92% Macro F1, while BN-only and hybrid RF+BN each achieved 60.00% accuracy. In an illustrative case, Bayesian updating shifted the leading diagnosis from LLG to LG (51.2%) and identified Vegetation as the most probable root cause (33.2%). In addition to classification performance, the three diagnostic routes were evaluated using calibration and uncertainty measures, including log loss, Brier score, ECE, and Top-2 accuracy. The results show that the main benefit of Bayesian reasoning lies in preserving competing fault hypotheses and providing probabilistic root-cause information rather than uniformly improving every performance metric. The framework therefore extends conventional fault classification toward more interpretable and uncertainty-aware diagnostic decision support, although further validation using real utility data or standard power-system test environments is required before operational deployment.