This chapter investigates the impact of data visualization on enhancing statistical thinking in higher education, emphasizing its significance in developing students' abilities to analyze and interpret data effectively. By utilizing visual tools such as graphs, charts, and interactive simulations, students can transform abstract statistical concepts into concrete understandings, facilitating deeper engagement and comprehension. The study explores various technologies, including R, Python, and Tableau, that support dynamic data manipulation and real-time feedback, promoting active learning experiences. While the benefits of visualization are substantial, challenges such as varying levels of technological proficiency among students and the potential for superficial understanding are also addressed. Ultimately, this research underscores the importance of integrating effective visualization techniques with pedagogical strategies to cultivate robust statistical literacy and critical thinking skills, preparing students for data-driven decision-making in their respective fields.
The rapid growth of voice-driven online shopping, powered by smart assistants such as Alexa, Google Assistant, and Siri, has transformed digital commerce by enabling personalized and hands-free shopping experiences. However, challenges related to trust, usability, and technological adaptability remain underexplored. This study aims to assess the effectiveness of smart assistance in enhancing consumer satisfaction, adoption, and engagement in voice-driven e-commerce. Methodologically, the study employs a predictive analysis framework integrating machine learning algorithms (Random Forest, SVM, Logistic Regression, among others) and multiple regression analysis to examine consumer behavior across diverse demographic and technological contexts. Exploratory Factor Analysis (EFA) and Principal Component Analysis (PCA) are also applied to identify key latent factors. Findings reveal that trust, accuracy of voice recognition, and seamless personalization strongly influence adoption, offering critical insights for optimizing smart assistant-enabled shopping systems.
In this work, we investigate an anisotropic Bianchi type–VII cosmological model in the framework of Lyra geometry filled with perfect fluid matter and Tsallis–Barrow holographic dark energy. The modified Einstein field equations are derived, and exact solutions are obtained by assuming a power-law average scale factor for a decelerating universe. Expressions for various cosmological parameters such as the Hubble parameter, expansion scalar, shear scalar, matter density, dark energy density, and density parameters are derived and analysed. The behaviour of these parameters indicates that the universe is expanding continuously, with the expansion rate decreasing with cosmic time. The anisotropy parameter decreases gradually, indicating that the universe evolves towards isotropy at late times. Energy conditions, stability analysis, and cosmological diagnostics, including the statefinder and Om parameters, are also examined to evaluate the model's physical viability. The results suggest that Tsallis–Barrow entropy corrections in Lyra geometry provide a consistent framework for studying anisotropic cosmological evolution and dark energy dynamics.
This study develops and empirically validates a structural model examining the determinants of alcohol misuse among adults aged 60 and above through a three-stage analytical framework of Confirmatory Factor Analysis, Structural Equation Modeling, and XGBoost machine learning. A survey of N = 612 elderly respondents spanning five living arrangement categories was conducted. Exploratory data analysis reveals that 72.1 percent of elderly alcohol misusers are never screened by their general practitioner, 67.4 percent are undetected by their families, and 81.3 percent receive no specialist referral. CFA confirms a six-factor measurement model with excellent fit (CFI = .963, RMSEA = .041). SEM establishes social isolation (beta = .46, p < .001) and mental health comorbidity as the co-dominant misuse drivers, with mental health comorbidity significantly moderating all six pathways. Findings advance theories and models in elderly substance misuse contexts and generate urgent policy implications for family education, primary care screening protocols, and geriatric mental health integration
Human capital agility, defined as the dynamic capacity to reconfigure workforce competencies amid technological disruption, has emerged as a critical competitive imperative, yet empirical literature linking specific technological enablers to agility outcomes remains limited. This study examined N = 512 knowledge workers across five industries and four countries using multiple regression and machine learning algorithms including Random Forest, Gradient Boosting, SVR, and XGBoost. Seven technological enablers significantly predicted human capital agility, with AI and automation adoption (𝛽= 0.44), cloud-based learning systems (𝛽 = 0.37), and digital collaboration platforms (𝛽= 0.33) as strongest predictors. The MLR model explained 65.5% of variance, F(14, 497) = 67.42, p < .001, while XGBoost achieved superior accuracy (R2= .791). Organizational learning culture significantly moderated all technology-agility relationships (𝛽= 0.19, p < .001), confirming that technological investment yields maximum returns only within supportive cultural infrastructures.