17504 research outputs found
Sort by
Build, Bust and Rebuild: Micropolitan Transportation in Wooster, Ohio
No full textThe preceding research pertains to transportation as experienced in micropolitan areas. Micropolitan areas are a designation that is under researched. This is done by way of triangulation of qualitative data surrounding the provision of transportation in the city of Wooster, Ohio. Special attention is given to the role of nonprofits, with Wooster’s case including the nonprofit Community Action of Wayne/Medina. Data is collected via online archives, current webpages, and data provided in the Wooster public library. It was found that the city of Wooster experiences a “build, bust, and rebuild” style of transportation provision, swinging from private contracting to public acquisition. This characterizes micropolitan areas as lacking a formal agency to aid in coordinating service
How Controller Designs Affect Video Game Accessibility and User Experience
No full textThe video game industry has vastly expanded in recent years, especially in entertainment. With the popularity of video games, various issues have also followed, particularly, accessibility in the design of these games. Since video games joined mainstream entertainment, garnering a huge amount of success the formulas for creating video games became somewhat uniform in style, specifically with controller design. Due to this uniformity, this study explores different controller types (i.e., keyboard and mouse, Xbox, voice control, and head tracking) to analyze to what extent controller type contributes to the user experience of video game play and its accessibility. This is examined by creating a simple game that consists of 4 scenes each with a slightly different task. Those tasks help users test the functionality, usability, and user experience each controller provides. The experiment had a total of 23 participants who completed a short demographics section, expressed how they viewed functionality, usability, user experience, and accessibility, and then tested each controller while filling out corresponding sections of a survey. The survey was then analyzed to see if game controllers affect user experience and accessibility, how user experience, accessibility, usability, and functionality are connected, and how familiarity plays a role in the users’ opinions of both user experience and accessibilit
How the Roman World was Restored; an Analysis of the Last Decades of the Crisis of the Third Century
No full textThis Independent Study examines the military recovery and political stabilization of the Roman empire from 260 to 283 AD. My project seeks to answer the following questions: How did Roman fortunes rebound so quickly during this turbulent period and what was the broader historical significance of the era, which I term the Late Crisis. To research this subject, I carefully read several ancient sources, most importantly the Historia Augusta and Zosimus’ New History, and dozens of modern publications on third-century Roman history and many related topics. The most important work for this period is David Potter’s The Roman Empire at Bay, AD 180-395. Potter contended that the third and fourth centuries can broadly be understood as a period of decline caused by poor leadership and a weakening military. I conclude that within this period, the Late Crisis represented a temporary reversal of the trend. While their economic policies were not always successful, I argue that Late Crisis emperors restored Roman dominance over neighboring territories because of an effective military doctrine pioneered by the emperor Gallienus in the 260’s and mastered in the 270’s by subsequent emperors Aurelian and Probus
Time Allocation and Inequality: A Study of Shifting Social Leisure Time Disparities
No full textThis study utilized the American Time Use Survey (ATUS) data from 2003 to 2023 to document trends in social leisure time allocation. In the two decades, the average weekly leisure time of U.S citizens increased by approximately 2 hours and 30 minutes. Previous studies have shown significant disparities in social leisure time based on factors such as gender, marital status, employment status, education level, and parental status. Comparing 2003 to 2023, these differences have largely persisted, with no significant changes observed in gender, marital status, employment status, or education level. However, a notable increase in the gap between parents and non-parents in terms of leisure time has been observed, with parents experiencing a substantial reduction in social leisure time. Regression models indicate that age and parental status have the strongest explanatory power in determining social leisure time. Time series predictions suggest that social leisure time will continue to increase in the future
Audio Signal Analysis for Songs’ Mood Recognition by Clustering and Neural Network
No full textIn the evolving landscape of digital music consumption, accurate mood classification plays a crucial role in enhancing user experience on streaming platforms. While services like Spotify and SoundCloud have introduced mood-based features, inconsistencies in labeling often result in disjointed listening experiences. This project proposes a music mood recognition algorithm based on audio processing analysis and machine learning, particularly focusing on clustering techniques and neural networks.
This study utilizes a dataset of 389 songs across eight genres from the Free Music Archive (FMA), with music features such as tempo, key, scale (surface-level), and audio deep learning features such as spectral centroid and Mel-spectrograms, which are extracted through extensive preprocessing, including denoising, trimming, normalization, and Short-Time Fourier Transform (STFT). Both audio features and deep spectral representations are analyzed together and individually to find the best models.
Two machine learning approaches are employed: unsupervised clustering (K-Means and Fuzzy C-Means) to group songs into mood categories without labels, and a supervised neural network trained with sparse categorical cross-entropy and optimized via Stochastic Gradient Descent. Principal Component Analysis (PCA) is used to reduce feature dimensionality. Mood categories are defined by energy and emotion levels (Energetic/Relaxing × Happy/Sad), forming four mood groups.
The results show K-Means provides clearer, interpretable clusters, while the neural network—though flexible—exhibits overfitting, with a test accuracy of 43.59%. Fuzzy clustering reveals the overlapping emotional content in music more effectively than hard clustering. ii Limitations include dataset size, subjectivity in mood labeling, and lack of lyrical or contextual data.
This project contributes a potential framework for mood-based recommendation, offering feasible applications in playlist generation and music discovery for users and artists alike. Future work may explore deep learning architectures, context-aware personalization, and integration of lyrics for a more comprehensive emotional modeling of music
Investigating the Effect of Shade on Crop Yield in Agrivoltaics
No full textAgrivoltaics, the integration of solar energy generation with agricultural production, offers a sustainable solution to the growing competition for land between food production and renewable energy. Solar panels installed a couple of meters above the ground allow space for crops to grow beneath them, providing shade that can affect crop yield. The extent and nature of this effect vary by crop type and environmental conditions, and understanding these dynamics is crucial to optimizing agrivoltaic systems. This study investigates the impact of shade on crop yield across different crop types and climatic zones, focusing on how varying levels of reduction in solar radiation (RSR) influence agricultural productivity in agrivoltaic systems. The research employed a multiple linear regression model to analyze data from a meta-analysis of crop yield responses to shading, exploring interactions between RSR, crop type, and plant hardiness zones. The findings reveal that yield responses to shade are significantly different across crop types. Additionally, while climatic factors such as plant hardiness zones were considered, they did not show clear trends in yield response, suggesting that other environmental variables may play a more significant role in explaining yield variability. This analysis classified crops into three categories based on their shade sensitivity: berries, fruits, and fruity vegetables are categorized as shade benefiting, C3 cereals, forages, leafy vegetables are shade tolerant, and maize and grain legumes are shade susceptible. These results highlight the importance of crop-specific considerations in agrivoltaic design to maximize agricultural productivity while supporting renewable energy generation
Transparency in Action: Evaluating the Influence of Taxpayer Receipt Initiatives on Public Trust in Government
No full textThis paper examines whether providing citizens with tax receipts—documents detailing how their taxes are spent—can improve trust in government. Drawing on public choice theory, rational ignorance, and principal-agent models, the study explores how information influences public perception. A difference-in-differences analysis across five countries that introduced tax receipt initiatives reveals that trust in government actually declined following their implementation. This decline intensified over time, suggesting that transparency may expose gaps between government actions and citizen expectations. While tax receipts may increase awareness, they do not inherently foster trust. The findings highlight that transparency tools like tax receipts must be paired with substantive reforms to build accountability and public confidence. This study contributes to ongoing discussions on transparency by emphasizing the context-dependent nature of information-based governance strategies
