Fecha de publicación: 30/01/2024 Fuente: Wipo "BigData" The invention provides a big data mining method and system based on digital services. Compared with a traditional preference viewpoint analysis technology, the preference viewpoint mining method and device have the advantage that the accuracy and timeliness of preference viewpoint mining can be improved. In detail, first session activity training text semantic moments of a plurality of business session activity training texts are fused; according to the embodiment of the invention, a derivative service preference demand vector between a service session activity training text without a service preference demand and a service session activity training text with a service preference demand or between the service session activity training text with the service preference demand and the service session activity training text with the service preference demand can be obtained; therefore, derivative expansion of the service preference demand vector can be realized, and diversity and comprehensiveness of debugging examples in a network debugging process are further guaranteed, so that the adaptability of the target preference viewpoint mining network to different digital services is improved.