TRAINING METHOD AND APPARATUS FOR TEXT SIMILARITY RECOGNITION MODEL, AND RELATED DEVICE

摘要

The present application relates to the technical field of text recognition in artificial intelligence. Provided are a training method and apparatus for a text similarity recognition model, and a related device. The method comprises: obtaining a plurality of first sample groups comprising first text samples and second text samples; using as third text samples elements having a literal similarity to the first text samples that reaches a preset threshold; labelling the third text samples to obtain negative text samples, and forming a plurality of second sample groups; representing each sample in each second sample group with a representation vector; calculating a first similarity and a second similarity; and adjusting parameters according to the first similarity and the second similarity, and repeatedly obtaining representation vectors up to the present step so as to obtain a trained text similarity recognition model. By implementing the present application, the problem in the prior art that text similarity recognition methods are low in recognition accuracy can be solved; moreover, the present application further relates to blockchain technology, and the first sample groups and the second sample groups can be stored in a blockchain node.