Cancel culture has emerged as a significant societal phenomenon, particularly in the context of consumer-brand relationships, where consumers leverage social media platforms to publicly sever ties with brands. This study explores cancel culture through a multi-method approach, integrating insights from survey data, qualitative research, and social media analysis. Leveraging advanced natural language processing (NLP) models, including transformer-based architectures, we analyze sentiment, emotional dynamics, and linguistic patterns underlying cancel behaviors. Our findings highlight the emotional distinction between canceling and boycotting, with sadness predominantly associated with canceling and anger driving boycotting behaviors. By employing state-of-the-art AI-driven tools to quantify and examine these behaviors, this research provides actionable insights into the mechanisms of consumer-brand disengagement. The study underscores the role of NLP in advancing business intelligence and offers new perspectives on digital-era consumer behavior and its impact on brand reputation and strategy.

Evaluating large artery atherosclerosis is critical for predicting and preventing ischemic strokes. Ultrasonographic assessment of the carotid arteries is the preferred first-line examination due to its ease of use, noninvasive, and absence of radiation exposure. This study proposed an automated classification model for the common carotid artery (CCA), carotid bulb, internal carotid artery (ICA), and external carotid artery (ECA) to enhance the quantification of carotid artery examinations. A total of 2,943 B-mode ultrasound images (CCA: 1,563; bulb: 611; ICA: 476; ECA: 293) from 288 patients were collected. Three distinct sets of embedding features were extracted from artificial intelligence networks including pre-trained DenseNet201, vision Transformer (ViT), and echo contrastive language-image pre-training (EchoCLIP) models using deep learning architectures for pattern recognition. These features were then combined in a support vector machine (SVM) classifier to interpret the anatomical structures in B-mode images. After ten-fold cross-validation, the models achieved accuracies between 76.1%-81.8%, which was significantly better than using individual feature sets, with a p-value of <0.001. The proposed model could make carotid artery examinations more accurate and consistent with the achieved classification accuracy.

Our study investigates sustainability and environmental disclosure in Saudi Arabian accounting, focusing on stakeholder accountability. Employing content analysis and close reading, we examine disclosures in the annual reports of 18 companies from 2008 to 2018. Findings reveal limited disclosures, influenced by interpretations of Islam and government emphasis. Despite sector-specific attention to environmental issues, disclosures lack depth. The study offers insights into managerial motivations, highlighting divergent stakeholder priorities and moral obligations. Policy implications suggest mandatory regulations could enhance accountability and encourage comprehensive disclosures. The research underscores the significance of contextual understanding in shaping accounting practices, particularly in regions like Saudi Arabia. Regulatory changes could improve corporate reporting on environmental and sustainability matters, addressing societal needs more effectively.

This study investigates the predictability of high-frequency returns for 170 cryptocurrencies

using machine learning models. By applying linear, boosting, and neural network models,

we focus on one-hour and eight-hour intervals for cross-sectional and time-series predictions. Key features, including skewness, kurtosis, and CAPM metrics (alpha and beta), significantly influenced the results. Lasso and Elastic Net models performed best, explaining up to 2% of one-hour return variations, while the eight-hour predictions showed reduced accuracy, suggesting the market trends toward weak-form efficiency over longer periods. The study highlights the profitability of machine learning-based trading strategies, particularly in exploiting market inefficiencies at shorter time scales. This research underscores the potential of machine learning models to forecast short-term cryptocurrency returns effectively while illustrating their limitations in longer intervals as market efficiency increases.