User Needs Identification and Design Strategies for Children's Picture Books Using Big Data and Artificial Intelligence

Despite the continuous decline in the domestic birth rate, the children’s product market has been expanding. A wide range of picture books has been published for children’s aesthetic enlightenment and cognitive development; however, their educational impact remains limited due to inconsistent content quality, prevalent homogenization, and the absence of systematic evaluation frameworks. In response, this study establishes a big data and AI-driven framework encompassing user needs identification, function mapping, and content validation, and systematically analyzes the content structure, modes of expression, and interaction of children’s picture books. A total of 844,640 data entries were collected from 12 data sources, and semantic analysis combined with the TF-IDF method was employed to extract key user requirements and determine their priorities. These elements were then integrated with the reading process and user personas to identify critical problem points, which informed the construction of a functional tree for designing visual styles and plot interactions. Subsequently, prototype picture books were developed and tested in parent–child settings to evaluate readability, engagement, and educational effectiveness. Ultimately, this research proposes a data-driven design methodology that optimizes the content structure and representational strategies of children’s picture books, providing a scholarly foundation for the systematic development of early childhood educational materials, while also suggesting new directions for personalized recommendation and structured evaluation systems.