Data classification model updating and application method and device, storage medium and product

Abstract

The embodiment of the invention provides a data classification model updating and application method and device, a storage medium and a product, and the method comprises the steps: obtaining a first labeled sample set and a first unlabeled sample set from a municipal governance event data pool, and enhancing preset data; obtaining a corresponding second labeled sample set and a second unlabeled sample set; obtaining a label prediction result corresponding to each unlabeled sample in the second unlabeled sample set according to the initial data classification model; performing mixed sample data enhancement on the second labeled sample set and the second unlabeled sample set after label prediction; and performing parameter updating on the initial data classification model according to the second labeled sample set and the second unlabeled sample set after mixed sample data enhancement processing to obtain a target data classification model. A semi-supervised learning method is adopted, the sample diversity is increased through data enhancement and mixing, the robustness and stability of the model are improved, the model prediction effect is improved, and the cost is saved.

Description

technical field [0001] The embodiments of the present invention relate to the field of artificial intelligence, and in particular to an update and application method, device, storage medium and product of a data classification model. Background technique [0002] At present, the classification of city events can be achieved by using the data classification model, so that the events can be dispatched to the corresponding departments for processing. As time goes by, the distribution of event data in the city area may change, such as changes in the environment over time, such as sudden epidemics, natural disasters, etc. In order to ensure the prediction effect of the sub-classification model, it is necessary to continuously update the data classification model at regular intervals. renew. [0003] In the existing technology, in order to ensure the predictive ability of the data classification model, it is usually necessary to update (fine-tune) the model with the latest collec...

AI Technical Summary

Problems solved by technology

[0004] In practical applications, it is often very difficult to obtain labeled data, or requires an online system to provide a timely feedback mechanism, or requires manual data labeling, resulting in a large amount of additional economic and time costs

If the latest labeled data cannot be obtained in time, the data classification model may face a decline in predictive ability; and even if the labeled data is obtained, if the amount of data is insufficient, the existing model update method will have limited improvement in the prediction ability of the data classification model. There is also a risk that the effect of the model will decline

Images