The invention discloses an adaptive field viewpoint data classification method, and belongs to the technical field of internets. The classification method includes the steps: determining a source field search word matrix and a target field search word matrix; determining a source field objective function and a target field objective function; determining a total objective function according to the source field objective function and the target field objective function; respectively determining target values of various parameters in the total objective function; training designated classification models according to the target values of various parameters and marked viewpoint data in a source field, and performing classification for viewpoint data in a target field by the designated classification models obtained in a training manner. The total objective function is connected with a special source field topic matrix, a special target field topic matrix and a hub matrix expressing sharing topics between the source field and the target field, so that the adaptive field viewpoint data classification method is achieved by the sharing topics, and differences between the source field and the target field can be decreased by the sharing topics, so that accuracy of classification results can be ensured.