|
Comparative Analysis of Human and AI-Generated Stories Using Rhetorical Devices Cristian Cordos, Ştefan Trăuşan-Matu |
39-49 |
|
MetaGallery: Enhancing Virtual Art Experiences with Personalized AI-driven Algorithms and 3D/VR Interactions Andra Maria Ciobanu, Adrian Iftene |
50-64 |
1 University Politehnica of Bucharest
313 Splaiul Independentei, Bucharest, Romania
2 Institutul de Cercetări în Inteligenţa Artificială
Calea 13 Septembrie nr. 13, Bucureşti
3 Academy of Romanian Scientists
Splaiul Indpendentei 54, Bucharest, Romania
Abstract. This paper presents the results of a study evaluating the stylistic and aesthetic differences between texts authored by humans and those generated by artificial intelligence (AI), specifically ChatGPT 4.0. The research focuses on comparing rhetorical devices such as anaphora, polyptotons, epistrophe, alliteration, sibilance, consonance, and assonance. AI stories were created using prompts, some designed by the authors and others generated by ChatGPT itself. Rhetorical devices were analyzed using Python programs, with counts normalized by word count for fair comparison. Initial analysis showed AI-generated stories were poorer in rhetorical devices, but normalization revealed similar counts between human and AI texts. AI-generated stories demonstrated stylistic richness comparable to human-authored ones by replicating specific linguistic features of human writing.
Keywords: Artificial Intelligence; Natural Language Processing; chatbots; Large Language Models; Text analysis; Rhetorical devices
Cite this paper as:
Cordos, C., Trăuşan-Matu, S. Comparative Analysis of Human and AI-Generated Stories Using Rhetorical Devices.
International Journal of User-System Interaction 17(2),
39-49, 2024.
“Alexandru Ioan Cuza” University
Bulevardul Carol I, Nr.11, 700506, Iaşi, România
Abstract. MetaGallery introduces an innovative approach to virtual museum visits by integrating 3D and VR interactive environments and personalized experiences driven by artificial intelligence (AI). Unlike traditional virtual galleries, MetaGallery tailors each session uniquely based on user feedback and preferences, employing advanced recommendation algorithms and text generation methods to ensure dynamic content. Users can interactively explore realistic 3D-modeled art spaces through web interfaces or immersive virtual reality, making art accessible and engaging for diverse audiences. This paper details the implementation strategies, presenting significant contributions to Human-Computer Interaction by demonstrating enhanced user engagement and personalized cultural exploration through technology.
Keywords: Virtual Reality; Artificial Intelligence; Recommendation Systems; Machine Learning; Clustering; 3D Modeling; Embeddings; Language Models
Cite this paper as:
Ciobanu, A. M., Iftene, A. MetaGallery: Enhancing Virtual Art Experiences with Personalized AI-driven Algorithms and 3D/VR Interactions.
International Journal of User-System Interaction 17(2),
50-64, 2024.