132 results about "Sentence similarity" patented technology

The invention provides an interrogative sentence similarity measurement method based on a deep convolutional neural network. The method comprises the following steps that: S1: through a knowledge domain related page, generating a raw corpus, crawling Chinese characters which appear in raw language materials, and generating a word vector corresponding to each Chinese character; S2: replacing each Chinese character in an interrogative sentence by the corresponding word vector to obtain a word vector set corresponding to the interrogative sentence, wherein the word vector set carries out calculation through the convolutional neural network to obtain a corresponding sentence meaning vector; and S3: carrying out pairwise combination on the interrogative sentences, and calculating the cosine function absolute values of the sentence meaning vectors corresponding to two interrogative sentences to obtain a similarity between two interrogative sentences. By use of the method, an individual-character analysis method is adopted to avoid an influence on subsequent analysis by word segmentation errors, the convolutional neural network takes the whole interrogative sentence as a whole to extractwhole sentence characteristics, and a sentence meaning splitting problem brought by using word similarity matrixes is avoided.

The invention provides a method and a device for calculating sentence similarity and a method and a device for machine translation, wherein the method for calculating the sentence similarity comprises the following steps that a first sentence and a second sentence are compared, so as to determine different word pairs; different words are marked by utilizing the matching probability of the different words in the different word pairs with other words in the first sentence or the second sentence in which the different words are contained, wherein the matching probability of two words is obtained by inquiring a matching probability model, and the matching probability of the two words in the matching probability model is obtained by counting the co-occurrence frequency of the two words in a preset corpus; the marking results of the different words in the different word pairs are utilized to mark the different word pairs; and the marking results of the different word pairs are utilized to determine the similarity of the first sentence and the second sentence. According to the method and the device, the matching degree of the two sentences can be more accurately reflected, thereby increasing the application quality of the method and the device for the machine translation and the like.