UPDATE TRAINING METHOD, APPARATUS AND DEVICE FOR TEXT CLASSIFICATION MODEL

摘要

An update training method, apparatus and device for a text classification model, the text classification model comprising a semantic extraction layer and a classification layer. Before a knowledge point is added, the training of the semantic extraction layer and the classification layer is completed according to sample data of the original knowledge point. The method comprises: acquiring a sample text corresponding to a newly added knowledge point and a labeling label for labeling the sample text (S110); constructing a feature vector of the sample text by means of the semantic extraction layer which completes the training according to the sample data (S130); and carrying out update training for the classification layer according to the feature vector of the sample text and the labeling label corresponding to the sample text, so as to implement update training for the text classification model (S150). When the text classification model needs to undergo update training, only the classification layer undergoes update training, which may greatly shorten the update training time of the text classification model and achieve timely updating of the text classification model.