传统的电影评价通常获取自专家评论或调查问卷, 但它们无法全面反映观众的真实意见。网络媒体和在线评论兴起使得网络文本成为一种可以帮助我们了解观众对电影的评价, 以评估电影质量的宝贵的资源。本研究以电影《热辣滚烫》为例, 收集了2024年2月10日至2024年6月10日期间大量网络评论数据, 基于ROST_CM6, 运用自然语言处理和文本挖掘技术对这些评论进行分析, 探究影响电影质量的因素。第一, 研究评价文本中的情感极性, 以确定观众对电影的整体情感倾向。第二, 分析评论中提及的电影特征, 如剧情、演员表现、视觉效果等, 并探究了这些特征与电影质量之间的关联。另外, 本研究借助议程设置理论的分析视角, 深入地探讨了媒体如何通过报道策略影响观众对这些话题的认知与评价标准, 以期为未来的电影创作与营销策略提供有益的参考。

Traditional film evaluations often stem from expert reviews or surveys, but they fail to comprehensively reflect the genuine opinions of audiences. The rise of online media and user comments has made web-based text a valuable resource that can assist us in understanding audience evaluations of films and assessing their quality. Taking the film “Spicy and Hot” as an example, this study collected a vast amount of online comment data from 2024. 2. 10 to 2024. 6. 10 and analyzed these comments using natural language processing and text mining techniques based on ROST_CM6 to explore the factors influencing film quality. Firstly, the study evaluated the sentiment polarity in the text to determine the overall emotional orientation of audiences towards the film. Secondly, it analyzed the film characteristics mentioned in the comments, such as plot, actor performance, visual effects, and explored the correlations between these characteristics and film quality. Additionally, this study employed the analytical perspective of the agenda-setting theory to deeply discuss how media influence audience perceptions and evaluation criteria of these topics through reporting strategies, aiming to provide beneficial insights for future film creation and marketing strategies.