Text recognition method and apparatus, computer-readable storage medium and electronic device

A non-transitory computer-readable storage medium, wherein the storage medium is stored with instructions that, when executed by a processor of an electronic device, enable the electronic device to perform a text recognition method,

• wherein the text recognition method comprises:
• determining, on the basis of a text to be recognized, a plurality of character strings corresponding to the text to be recognized, wherein adjacent character strings in the plurality of character strings partially overlap;
• performing word vector conversion on the plurality of character strings to obtain a plurality of word vectors, wherein the plurality of word vectors correspond to the plurality of character strings on a one-to-one basis;
• generating word vector recognition results respectively corresponding to the plurality of word vectors on the basis of the plurality of word vectors, wherein the word vector recognition result is an effect text or a non-effect text, and the effect text is a text representing technical effects in a patent text;
• determining a text recognition result of the text to be recognized on the basis of the word vector recognition results respectively corresponding to the plurality of word vectors, the determining improving the accuracy of text recognition;
• and
• classifying the patent text according to the text recognition result,
• wherein the generating word vector recognition results respectively corresponding to the plurality of word vectors on the basis of the plurality of word vectors comprises:
• determining a training text and a text recognition result corresponding to the training text, wherein the training text is a patent text, and comprises a first language training text and a second language training text, the first language training text comprises text contents written in the first language, the second language training text comprises text contents written in the second language, the text recognition result corresponding to the first language training text is a first text recognition result, and the text recognition result corresponding to the second language training text is a second text recognition result;
• on the basis of the training text and the text recognition result corresponding to the training text, determining a plurality of word vector samples corresponding to the training text and word vector recognition results respectively corresponding to the plurality of word vector samples;
• establishing an initial network model, and training the initial network model on the basis of the plurality of word vector samples and the word vector recognition results respectively corresponding to the plurality of word vector samples to generate an effect recognition model used for generating, on the basis of an input word vector, a word vector recognition result corresponding to the input word vector;
• and
• generating the word vector recognition results respectively corresponding to the plurality of word vectors on the basis of the plurality of word vectors using the effect recognition model, the generating the word vector recognition results using the effect recognition model improving an efficiency of text recognition, wherein the determining a training text and a text recognition result corresponding to the training text comprises:
• acquiring the first language training text and the second language training text;
• obtaining a first effect marking text corresponding to the first language training text on the basis of the first language training text and the first text recognition result;
• and
• determining the second text recognition result corresponding to the second language training text on the basis of the first effect marking text, wherein the acquiring the first language training text and the second language training text comprises:
• acquiring text data comprising patent family text data in a plurality of different languages;
• and
• screening the first language training text and the second language training text on the basis of the patent family text data, wherein the first language training text comprises effect identification paragraph information, and the second language training text comprises patent text to be marked with effect.

An electronic device, comprising:

• a processor;
• and
• a memory for storing computer-executable instructions;
• wherein the processor is configured to execute the computer-executable instructions to implement a text recognition method,
• wherein the text recognition method comprises:
• determining, on the basis of a text to be recognized, a plurality of character strings corresponding to the text to be recognized, wherein adjacent character strings in the plurality of character strings partially overlap;
• performing word vector conversion on the plurality of character strings to obtain a plurality of word vectors, wherein the plurality of word vectors correspond to the plurality of character strings on a one-to-one basis;
• generating word vector recognition results respectively corresponding to the plurality of word vectors on the basis of the plurality of word vectors, wherein the word vector recognition result is an effect text or a non-effect text, and the effect text is a text representing technical effects in a patent text;
• determining a text recognition result of the text to be recognized on the basis of the word vector recognition results respectively corresponding to the plurality of word vectors, the determining improving the accuracy of text recognition;
• and
• classifying the patent text according to the text recognition result,
• wherein the generating word vector recognition results respectively corresponding to the plurality of word vectors on the basis of the plurality of word vectors comprises:
• determining a training text and a text recognition result corresponding to the training text, wherein the training text is a patent text, and comprises a first language training text and a second language training text, the first language training text comprises text contents written in the first language, the second language training text comprises text contents written in the second language, the text recognition result corresponding to the first language training text is a first text recognition result, and the text recognition result corresponding to the second language training text is a second text recognition result;
• on the basis of the training text and the text recognition result corresponding to the training text, determining a plurality of word vector samples corresponding to the training text and word vector recognition results respectively corresponding to the plurality of word vector samples;
• establishing an initial network model, and training the initial network model on the basis of the plurality of word vector samples and the word vector recognition results respectively corresponding to the plurality of word vector samples to generate an effect recognition model used for generating, on the basis of an input word vector, a word vector recognition result corresponding to the input word vector;
• and
• generating the word vector recognition results respectively corresponding to the plurality of word vectors on the basis of the plurality of word vectors using the effect recognition model, the generating the word vector recognition results using the effect recognition model improving an efficiency of text recognition,
• wherein the determining a training text and a text recognition result corresponding to the training text comprises:
• acquiring the first language training text and the second language training text;
• obtaining a first effect marking text corresponding to the first language training text on the basis of the first language training text and the first text recognition result;
• and
• determining the second text recognition result corresponding to the second language training text on the basis of the first effect marking text,
• wherein the acquiring the first language training text and the second language training text comprises:
• acquiring text data comprising patent family text data in a plurality of different languages;
• and
• screening the first language training text and the second language training text on the basis of the patent family text data, wherein the first language training text comprises effect identification paragraph information, and the second language training text comprises patent text to be marked with effect.