TEXT SIMILARITY ACQUISITION METHOD, APPARATUS, DEVICE, AND READABLE STORAGE MEDIUM

摘要

Disclosed are a text similarity acquisition method, an apparatus, a device, and a readable storage medium. Said method comprises: obtaining, according to a data set of a text pair, numerical features of the text pair; constructing, by means of the numerical features of the text pair, a sample feature matrix; performing model training according to the sample feature matrix and a prediction vector, to obtain a prediction model; and acquiring a target text pair, and acquiring, according to the sample feature matrix and the prediction model, a similarity score of the target text pair. Figure 1