Abstract: A biographical visualization is a graphical description of a person’s life to identify key information and gain insights from massive biographical data through visualization and visual analytics. A narrative visualization generation system called BioVistory based on biographical text data is developed to addressing challenges in automatically generating narrative visualizations from biographical text data, which includes distilling concise design patterns, designing an automated workflow for biographical visualization generation, and establishing universally applicable visualization design solutions. This system takes biographical text data and other auxiliary information as input and applies natural language processing techniques for preprocessing. It extracts events from the text, categorizing them into four different perspectives: professional career, personal activities, interpersonal relationships, and chronological information, dividing them into various life stages. The extracted events, perspectives, and stages are then automatically transformed into biographical visualizations using visual metaphors. The system offers users three levels of semantic interaction exploration: event interaction, perspective interaction, and stage interaction. Case studies and user experiments show that BioVistory can automatically generate high-quality biographical visualizations, effectively supporting data-driven narrative visualization and enhancing the expressiveness and impact of biographical texts. It supports reader-driven reading experiences, improves learning outcomes, increases user engagement, and reduces cognitive load.