Cyberbullying involves the use of social media platforms to harm or humiliate people online. Victims may resort to self-harm due to the abuse they experience on these platforms, where users can remain anonymous and spread malicious content. This highlights an urgent need for efficient systems to identify and classify cyberbullying. Many researchers have approached this problem using various methods such as binary and multi-class classification, focusing on text, image, or multi-modal data. While deep learning has advanced cyberbullying detection and classification, the multi-class classification of cyberbullying using multi-modal data, such as memes, remains underexplored. This paper addresses this gap by proposing several multi-modal hybrid deep learning models, such as LSTM+ResNet, LSTM+CNN, LSTM+ViT, GRU+ResNet, GRU+CNN, GRU+ViT, BERT+ResNet, BERT+CNN, BERT+ViT, DistilBERT+ResNet, DistilBERT+CNN, DistilBERT+ViT, RoBERTa+ResNet, RoBERTa+CNN, and RoBERTa+ViT, for classifying multi-classes of cyberbullying. The proposed model incorporates a late fusion process, combining the LSTM, GRU, BERT, DistilBERT, and RoBERTa models for text extraction and the ResNet, CNN, and ViT models for image extraction. These models are trained on two datasets: a private dataset, collected from various social media platforms, and a public dataset, obtained from previously published research. Our experimental results demonstrate that the RoBERTa+ViT model achieves an accuracy of 99.20% and an F1-score of 0.992 on the public dataset, and an accuracy of 96.10% and an F1-score of 0.959 on the private dataset when compared with other hybrid models.

Air pollution poses a significant threat to human health and the environment, making accurate air quality assessment crucial. In this study, we propose a deep learning-based approach for multi-class classification of air quality levels using image data. We evaluate and compare the performance of four prominent models, Convolutional Neural Network (CNN), ResNet-50, Vision Transformer (ViT), and Swin Transformer, using the Smartphone-Based Air Pollution Image Dataset (SAPID). Additionally, we assess two traditional machine learning models, Support Vector Machine (SVM) and Random Forest classifier, for baseline comparison. Each model was trained to classify images into five air quality categories. Among the models, ViT achieved the highest accuracy of 97%, followed closely by Swin Transformer at 96%, both significantly outperforming the traditional CNN and ResNet-50 models, which attained accuracies of 78% and 64%, respectively. The traditional classifiers, SVM and Random Forest, achieved accuracies of 88% and 91%, indicating lower performance compared to the ViT and Swin Transformer models. These results demonstrate the superior generalization capabilities of transformer-based models in handling complex visual patterns associated with varying pollution levels. This work highlights the effectiveness of vision transformers in environmental monitoring and offers a promising direction for scalable, image-based air quality assessment systems.

Cyberbullying is defined as the use of social media platforms to hurt or humiliate people online. The anonymity on these platforms makes it easy to spread hurtful content, sometimes leading victims to self-harm. This highlights the urgent need for efficient methods to identify to prevent cyberbullying. Numerous studies have addressed this issue by focusing on cyberbullying classification, primarily through binary classification using multi-modal data or multi-class approaches targeting either text or image data. While deep-learning advancements have improved cyberbullying identification, a gap remains in the multi-class classification of cyberbullying utilizing multi-modal data such as memes and this research aims to bridge this gap by accurately classifying cyberbullying across multiple data modalities through a hybrid deep-learning model that combines RoBERTa for text extraction and Vision Transformer (ViT) for images extraction as hybrid(RoBERTa+ViT) model using late fusion module process. Two datasets were utilized: a Private-dataset was collected from comments on social media videos and a Public-dataset that was downloaded from existing research. The hybrid model was trained on these two datasets and the model demonstrated notable performance, achieving an an accuracy of 99.24% on the Public-dataset and 96.1% on the Private-dataset, with F1-scores of 0.9924 and 0.9599, respectively.

Rice (Oryza sativa L.) is a staple food for over half of the global population, with significant economic, agricultural, and cultural importance, particularly in Asia. Thousands of rice varieties exist worldwide, differing in size, shape, color, and texture, making accurate classification essential for quality control, breeding programs, and authenticity verification in trade and research. Traditional manual identification of rice varieties is time-consuming, error-prone, and heavily reliant on expert knowledge. Deep learning provides an efficient alternative by automatically extracting discriminative features from rice grain images for precise classification. While prior studies have primarily employed deep learning models such as CNN, VGG, InceptionV3, MobileNet, and DenseNet201, transformer-based models remain underexplored for rice variety classification. This study addresses this gap by applying two deep learning models such as Swin Transformer and Vision Transformer for multi-class classification of rice varieties using the publicly available PRBD dataset from Bangladesh. Experimental results demonstrate that the ViT model achieved an accuracy of 99.86% with precision, recall, and F1-score all at 0.9986, while the Swin Transformer model obtained an accuracy of 99.44% with a precision of 0.9944, recall of 0.9944, and F1-score of 0.9943. These results highlight the effectiveness of transformer-based models for high-accuracy rice variety classification.