文献类型： 会议论文
第一作者： Melody Yu
作者： Melody Yu 1 ; Yu Sun 2 ;
作者机构： 1.Sage Hill School, Newport Coast, USA
2.Sage Hill School, Newport Coast, USA|California State Polytechnic University, Pomona, Pomona, USA
关键词： TV;Correlation;Reviews;Density measurement;Network analyzers;Production;Games
会议名称： International Conference on Computer Communication and Artificial Intelligence
主办单位：
页码： 175-180
摘要： Viewer ratings are a key factor in shaping the success and continued viability of television shows. However, the factors contributing to high or low ratings are complex. This paper investigates whether character interactions in TV episodes, as captured through character network analysis, correlate with viewer ratings. Character networks are graphs created from the plot of a TV show that represent the interactions of characters in scenes. We constructed character networks from episode plots of three popular TV series and extracted graph metrics such as network density and centrality measures from these networks. We explored the hypothesis that interactions between the characters in a TV episode, as quantified by the metrics of the episode character networks, are related to the overall reception of the episode, as indicated by the TV episode reviews. We conducted regression analysis between character network metrices and the TV reviews. The findings revealed significant, albeit varied, correlations between certain network metrics and episode reviews across the different series. For instance, Game of Thrones episodes with fewer active characters received higher reviews, suggesting viewer preference for simpler character interactions. Similarly, Breaking Bad also gained popularity by centering on a smaller group of main characters. Our study contributes to understanding how character dynamics influence TV reviews, providing valuable insights for TV producers and writers in developing their shows. We have shown that network analysis has the capability to quantify character interactions. This could potentially allow producers to consider adjusting character dynamics in future episodes, which may enhance audience engagement.
分类号： tp18
- 相关文献
[1]Gene duplication, transfer, and evolution in the chloroplast genome. Xiong, Ai-Sheng,Peng, Ri-He,Zhuang, Jing,Gao, Feng,Zhu, Bo,Fu, Xiao-Yan,Xue, Yong,Jin, Xiao-Feng,Tian, Yong-Sheng,Zhao, Wei,Yao, Quan-Hong.
[2]A Review of Smart Traffic Operation System for Traffic Control Using Internet of effects & Reinforcement Learning. Bharat Pahadiya,Rekha Ranawat. 2023
[3]Approaches for Identifying Suicide Ideation in Social Media Texts: Comprehensive Review. Jayshri Suresh Sonawane,Dinesh Jain. 2024
[4]Revolutionizing Healthcare with Federated Learning: A Comprehensive Review. Snehlata Mishra,Ritu Tondon,Narendra Pal Singh Rathore. 2024
[5]Spam Free Social Media User Feed Generation Using ML Technique. Anubha Sharma,Manoj Ramaiya. 2023
[6]A Review of Protein Sequences of COVID-19 Using Machine Learning and Deep Learning Approaches. Anurag Golwelkar,Abhay Kothari. 2023
[7]Estimation of grassland productivity by using climate parameters in the French Piedmonts Pyrenean Mountains. Protin P.V.,Pommelet A.,Caperaa P.. 2011
[8]Evaluation of Solar Energy Generation and Radiation Prediction Using Machine Learning. Santosh S Raghuwanshi,Jagrati Gupta,Gourav Shrivastava. 2023
[9]Automated Tomato Inspection and Harvesting System Using Robotic Arm and Computer Vision in Greenhouses. Gaby Abou Haidar,Maher Alaa Deen,Roger Achkar,Michel Owayjan,Roy Abi Zeid Daou. 2023
