Emerging media technologies, epitomized by short video intelligent recommendation algorithms, are catalyzing substantial transformations within the realm of film and television culture. This study delves into the secondary creation of film and television works facilitated by algorithmic recommendation technologies. Specifically, it introduces a novel short video intelligent recommendation algorithm founded on deep learning principles. This algorithm harnesses joint features derived from the scene and behavioral attributes of target short videos, employing a graph convolutional neural network to model the long-term and short-term preferences, thereby enabling intelligent recommendations for the secondary creation of film and television works. Further, this research designs and implements questionnaires to formulate research hypotheses and conducts surveys targeting both general film and television audiences and industry professionals. The objective is to scrutinize the profound alterations instigated by emerging algorithmic recommendation technologies in film and television culture and to assess the adaptability and acceptance of these changes among media practitioners. The findings reveal that both explicit personalization (t=9.26, P<0.001) and implicit personalization (t=8.107, P<0.001) significantly enhance the immersive experience of viewers. The application of such technologies has rendered film and television audiences more discerning in their evaluation of various aspects of media productions. Notably, film and television media practitioners aged 30 years or older exhibit a more favorable disposition towards algorithmic recommendation technologies. This investigation not only underscores the significant impact of algorithmic recommendation technologies on film and television culture but also proposes new avenues for the seamless integration and innovative development of film and television media with information technology.
