Kenyon College
Kenyon College: Digital Kenyon - Research, Scholarship, and Creative ExchangeNot a member yet
135065 research outputs found
Sort by
Human-Chatbot Interaction Patterns: A Topic Modeling Analysis of 3,275 Conversations with ChatGPT
Get PDFThis research presents a topic modeling analysis of the WILDCHAT-FULL dataset, examining user interactions with ChatGPT across over one million conversations. The study focused specifically on extended conversations (five or more exchanges) between U.S.-based users and ChatGPT in English. Using Latent Dirichlet Allocation (LDA), I identified 50 distinct conversational topics with coherence scores ranging from -1.5 to -14.92 (mean: -4.47). The analysis revealed diverse interaction patterns spanning creative writing, jailbreaking (attempting to get ChatGPT to do something against its guidelines), technical discussions, business applications, and educational queries. A particularly striking finding was that creative writing and role-play scenarios dominated the interactions, comprising over 25% of the identified topics
Unraveling the March Sisters: A Digital Network Analysis of Little Women\u27s Evolving Bonds
Get PDFLittle Women is one of the most female-focused novels of the 19th century. The book, being so popular, inspired numerous film and television adaptations. The three most recognized film adaptations are: 1933 directed by George Cukor, 1994 directed by Gillian Armstrong, and 2019 directed by Greta Gerwig. This project aims to analyze and compare the social networks in these three adaptations to assess how genuinely female-centric the films are. I am curious to see if there have been any changes in the social networks and if they have evolved in response to cultural context. Using AI for coding assistance and Google CoLab, I mapped social dynamics based on dialogue and character interactions. I hope to uncover trends in female relationships in each adaptation and explore where shifts in the focus of relationships reflect broader societal changes. This project combines literary and film analysis with digital humanities by using a new method of social network visualization for character analysis
Instagram\u27s 2025 Terms of Service: The Evolution of Surveillance Capitalism and AI Training Data
Get PDFThis research examines how social media platforms like Instagram’s 2025 Terms of Services (TOS) updates reflect the increasing demand for AI training data, illustrating the critical concerns about user privacy and data rights. Technology companies like Meta progressively require vast amounts of user-generated content to develop competitive AI systems to generate financial gain. Forming their data collection practices has led to their updated terms and services. In particular, their expanded AI training rights and modified dispute resolution procedures reveal the complex tension between technological advancement and user privacy. AI training data raises urgent questions about consent, ownership, and the commodification of personal experience for these algorithms. Shoshana Zuboff refers to this issue as surveillance capitalism
注满欲望和能量的意象(Imagery Full of Desire and Energy)
Get PDFThis article, written by female critic Liao Wen, was published in China Culture Newspaper and focuses on Yang Ke-Qing\u27s stylistic approach to imagery in her oil paintings. For more information on the featured artwork: https://digital.kenyon.edu/zhou/450/ (Nicole Wang \u2726).https://digital.kenyon.edu/zhoudocs/1406/thumbnail.jp
The Effect of NaCl Stress on Root Growth, Stomatal Density, and Photosynthetic Activity of Phaseolus vulgaris
No full text
