CN114328815A

CN114328815A - Text mapping model processing method and device, computer equipment and storage medium

Info

Publication number: CN114328815A
Application number: CN202111376101.5A
Authority: CN
Inventors: 周辉阳; 闫昭
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2021-11-19
Filing date: 2021-11-19
Publication date: 2022-04-12

Abstract

The embodiment of the application discloses a processing method and device of a text mapping model, computer equipment and a storage medium, and belongs to the technical field of computers. The method comprises the following steps: acquiring sample text information and first label information, and mapping the sample text information based on a text mapping model to obtain predicted text information; and determining second label information based on a first similarity between the predicted text information and the sample text information and a second similarity between the first label information and the sample text information, and training the text mapping model based on the second label information, the predicted text information and the first similarity corresponding to the predicted text information. In the embodiment of the application, the text information most similar to the sample text information is used as the second label information, and the text mapping model is trained based on the predicted text information, the second label text information and the first similarity between the predicted text information and the sample text information, so that the mapping effect of the text mapping model is improved.

Description

Text mapping model processing method and device, computer equipment and storage medium

Technical Field

The embodiment of the application relates to the technical field of computers, in particular to a processing method and device of a text mapping model, computer equipment and a storage medium.

Background

With the development of computer technology, artificial intelligence models are widely applied in a variety of scenes. In a natural language processing scenario, text information is usually mapped based on a text mapping model to obtain text information similar to the text information, so as to implement expansion of the text information, but the mapping effect of the current text mapping model is still poor.

Disclosure of Invention

The embodiment of the application provides a processing method and device of a text mapping model, a computer device and a storage medium, which can improve the mapping effect of the text mapping model. The technical scheme is as follows:

in one aspect, a method for processing a text mapping model is provided, where the method includes:

acquiring sample text information and first label information, wherein the first label information is the text information of which the similarity with the sample text information is not less than a similarity threshold value;

mapping the sample text information based on the text mapping model to obtain predicted text information;

determining second label information based on a first similarity between the predicted text information and the sample text information and a second similarity between the first label information and the sample text information, wherein the second label information is text information with larger similarity between the predicted text information and the sample text information;

training the text mapping model based on the second label information, the predicted text information and the first similarity corresponding to the predicted text information, wherein the text mapping model is used for mapping similar text information of any text information.

In another aspect, an apparatus for processing a text mapping model is provided, the apparatus including:

the acquisition module is used for acquiring sample text information and first label information, wherein the first label information is the text information of which the similarity with the sample text information is not less than a similarity threshold value;

the mapping module is used for mapping the sample text information based on the text mapping model to obtain predicted text information;

a determining module, configured to determine second label information based on a first similarity between the predicted text information and the sample text information and a second similarity between the first label information and the sample text information, where the second label information is text information with a larger similarity between the predicted text information and the sample text information;

and the training module is used for training the text mapping model based on the second label information, the predicted text information and the first similarity corresponding to the predicted text information, wherein the text mapping model is used for mapping similar text information of any text information.

In one possible implementation, the apparatus further includes:

the obtaining module is further configured to obtain a third similarity and a fourth similarity between the predicted text information and the sample text information, where the third similarity indicates a difference between words included in the predicted text information and the sample text information, and the fourth similarity indicates a semantic similarity between the predicted text information and the sample text information;

and the fusion module is used for performing weighted fusion on the third similarity and the fourth similarity to obtain the first similarity.

In another possible implementation manner, the obtaining module is configured to divide the predicted text information based on at least one character number to obtain at least one first term set, where terms belonging to the same first term set have the same number of characters; based on the number of at least one character, dividing the sample text information respectively to obtain at least one second word set, wherein the words belonging to the same second word set contain the same number of characters; determining a first number and a second number, the first number indicating a sum of a number of different words in the first set of words and a second set of words for each of the character numbers, the second number indicating a total number of words in at least one of the first set of words and at least one of the second set of words; determining a ratio of the first number to the second number as the third degree of similarity between the predicted text information and the sample text information.

In another possible implementation manner, the obtaining module is configured to perform semantic extraction on the predicted text information and the sample text information respectively to obtain a first semantic feature of the predicted text information and a second semantic feature of the sample text information; determining a similarity between the first semantic feature and the second semantic feature as the fourth similarity.

In another possible implementation manner, the obtaining module is configured to splice the predicted text information and the sample text information to obtain spliced text information; semantic extraction is carried out on the spliced text information to obtain a third semantic feature corresponding to the spliced text information; classifying the third semantic features to obtain a classification result; determining the classification result as the fourth similarity.

In another possible implementation manner, the training module is configured to obtain a first loss value corresponding to the predicted text information based on the second label information and the predicted text information; and training the text mapping model based on the first similarity and the first loss value corresponding to the predicted text information.

In another possible implementation manner, the training module is configured to obtain a weight parameter corresponding to each piece of predicted text information based on a difference between a target similarity and the first similarity corresponding to each piece of predicted text information; based on the weight parameter corresponding to each predicted text message, carrying out weighted average on the first loss values corresponding to the plurality of predicted text messages to obtain a second loss value; determining an average value of the first loss values corresponding to the plurality of predicted text information as a third loss value; training the text mapping model based on the second loss value and the third loss value.

In another possible implementation manner, the obtaining module includes:

the device comprises an acquisition unit, a processing unit and a display unit, wherein the acquisition unit is used for acquiring a first text information set, and the first text information set comprises a plurality of first text information and label information corresponding to each first text information;

a determining unit, configured to determine a similarity between each of the first text information and corresponding tag information;

and the screening unit is used for screening the sample text information and the first label information with the similarity greater than the similarity threshold value from the first text information set based on the similarity corresponding to each piece of the first text information.

In another possible implementation manner, the screening unit is configured to screen, based on a similarity corresponding to each first text message, at least one second text message of which the similarity is greater than the similarity threshold from the first text message set, and configure a second text message set with the screened second text message and corresponding tag information; and acquiring the sample text information and the first label information from the second text information set, wherein the sample text information is any one of the second text information.

In another possible implementation manner, the obtaining module is further configured to train the text mapping model based on the first text information in the first text information set and the corresponding label information.

In another possible implementation manner, the determining module is configured to determine, in response to that a largest first similarity among the plurality of first similarities is greater than the second similarity, predicted text information corresponding to the largest first similarity as the second label information, where the plurality of first similarities are similarities between the plurality of predicted text information and the sample text information; or, in response to that none of the plurality of first similarities is greater than the second similarity, determining the first tag information as the second tag information.

In another possible implementation manner, the mapping module is further configured to map the target text information based on the text mapping model to obtain similar text information of the target text information.

In another aspect, a computer device is provided, which includes a processor and a memory, wherein at least one computer program is stored in the memory, and the at least one computer program is loaded and executed by the processor to implement the operations performed by the processing method of the text mapping model according to the above aspect.

In another aspect, a computer-readable storage medium is provided, in which at least one computer program is stored, the at least one computer program being loaded and executed by a processor to implement the operations performed by the processing method of the text mapping model according to the above aspect.

In a further aspect, a computer program product is provided, which comprises a computer program, which when executed by a processor, performs the operations performed by the processing method of the text mapping model according to the above aspect.

The beneficial effects brought by the technical scheme provided by the embodiment of the application at least comprise:

according to the method, the device, the computer equipment and the storage medium provided by the embodiment of the application, the text information which is most similar to the sample text information in the predicted text information and the first label information is taken as the second label information according to the similarity between the predicted text information and the sample text information respectively, the difference between the predicted text information and the second label text information is considered, the first similarity between the predicted text information and the sample text information is also considered, and the text mapping model is trained based on the second label information, the predicted text information and the first similarity corresponding to the predicted text information, so that the subsequent text mapping model based on the trained text mapping model can map out the similar text information of any text information, and the mapping effect of the text mapping model is improved.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a schematic diagram of an implementation environment provided by an embodiment of the present application;

FIG. 2 is a flowchart of a processing method of a text mapping model according to an embodiment of the present disclosure;

FIG. 3 is a flowchart of another processing method for a text mapping model according to an embodiment of the present disclosure;

FIG. 4 is a schematic structural diagram of a text mapping model provided in an embodiment of the present application;

FIG. 5 is a flowchart of a processing method of another text mapping model provided in an embodiment of the present application;

FIG. 6 is a diagram illustrating a training text mapping model according to an embodiment of the present disclosure;

FIG. 7 is a schematic diagram of a data comparison provided by an embodiment of the present application;

FIG. 8 is a schematic diagram of an editing interface of a knowledge base of questions and answers provided by an embodiment of the present application;

FIG. 9 is a schematic structural diagram of a processing apparatus of a text mapping model according to an embodiment of the present application;

FIG. 10 is a schematic structural diagram of a processing apparatus of another text mapping model provided in an embodiment of the present application;

fig. 11 is a schematic structural diagram of a terminal according to an embodiment of the present application;

fig. 12 is a schematic structural diagram of a server according to an embodiment of the present application.

Detailed Description

To make the objects, technical solutions and advantages of the embodiments of the present application more clear, the embodiments of the present application will be further described in detail with reference to the accompanying drawings.

The terms "first," "second," "third," "fourth," and the like as used herein may be used herein to describe various concepts, but these concepts are not limited by these terms unless otherwise specified. These terms are only used to distinguish one concept from another. For example, the first similarity can be referred to as a second similarity, and similarly, the second similarity can be referred to as the first similarity, without departing from the scope of the present application.

As used herein, the terms "at least one," "a plurality," "each," and "any," at least one of which includes one, two, or more than two, and a plurality of which includes two or more than two, each of which refers to each of the corresponding plurality, and any of which refers to any of the plurality. For example, the predicted text information includes 3 predicted text information, each of the 3 predicted text information is referred to, and any one of the 3 predicted text information is referred to as any one of the 3 predicted text information, which can be the first predicted text information, the second predicted text information, or the third predicted text information.

Artificial Intelligence (AI) is a theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human Intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technique of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a manner similar to human intelligence. Artificial intelligence is the research of the design principle and the realization method of various intelligent machines, so that the machines have the functions of perception, reasoning and decision making.

The artificial intelligence technology is a comprehensive subject and relates to the field of extensive technology, namely the technology of a hardware level and the technology of a software level. The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.

Natural Language Processing (NLP) is an important direction in the fields of computer science and artificial intelligence. It studies various theories and methods that enable efficient communication between humans and computers using natural language. Natural language processing is a science integrating linguistics, computer science and mathematics. Therefore, the research in this field will involve natural language, i.e. the language that people use everyday, so it is closely related to the research of linguistics. Natural language processing techniques typically include text processing, semantic understanding, machine translation, robotic question and answer, knowledge mapping, and the like.

Machine Learning (ML) is a multi-domain cross discipline, and relates to a plurality of disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and the like. The special research on how a computer simulates or realizes the learning behavior of human beings so as to acquire new knowledge or skills and reorganize the existing knowledge structure to continuously improve the performance of the computer. Machine learning is the core of artificial intelligence, is the fundamental approach for computers to have intelligence, and is applied to all fields of artificial intelligence. Machine learning and deep learning generally include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and teaching learning.

According to the scheme provided by the embodiment of the application, the text mapping model can be trained based on the machine learning technology of artificial intelligence, similar text information of any text information can be mapped by using the trained text mapping model, and therefore the processing method of the text mapping model is achieved.

The processing method of the text mapping model provided by the embodiment of the application is executed by computer equipment. Optionally, the computer device is a terminal or a server. Optionally, the server is an independent physical server, or a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a web service, cloud communication, a middleware service, a domain name service, a security service, a CDN (Content Delivery Network), a big data and artificial intelligence platform, and the like. Optionally, the terminal is a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, a smart voice interaction device, a smart home appliance, a vehicle-mounted terminal, and the like, but is not limited thereto.

In some embodiments, the computer program according to the embodiments of the present application may be deployed to be executed on one computer device or on multiple computer devices located at one site, or may be executed on multiple computer devices distributed at multiple sites and interconnected by a communication network, and the multiple computer devices distributed at the multiple sites and interconnected by the communication network can form a block chain system.

In some embodiments, the computer device is provided as a server. Fig. 1 is a schematic diagram of an implementation environment provided by an embodiment of the present application. Referring to fig. 1, the implementation environment includes a terminal 101 and a server 102. The terminal 101 and the server 102 are connected via a wireless or wired network.

The server 102 is configured to train the text mapping model, store the text mapping model after the text mapping model is trained, send target text information to the server 102 through a network connection between the terminal 101 and the server, receive the target text information sent by the terminal 101, map similar text information of the target text information based on the text mapping model, send the similar text information to the terminal 101, and receive the similar text information by the terminal 101.

In one possible implementation, the terminal 101 has installed on it a target application served by the server 102, through which the terminal 101 can implement, for example, a text mapping function. Optionally, the target application is a target application in an operating system of the terminal 101, or a target application provided by a third party. For example, the target application is a text mapping application having a function of text mapping, but of course, the text mapping application can also have other functions, such as a comment function, a question and answer function, a navigation function, a game function, and the like.

The terminal 101 is configured to log in a target application based on a user identifier, send a text expansion request to the server 102 through the target application, where the text expansion request carries target text information, and the server 102 is configured to receive the text expansion request sent by the terminal 101, map similar text information of the target text information based on a text mapping model, send the similar text information to the terminal 101, and receive the similar text information by the terminal 101.

Fig. 2 is a flowchart of a processing method of a text mapping model according to an embodiment of the present application, where the processing method is executed by a computer device, where the computer device is a terminal or a server, and as shown in fig. 2, the method includes:

201. the computer equipment acquires sample text information and first label information, wherein the first label information is the text information of which the similarity with the sample text information is not less than a similarity threshold value.

The sample text information is any type of text information, and for example, the sample text information includes a query sentence, an answer sentence, or a disease description sentence. The similarity threshold is an arbitrary value, and for example, the similarity threshold is 0.8 or 3. The similarity between the first label information and the sample text information is not less than the similarity threshold value, which indicates that the first label information and the sample text information have the same or similar meanings. For example, the sample text information is "simple little problem", and the first tag information is "ultra simple problem".

202. And the computer equipment maps the sample text information based on the text mapping model to obtain predicted text information.

The text mapping model is used for mapping similar text information of any text information. Since the text mapping model is a model to be trained, the current text mapping model may be inaccurate, and predicted text information mapped based on the text mapping model may be similar to the sample text information.

203. The computer device determines second label information based on a first similarity between the predicted text information and the sample text information and a second similarity between the first label information and the sample text information, wherein the second label information is text information with larger similarity between the predicted text information and the sample text information.

Wherein the first similarity is used for representing the similarity between the predicted text information and the sample text information, and the second similarity is used for representing the similarity between the first label information and the sample text information. In the embodiment of the application, after the predicted text information is obtained based on the text mapping model, according to the similarity between the predicted text information and the sample text information and the similarity between the first label information and the predicted text information, the text information most similar to the sample text information is selected from the predicted text information and the first label information as the second label information, that is, the second label information is the text information most similar to the sample text information in the predicted text information and the first label information, so as to ensure the training effect of the subsequent text mapping model. For example, if the first similarity is greater than the second similarity, the second label information is predicted text information, and if the first similarity is not greater than the second similarity, the second label information is the first label information.

204. And training a text mapping model by the computer equipment based on the second label information, the predicted text information and the first similarity corresponding to the predicted text information, wherein the text mapping model is used for mapping similar text information of any text information.

According to the method provided by the embodiment of the application, the text information which is most similar to the sample text information in the predicted text information and the first label information is used as the second label information according to the similarity between the predicted text information and the sample text information, the difference between the predicted text information and the second label text information is considered, the first similarity between the predicted text information and the sample text information is also considered, and the text mapping model is trained based on the second label information, the predicted text information and the first similarity corresponding to the predicted text information, so that the subsequent text mapping model based on the trained text mapping model can map the similar text information of any text information, and the mapping effect of the text mapping model is improved.

On the basis of the embodiment shown in fig. 2, the first similarity and the first loss value corresponding to the predicted text information can be further obtained, and the text mapping model is trained based on the first similarity and the first loss value corresponding to the predicted text information, and the training process is described in the following embodiments.

301. The computer equipment acquires sample text information and first label information, wherein the first label information is the text information of which the similarity with the sample text information is not less than a similarity threshold value.

302. And the computer equipment maps the sample text information based on the text mapping model to obtain predicted text information.

The predicted text information is text information which is mapped by the text mapping model and is possibly similar to the sample text information.

In one possible implementation, this step 302 includes: and the computer equipment maps the sample text information based on the text mapping model to obtain at least one piece of predicted text information.

In the embodiment of the present application, the text mapping model is used for mapping at least one similar text message of any text message.

In one possible implementation, this step 302 includes: the computer equipment encodes sample text information based on a text mapping model to obtain encoding characteristics, decodes the encoding characteristics to obtain a first word, fuses the word characteristics of the first word with the encoding characteristics to obtain a first fusion characteristic, decodes the first fusion characteristic to obtain a second word, fuses the first fusion characteristic with the word characteristics of the second word to obtain a second fusion characteristic, decodes the second fusion characteristic to obtain a third word, and so on until an nth word is obtained, and forms the obtained plurality of words into the predicted text information.

In the embodiment of the application, in the process of generating the predicted text information corresponding to the sample text information based on the text mapping model, each word of the predicted text information is sequentially output, and when each word is output, a plurality of words before the word are relied on to ensure that the output predicted text information is as similar as possible to the sample text information, so that the mapping effect of the text mapping model is ensured. As shown in fig. 4, the text mapping model includes a coding layer and a decoding layer, based on the coding layer, the sample text information is coded to obtain a coding feature, based on the decoding layer, the coding feature is decoded to obtain a first word, based on the decoding layer, the word feature of the first word is fused with the coding feature to obtain a first fusion feature, the first fusion feature is decoded to obtain a second word, based on the decoding layer, the first fusion feature is fused with the word feature of the second word to obtain a second fusion feature, the second fusion feature is decoded to obtain a third word, and so on until an nth word is obtained, the obtained words form the predicted text information, and the predicted text information is obtained by using the obtained words

303. The computer device obtains a third similarity and a fourth similarity between the predicted text information and the sample text information.

The third similarity indicates the difference between words contained in the predicted text information and words contained in the sample text information, and can represent the similarity of the predicted text information and the sample text information in syntax. The fourth similarity indicates the semantic similarity between the predicted text information and the sample text information, and can represent the semantic similarity between the predicted text information and the sample text information.

In a possible implementation manner, a plurality of predicted text information corresponding to the sample text information is obtained based on the text mapping model, and then the step 303 includes: the computer device obtains a third similarity and a fourth similarity between each of the predicted text information and the sample text information.

In one possible implementation manner, the process of obtaining the third similarity between the predicted text information and the sample text information includes the following steps 3031-3034:

3031. and respectively dividing the predicted text information based on the number of at least one character to obtain at least one first word set, wherein the words belonging to the same first word set contain the same number of characters.

The number of characters is any number, for example, the number of characters is 1, 2, 3, or the like. Dividing the predicted text information according to at least one character number to obtain at least one first word set, wherein each first word set corresponds to one character number, the number of the words containing characters belonging to the same first word set is the same, and the number of the words containing characters belonging to the same first word set is the same as the number of the characters corresponding to the first word set. For example, at least one character number includes 1, 2, and 3, and the predicted text information is "how fast i go to the belly", the predicted text information is divided based on the 3 character numbers to obtain 3 first term sets, where the first term set includes "i", "go", and "hungry", that is, each term in the first term set includes 1 character; the second first word set comprises "i'm belly", "belly" and "hungry", i.e. each word in the second first word set comprises 2 characters; the third first word set includes "i'm belly" and "hungry", i.e., each word in the third first word set contains 3 characters.

In one possible implementation, this step 3031 includes: and for each character number, dividing the predicted text information based on the character number and the target step length to obtain a first word set corresponding to the character number.

The target step is the number of words moved each time a word is divided, and the target step is an arbitrary number, for example, the target step is 1. The predictive text information is partitioned, for example based on N-Gram (a statistical language model), resulting in at least one first set of words. For example, starting from the first character of the predicted text information, the marquee with the length equal to the number of the characters is gradually moved according to the target step length to frame the characters in the predicted text information, the characters with the number of the characters in the marquee are taken as a word every time the marquee is moved, and the obtained words form a first word set.

3032. And respectively dividing the sample text information based on the number of at least one character to obtain at least one second word set, wherein the words belonging to the same second word set contain the same number of characters.

And each second word set corresponds to one character number, and the number of characters contained in the words in each second word set is equal to the corresponding character number. Step 3032 is similar to step 3031, and will not be described herein again.

3033. A first number and a second number are determined, the first number indicating a sum of a number of different words in the first set of words and the second set of words for each number of characters, the second number indicating a total number of words in the at least one first set of words and the at least one second set of words.

And for a first word set and a second word set corresponding to any character number, determining the number of different words contained in the first word set and the second word set corresponding to the character number, and determining the sum of the numbers corresponding to at least one character number as the first number.

3034. A ratio of the first number to the second number is determined as a third similarity between the predicted text information and the sample text information.

According to the number of at least one character, dividing the predicted text information into at least one first word set, dividing the sample text information into at least one second word set, wherein the number of words containing characters in the first word sets corresponding to different numbers of characters is different, and determining a third similarity between the predicted text information and the sample text information according to the word sets with different granularity levels, so that the grammatical similarity between the predicted text information and the sample text information can be obtained, and the accuracy of the obtained third similarity is ensured.

In one possible implementation manner, the process of obtaining the fourth similarity between the predicted text information and the sample text information includes the following two manners:

the first mode is as follows: semantic extraction is respectively carried out on the predicted text information and the sample text information to obtain a first semantic feature of the predicted text information and a second semantic feature of the sample text information; and determining the similarity between the first semantic feature and the second semantic feature as a fourth similarity.

The first semantic feature is used for representing the predicted text information, the second semantic feature is used for representing the sample text information, and the first semantic feature and the second semantic feature can be represented in any form, for example, the first semantic feature and the second semantic feature are both represented in a feature vector form.

Optionally, the first semantic feature and the second semantic feature are both expressed in the form of a feature vector, and the determining the fourth similarity includes: and determining the fourth similarity as the product of the first semantic feature vector and the second semantic feature vector.

The second mode is as follows: splicing the predicted text information and the sample text information to obtain spliced text information; semantic extraction is carried out on the spliced text information to obtain a third semantic feature corresponding to the spliced text information; classifying the third semantic features to obtain a classification result; and determining the classification result as a fourth similarity.

The third semantic features are used for representing the spliced text information, and the features of the predicted text information and the features of the sample text information are blended into the third semantic features, namely the third semantic features can represent the predicted text information and the sample text information.

The predicted text information and the sample text information are processed in a splicing and reclassifying mode, so that the change between the predicted text information and the core words in the sample text information can be captured, and whether the predicted text information is the same as the sample text information or not is determined semantically, and the accuracy of the obtained fourth similarity is guaranteed.

For example, the predicted text information is "why the label displays gray", and the sample text information is "why the label displays brown", that is, the core words "gray" and "brown" of the two text information are different, and the fourth similarity obtained in the manner of stitching and reclassifying first is small, so that it can be determined that the predicted text information and the sample text information have different meanings.

Optionally, semantic extraction is performed on the spliced text information in a self-attention mode to obtain a third semantic feature, normalization processing is performed on the third semantic feature to obtain a classification result, and the classification result is determined as a fourth similarity.

Through a self-attention mode, in the process of performing semantic extraction on spliced text information, the predicted text information and the sample text information can be subjected to cross coding, so that a third semantic feature is guaranteed to be a feature obtained by fusing the features of the predicted text information and the features of the sample text information, and then normalization processing is performed on the third semantic feature to determine whether the core words in the predicted text information and the sample text information are the same or not, wherein the classification result represents the similarity degree of the predicted text information and the core words in the sample text information, and therefore the fourth similarity degree is determined.

In the embodiment of the present application, the fourth similarity is obtained in the above two manners, respectively, but in another embodiment, the above two manners can be combined, and the similarity obtained in the above two manners is weighted and fused to obtain the fourth similarity. In a possible implementation manner, the similarity obtained according to the first manner is a fifth similarity, the similarity obtained according to the second manner is a sixth similarity, and the fifth similarity and the sixth similarity are weighted and fused to obtain a fourth similarity.

304. And the computer equipment performs weighted fusion on the third similarity and the fourth similarity to obtain a first similarity between the predicted text information and the sample text information.

Wherein the first similarity is used to represent a degree of similarity between the predicted text information and the sample text information. And considering the similarity between the first text information and the label information in multiple ways to obtain a third similarity and a fourth similarity, and performing weighted fusion on the third similarity and the fourth similarity to ensure the accuracy of the similarity between the obtained predicted text information and the sample text information. The application provides a semantic and grammar combined scoring mechanism, and based on the scoring mechanism, the similarity between predicted text information and sample text information is determined, so that the accuracy of the determined first similarity is ensured.

In one possible implementation manner, the fourth similarity is weighted based on the fifth similarity and the sixth similarity, and the step 303 includes: and performing weighted fusion on the third similarity, the fifth similarity and the sixth similarity to obtain the first similarity.

Optionally, the third similarity, the fifth similarity, the sixth similarity, and the first similarity satisfy the following relationship:

Score(all)＝2*Score(sbert)+2*Score(common_bert)+1*Score(diversity)

wherein Score (all) is used for representing the first similarity, Score (sbert) is used for representing the fifth similarity, and Score (common)_berr) is used to represent the sixth similarity, and score (diversity) is used to represent the third similarity.

It should be noted that, in the embodiment of the present application, the third similarity and the fourth similarity between the predicted text information and the sample text information are obtained first, and then the third similarity and the fourth similarity are weighted and fused to obtain the fourth similarity between the predicted text information and the sample text information, but in another embodiment, the step 303 and the step 304 need not be executed, and the first similarity between the predicted text information and the sample text information can be obtained in other manners.

305. The computer device determines second label information based on a first similarity between the predicted text information and the sample text information and a second similarity between the first label information and the sample text information, wherein the second label information is text information with larger similarity between the predicted text information and the sample text information.

The first similarity is used for representing the similarity between the predicted text information and the sample text information, the second similarity is used for representing the similarity between the first label information and the sample text information, the text information most similar to the sample text information can be determined from the predicted text information and the first label information by comparing the first similarity with the second similarity, and the most similar text information is used as the second label information. The manner of obtaining the second similarity between the first label information and the sample text information is the same as the manner of obtaining the first similarity between the predicted text information and the sample text information, and is not repeated herein.

In one possible implementation, this step 305 includes the following two ways:

the first mode is as follows: and in response to the fact that the largest first similarity in the plurality of first similarities is larger than the second similarity, determining the predicted text information corresponding to the largest first similarity as the second label information, wherein the plurality of first similarities are similarities between the plurality of predicted text information and the sample text information.

The second mode is as follows: and determining the first label information as the second label information in response to the plurality of first similarities not being greater than the second similarity.

In the embodiment of the application, based on a text mapping model, sample text information is mapped to obtain a plurality of predicted text information, a first similarity between each predicted text information and the sample text information and a second similarity between the first label information and the sample text information are determined, the plurality of first similarities are compared with the second similarity to determine the maximum similarity, and information corresponding to the maximum similarity is determined as the second label information. For example, if the maximum similarity is the second similarity, the first tag information corresponding to the second similarity is determined as the second tag information, and if the maximum similarity is a first similarity, the predicted text information corresponding to the first similarity is determined as the second tag information.

In one possible implementation, the method further comprises: and updating the first label information corresponding to the sample text information in the second text information to the second label information when the second label information is the predicted text information.

If the first similarity between the predicted text information and the sample text information is greater than the second similarity between the first label information and the sample text information, the predicted text information is more similar to the sample text information, namely the predicted text information is more suitable to be used as the label information of the sample text information, therefore, the first label information corresponding to the sample text information in the second text information set is updated, so that the second text information in the second text information set and the corresponding label information are more accurate, the quality of the second text information set is ensured to be higher, and the mapping effect of the text mapping model can be ensured when the text mapping model is trained based on the second text information in the second text information set and the corresponding label information.

306. And the computer equipment acquires a first loss value corresponding to the predicted text information based on the second label information and the predicted text information.

The difference between the predicted text information and the second label information can represent the accuracy of the text mapping model, so that a first loss value is determined based on the predicted text information and the second label information, and the text mapping model is trained based on the first loss value.

In one possible implementation, this step 306 includes: and acquiring a first loss value corresponding to each predicted text information based on the second label information and each predicted text information.

In the embodiment of the application, a plurality of predicted text information corresponding to the sample text information can be mapped based on the text mapping model, and then a first loss value corresponding to each predicted text information can be determined based on each predicted text information and the second label information, so that the text mapping model can be trained based on a plurality of first loss values in the following process.

307. And the computer equipment trains the text mapping model based on the first similarity and the first loss value corresponding to the predicted text information.

The first similarity and the first loss value corresponding to the predicted text information can reflect the accuracy of the text mapping model, and the text mapping model is trained based on the first similarity and the first loss value corresponding to the predicted text information so as to improve the mapping effect of the text mapping model.

In one possible implementation, the step 307 comprises the following steps 3071-3074:

3071. and the computer equipment acquires the weight parameter corresponding to each predicted text message respectively based on the difference between the target similarity and the first similarity corresponding to each predicted text message.

The target similarity is an arbitrary value and is used to represent the maximum value of the similarity between the expected predicted text information and the sample text information. For example, the target similarity is 5, and the first similarity corresponding to each predicted text information is not greater than 5. The weight parameter indicates whether each predicted text information is accurate or not and can also reflect the mapping effect of the text mapping model, the loss value can be adjusted based on the weight parameter, the larger the weight parameter is, the more inaccurate the predicted text information is, and the larger the subsequently calculated loss value is; the smaller the weight parameter is, the more accurate the predicted text information is represented, and the smaller the loss value calculated subsequently is.

In a possible implementation manner, the target similarity, the first similarity corresponding to any predicted text information, and the weight parameter satisfy the following relationship:

wherein, forwards is used for representing any weight parameter corresponding to the predicted text information, mu is used for representing a hyper parameter, mu is a constant, score_maxFor representing the similarity of objects, score_realThe similarity calculation unit is used for expressing the first similarity corresponding to the predicted text information.

3072. And the computer equipment carries out weighted average on the first loss values corresponding to the plurality of predicted text information based on the weight parameter corresponding to each predicted text information to obtain a second loss value.

In consideration of the accuracy of each predicted text message, the weighted average is carried out on the first loss values corresponding to the plurality of predicted text messages through the weight parameter corresponding to each predicted text message, so that the accuracy of the obtained second loss value is ensured.

In one possible implementation manner, the weight parameter corresponding to each piece of predicted text information, and the first loss value and the second loss value corresponding to the plurality of pieces of predicted text information satisfy the following relationship:

loss_rl＝mean(loss_mle-batch·rewards)

therein, loss_rlFor representing a second loss value, loss_mle-batchThe first loss value is used for representing the corresponding predicted text information, the rewards is used for representing the corresponding weight parameter of the predicted text information, and mean () is used for representing the averaging function.

3073. The computer device determines an average value of the first loss values corresponding to the plurality of predicted text information as a third loss value.

The first loss value corresponding to each predicted text message can reflect the accuracy of the text mapping model, the first loss values corresponding to a plurality of predicted text messages are comprehensively considered, the average value of the first loss values corresponding to the predicted text messages is used as a third loss value, and the text mapping model is trained by the third loss value subsequently, so that the accuracy of the trained text mapping model is ensured, and the mapping effect of the text mapping model can be improved.

In one possible implementation manner, the third loss value and the first loss values corresponding to the predicted text information satisfy the following relationship:

loss_mle＝mean(loss_mle-batch)

therein, loss_mleFor representing a third loss value, loss_mle-batchThe first loss value is used for representing the corresponding of a plurality of predicted text information, and mean () is used for representing the average value.

3074. The computer device trains the text mapping model based on the second loss value and the third loss value.

The second loss value and the third loss value can reflect the accuracy of the text mapping model, and the text mapping model is trained based on the second loss value and the third loss value so as to improve the mapping effect of the text mapping model.

In one possible implementation, this step 3074 includes: and performing weighted fusion on the second loss value and the third loss value to obtain a fourth loss value, and training the text mapping model based on the fourth loss value.

Optionally, the second loss value, the third loss value, and the fourth loss value satisfy the following relationship:

loss＝λloss_rl+(1-λ)loss_mle

wherein loss is used to represent the fourth loss value, λ is used to represent the hyperparameter, λ is a constant, and loss_rlFor representing a second loss value, loss_mleFor representing the third loss value.

And training the text mapping model through the second loss value and the third loss value, considering the influence of the first similarity between the predicted text information and the sample text information, so that the loss value is reduced under the condition that the predicted text information is sufficiently similar to the sample text information, the adjustment amplitude of the adjusted text model is reduced, and the loss value is increased under the condition that the predicted text information is not similar to the sample text information, so that the convergence and optimization of the text mapping model are accelerated, and the mapping effect of the finally trained text mapping model is ensured.

It should be noted that, in the embodiment of the present application, the text mapping model is trained through the first loss value and the first similarity corresponding to the predicted text information, but in another embodiment, step 306 and step 307 do not need to be executed, and other manners can be adopted to train the text mapping model based on the second label information, the predicted text information, and the first similarity corresponding to the predicted text information.

It should be noted that, in the embodiment of the present application, the text mapping model is only iteratively trained once based on the sample text information and the corresponding first label information, and in another embodiment, the text mapping model can be iteratively trained multiple times according to the

above steps

301 and 307. In a possible implementation manner, in the process of performing iterative training on the text mapping model according to the above step 301-307, in response to the number of iterations reaching the first value, the training on the text mapping model is stopped; or stopping training the text mapping model in response to the fourth loss value of the current iteration number being less than the second numerical value.

The first value is used to represent the maximum value of the iteration number, and both the first value and the second value are arbitrary values, for example, the first value is 100, and the second value is 0.3.

It should be noted that, in the embodiment of the present application, the text mapping model is trained based on only one iteration of one sample text message and corresponding first label information, and in another embodiment, when the text mapping model is trained based on one iteration of a plurality of sample text messages and corresponding first label information of each sample text message, the text mapping model can be trained based on a plurality of sample text messages and corresponding first label information of each sample text message. For example, a plurality of sample text messages and first label information corresponding to each sample text message are obtained from the second text message set, then the predicted text message corresponding to each sample text message is obtained according to the above step 301-306, the first similarity and the first loss value corresponding to each predicted text message are obtained, and the text mapping model is trained according to the above step 307 based on the first similarity and the first loss value corresponding to the plurality of predicted text messages.

In one possible implementation, after step 307, the method further comprises: and mapping the target text information based on the text mapping model to obtain similar text information of the target text information.

The target text information is any text information. After the text mapping model is obtained through training, similar text information of the target text information can be mapped based on the text mapping model.

Optionally, based on the text mapping model, the target text information is mapped to obtain a plurality of similar text information of the target text information.

Moreover, the method and the device provide a scoring mechanism combining semantics and grammar to determine the similarity between the first text information and the corresponding label information, so that the accuracy of the determined similarity is guaranteed.

And in the process of training the text mapping model, the label information is updated, so that the updated label information is the text information which is most similar to the sample text information in the predicted text information and the label text information, the accuracy of the label information is ensured, the text mapping model is trained based on the updated label information, the predicted text information and the first similarity corresponding to the predicted text information, the mapping effect of the text mapping model can be improved, the model is prevented from being trained in a fixed label mode, and the text mapping model has better generalization capability.

On the basis of the embodiment shown in fig. 2, the text mapping model can also be trained in multiple stages, that is, the text mapping model is trained in a first stage based on the first text information in the first text information set and the corresponding label information, then the first text information set is screened, and the text mapping model is trained in a second stage based on the second text information set obtained by screening.

Fig. 5 is a flowchart of a processing method of a text mapping model according to an embodiment of the present application, which is executed by a computer device, and as shown in fig. 5, the method includes:

501. a computer device obtains a first set of textual information.

The first text information set includes a plurality of first text information and tag information corresponding to each first text information, and the first text information is any type of text information, for example, the first text information includes an inquiry sentence, an answer sentence, a disease description sentence, or the like. The label information corresponding to each first text information is text information which is possibly similar to the first text information.

502. The computer device trains the text mapping model based on the first text information in the first text information set and the corresponding label information.

The text mapping model is an arbitrary network model, for example, the text mapping model is Sequence to Sequence (Sequence to Sequence, a neural network model), or is a transform (an attention network model).

And training the text mapping model through the first text information in the first text information set and the corresponding label information, so that the trained text mapping model has preliminary text mapping capability and can map similar text information of any text information.

In one possible implementation, this step 502 includes: and based on the text mapping model, mapping the first text information to obtain predicted text information corresponding to the first text information, and training the text mapping model based on the predicted text information and the label information corresponding to the first text information.

Based on the text mapping model, the predicted text information corresponding to the first text information is mapped, the difference between the predicted text information corresponding to the first text information and the label information can reflect the quality of the mapping effect of the text mapping model, and the text mapping model is trained based on the predicted text information corresponding to the first text information and the label information so as to improve the mapping effect of the text mapping model.

Optionally, the training of the text mapping model includes: and determining a fifth loss value based on the predicted text information and the label information corresponding to the first text information, and training the text mapping model based on the fifth loss value.

The fifth loss value indicates the difference degree between the predicted text information corresponding to the first text information and the label information, and the mapping effect of the text mapping model can be reflected. For example, a fifth loss value is determined based on the predicted text information and the tag information corresponding to the first text information by using a cross entropy loss function. And training the text mapping model through a fifth loss value so as to improve the mapping effect of the text mapping model.

In one possible implementation, this step 502 includes: and performing iterative training on the text mapping model based on the plurality of first text messages and the label information corresponding to each first text message.

And performing iterative training on the text mapping model for multiple times to improve the mapping effect of the text mapping model as much as possible.

Optionally, in the process of performing iterative training on the text mapping model, in response to the number of iterations reaching a third value, stopping training the text mapping model; or stopping training the text mapping model in response to the fifth loss value of the current iteration number being less than the fourth numerical value.

The third value is used to represent the maximum value of the iteration number, and both the third value and the fourth value are arbitrary values, for example, the third value is 100, and the fourth value is 0.3.

503. The computer device determines a similarity between each first text information and the corresponding tag information.

The similarity between each piece of first text information and the corresponding label information can embody the similarity between the first text information and the corresponding label information.

In one possible implementation, this step 503 includes: and for any first text message and corresponding label information, acquiring seventh similarity and eighth similarity between the first text message and the label information, and performing weighted fusion on the seventh similarity and the eighth similarity to obtain the similarity between the first text message and the label information.

The seventh similarity indicates a difference between words contained in the first text information and the tag information, and can indicate a grammatical similarity between the first text information and the tag information. The eighth similarity indicates semantic similarity between the first text information and the tag information, and can indicate a semantic similarity between the first text information and the tag information. And considering the similarity between the first text information and the label information in multiple ways to obtain a seventh similarity and an eighth similarity, and performing weighted fusion on the seventh similarity and the eighth similarity to ensure the accuracy of the obtained similarity between the first text information and the label information. The application provides a scoring mechanism combining semantics and grammar, and the similarity between first text information and corresponding label information is determined based on the scoring mechanism, so that the accuracy of the determined similarity is ensured.

Optionally, the process of obtaining the seventh similarity between the first text information and the tag information includes:

5031. and respectively dividing the first text information based on the number of at least one character to obtain at least one third word set, wherein the words belonging to the same third word set contain the same number of characters.

5032. And respectively dividing the label information based on the number of at least one character to obtain at least one fourth word set, wherein the words belonging to the same fourth word set contain the same number of characters.

5033. A third number and a fourth number are determined, the third number indicating a sum of numbers of different words in the third set of words and the fourth set of words for each character number, the fourth number indicating a total number of words in the at least one third set of words and the at least one fourth set of words.

5034. And determining the ratio of the third number to the fourth number as a seventh similarity between the first text information and the label information.

The steps 5031-5034 are the same as the steps 3031-3034, and will not be described herein again.

Optionally, the process of obtaining the eighth similarity between the first text information and the tag information includes the following two ways:

the first mode is as follows: semantic extraction is respectively carried out on the first text information and the label information to obtain a fourth semantic feature of the first text information and a fifth semantic feature of the label information; and determining the similarity between the fourth semantic feature and the fifth semantic feature as an eighth similarity.

The first way of obtaining the eighth similarity is similar to the first way of obtaining the fourth similarity, and is not repeated herein.

The second mode is as follows: splicing the first text information and the label information to obtain first spliced text information; semantic extraction is carried out on the first spliced text information to obtain a sixth semantic feature corresponding to the first spliced text information; classifying the sixth semantic features to obtain a classification result; and determining the classification result as the eighth similarity.

The second way of obtaining the eighth similarity is similar to the second way of obtaining the fourth similarity, and is not repeated herein.

In the embodiment of the present application, the eighth similarity is obtained in the two manners described above, respectively, but in another embodiment, the two manners described above can be combined, and the similarities obtained in the two manners are weighted and fused to obtain the eighth similarity. In a possible implementation manner, the similarity obtained according to the first manner is a ninth similarity, the similarity obtained according to the second manner is a tenth similarity, and the ninth similarity and the tenth similarity are weighted and fused to obtain an eighth similarity.

504. And the computer equipment screens out at least one piece of second text information with the similarity larger than a similarity threshold value from the first text information set based on the similarity corresponding to each piece of first text information, and the screened out second text information and the corresponding label information form a second text information set.

The similarity threshold is an arbitrary threshold, and for example, the similarity threshold is 0.8, or 3.

In this embodiment of the application, the first text information set includes a plurality of first text information and label information corresponding to each first text information, each first text information and corresponding label information form an information combination, the quality of the information combination in the first text information set is different, and a poor information combination may exist, that is, the first text information in the information combination is not similar to the label information, and if the text mapping model is trained according to the poor information combination, the mapping effect of the text mapping model is poor. Therefore, the text information in the first text information set and the corresponding label information are screened to generate a second text information set, so that the similarity between each second text information in the second text information set and the corresponding label information is greater than the similarity threshold value, that is, the second text information set comprises a high-quality information combination, and then the text mapping model is trained based on the second text information in the second text information set and the corresponding label information, so that the mapping effect of the text mapping model can be ensured.

It should be noted that, in the embodiment of the present application, the text mapping model is trained based on the first text information in the first text information set and the corresponding label information, and then the second text information set is screened out based on the first text information set, in another embodiment, the above-mentioned step 503 and step 504 can be performed first, the second text information set is screened out, and then the above-mentioned step 502 is performed, and the order of performing the steps is not limited in the present application.

505. And the computer equipment acquires sample text information and the first label information from the second text information set, wherein the sample text information is any one of the second text information.

In this embodiment of the application, the second text information set includes at least one piece of second text information and tag information corresponding to each piece of second text information, any one piece of second text information is obtained from the second text information set as sample text information, and first tag information corresponding to the sample text information is obtained.

It should be noted that in the embodiment of the present application, the first text information set is first screened to obtain the second text information set, and the sample text information and the first tag information are obtained from the second text information set, but in another embodiment, step 504 and step 505 do not need to be executed, and other manners can be adopted, and based on the similarity corresponding to each first text information, the sample text information and the first tag information with the similarity greater than the similarity threshold value are screened from the first text information set.

It should be noted that, in the embodiment of the present application, a text mapping model is initially trained based on a first text information set, and then, first text information in the first text information set is screened to construct a second text information set, and sample text information and first label information are obtained from the second text information set, while in another embodiment, it is not necessary to train the text mapping model based on the first text information set, and it is also not necessary to perform

step

501 and 504, and it is possible to obtain the sample text information and the first label information in other ways, where the similarity between the first label information and the sample text information is not less than a similarity threshold.

506. And the computer equipment maps the sample text information based on the text mapping model to obtain predicted text information.

507. The computer device determines second label information based on a first similarity between the predicted text information and the sample text information and a second similarity between the first label information and the sample text information, wherein the second label information is text information with larger similarity between the predicted text information and the sample text information.

508. And training a text mapping model by the computer equipment based on the second label information, the predicted text information and the first similarity corresponding to the predicted text information, wherein the text mapping model is used for mapping similar text information of any text information.

The

steps

506 and 508 are similar to the

steps

302 and 307, and are not described herein again.

In addition, in the process of training the text mapping model, the label information corresponding to the sample text information in the second text information set is updated, so that the updated label information is the text information which is most similar to the sample text information in the predicted text information and the label text information, the accuracy of the label information is ensured, the text mapping model is trained, the mapping effect of the text mapping model can be improved, the model is prevented from being trained in a fixed label mode, and the text mapping model has better generalization capability.

And the initial first text information set is screened, the low-quality information combination in the first text information set is cleaned, so that the similarity between the second text information in the screened second text information and the corresponding label information is greater than a similarity threshold value, the quality of the information combination in the second text information set is improved, the second text information in the second text information set and the corresponding label information are used for training the text mapping model, the accuracy of the sample information of the text mapping model is guaranteed, and the training effect of the subsequent text mapping model is guaranteed.

In addition, the text mapping model is trained in multiple stages, the text mapping model is initially trained on the basis of the first text information set, so that the text mapping model has initial text mapping capacity, then the text mapping model is trained through the screened second text information set, and the mapping effect of the text mapping model is further improved.

As shown in fig. 6, the text mapping model is trained in two stages, the first training stage trains the text mapping model based on the first text information set, the second training stage screens out the second text information set based on the first text information set by using a scoring mechanism, and trains the text mapping model based on the second text information set. And in the training process, a scoring mechanism is adopted to determine a first similarity between the sample text information and the predicted text information and a second similarity between the first label information and the sample text information, the second label information is determined based on the first similarity and the second similarity, and the text mapping model is subjected to reinforced training based on the second label information, the predicted text information and the first similarity corresponding to the predicted text information.

As shown in fig. 7, the processing method of the text mapping model provided in the embodiment of the present application is compared with the processing method of the text mapping model in the related art. The related art 1 is to train the text mapping model directly based on the original training data set, and the related art 2 is to filter the original training data set, adopt the cross entropy loss function, and train the text mapping model through the filtered training data set. By comparison, the training data volume obtained by the method provided by the embodiment of the application is minimum, but the quantity of the similar text information mapped by the text mapping model provided by the embodiment of the application is large, and the quality of the mapped similar text information is high, namely the text mapping model provided by the embodiment of the application has better performance and better mapping effect.

The processing method of the text mapping model provided by the embodiment of the application can be applied to various scenes, for example, the text mapping model can be used for expanding a question and answer knowledge base in an intelligent question and answer scene. As shown in fig. 8, a question-answer knowledge base of an intelligent dialogue robot is configured with a plurality of questions and answers to each question, a question-answer editing interface is displayed on an editing interface of the question-answer knowledge base in response to a trigger operation of an editing option corresponding to any question, a user can edit the question and the corresponding answer in the question-answer editing interface, the question input in the question-answer editing interface is mapped based on a text mapping model to obtain a plurality of similar questions of the question, a prompt mark 801 is displayed to prompt the user to map the plurality of similar questions of the currently input question, the user clicks the similar question option, the plurality of similar questions are displayed, the user clicks a save option, the input question is associated with the corresponding plurality of similar questions, the plurality of questions corresponding to each answer in the question-answer knowledge base are enriched, and when a subsequent user has a dialogue with the intelligent dialogue robot, the intelligent dialogue robot can search the answer of any question from the question-answer knowledge base when the user inputs any question, so that the experience of the user is improved, and the situation that the intelligent dialogue robot cannot answer the question due to few questions configured in the question-answer knowledge base is avoided.

Fig. 9 is a schematic structural diagram of a processing apparatus for a text mapping model according to an embodiment of the present application, and as shown in fig. 9, the apparatus includes:

an obtaining module 901, configured to obtain sample text information and first tag information, where the first tag information is text information whose similarity with the sample text information is not less than a similarity threshold;

a mapping module 902, configured to map the sample text information based on a text mapping model to obtain predicted text information;

a determining module 903, configured to determine second label information based on a first similarity between the predicted text information and the sample text information and a second similarity between the first label information and the sample text information, where the second label information is text information with a larger similarity between the predicted text information and the sample text information in the first label information;

the training module 904 is configured to train a text mapping model based on the second label information, the predicted text information, and the first similarity corresponding to the predicted text information, where the text mapping model is configured to map similar text information of any text information.

In one possible implementation, as shown in fig. 10, the apparatus further includes:

the obtaining module 901 is further configured to obtain a third similarity and a fourth similarity between the predicted text information and the sample text information, where the third similarity indicates a difference between words included in the predicted text information and the sample text information, and the fourth similarity indicates a semantic similarity between the predicted text information and the sample text information;

and a fusion module 905, configured to perform weighted fusion on the third similarity and the fourth similarity to obtain the first similarity.

In another possible implementation manner, the obtaining module 901 is configured to divide the predicted text information based on the number of at least one character to obtain at least one first term set, where the terms belonging to the same first term set include the same number of characters; respectively dividing the sample text information based on the number of at least one character to obtain at least one second word set, wherein the words belonging to the same second word set contain the same number of characters; determining a first number and a second number, the first number indicating a sum of numbers of different words in the first set of words and the second set of words corresponding to each number of characters, the second number indicating a total number of words in at least one of the first set of words and the at least one of the second set of words; a ratio of the first number to the second number is determined as a third similarity between the predicted text information and the sample text information.

In another possible implementation manner, the obtaining module 901 is configured to perform semantic extraction on the predicted text information and the sample text information respectively to obtain a first semantic feature of the predicted text information and a second semantic feature of the sample text information; and determining the similarity between the first semantic feature and the second semantic feature as a fourth similarity.

In another possible implementation manner, the obtaining module 901 is configured to splice the predicted text information and the sample text information to obtain spliced text information; semantic extraction is carried out on the spliced text information to obtain a third semantic feature corresponding to the spliced text information; classifying the third semantic features to obtain a classification result; and determining the classification result as a fourth similarity.

In another possible implementation manner, the training module 904 is configured to obtain a first loss value corresponding to the predicted text information based on the second label information and the predicted text information; and training the text mapping model based on the first similarity and the first loss value corresponding to the predicted text information.

In another possible implementation manner, the training module 904 is configured to obtain a weight parameter corresponding to each piece of predicted text information based on a difference between the target similarity and the first similarity corresponding to each piece of predicted text information; based on the weight parameter corresponding to each predicted text message, carrying out weighted average on the first loss values corresponding to the plurality of predicted text messages to obtain a second loss value; determining the average value of the first loss values corresponding to the plurality of predicted text information as a third loss value; and training the text mapping model based on the second loss value and the third loss value.

In another possible implementation manner, as shown in fig. 10, the obtaining module 901 includes:

an obtaining unit 9011, configured to obtain a first text information set, where the first text information set includes a plurality of first text information and tag information corresponding to each first text information;

a determining unit 9012, configured to determine a similarity between each piece of first text information and corresponding piece of tag information;

the screening unit 9013 is configured to screen out, from the first text information set, sample text information and first label information, of which the similarity is greater than a similarity threshold, based on the similarity corresponding to each piece of first text information.

In another possible implementation manner, the screening unit 9013 is configured to screen, based on the similarity corresponding to each first text message, at least one second text message of which the similarity is greater than a similarity threshold from the first text message set, and configure the screened second text message and the corresponding tag information into a second text message set; and acquiring sample text information and first label information from the second text information set, wherein the sample text information is any one of the second text information.

In another possible implementation manner, the obtaining module 901 is further configured to train a text mapping model based on the first text information in the first text information set and the corresponding label information.

In another possible implementation manner, the determining module 903 is configured to determine, in response to that a maximum first similarity among the multiple first similarities is greater than the second similarity, the predicted text information corresponding to the maximum first similarity as the second label information, where the multiple first similarities are similarities between the multiple predicted text information and the sample text information; or, in response to the plurality of first similarities not being greater than the second similarity, determining the first tag information as the second tag information.

In another possible implementation manner, the mapping module 902 is further configured to map the target text information based on a text mapping model to obtain similar text information of the target text information.

It should be noted that: the processing apparatus of the text mapping model provided in the foregoing embodiment is only illustrated by the division of the functional modules, and in practical applications, the functions may be distributed by different functional modules according to needs, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the processing apparatus of the text mapping model and the processing method of the text mapping model provided in the above embodiments belong to the same concept, and specific implementation processes thereof are described in detail in the method embodiments and are not described herein again.

The embodiment of the present application further provides a computer device, where the computer device includes a processor and a memory, where the memory stores at least one computer program, and the at least one computer program is loaded by the processor and executed to implement the operations performed by the processing method of the text mapping model according to the above embodiments.

Optionally, the computer device is provided as a terminal. Fig. 11 shows a block diagram of a terminal 1100 according to an exemplary embodiment of the present application. The terminal 1100 may be a portable mobile terminal such as: a smart phone, a tablet computer, an MP3 player (Moving Picture Experts Group Audio Layer III, motion video Experts compression standard Audio Layer 3), an MP4 player (Moving Picture Experts Group Audio Layer IV, motion video Experts compression standard Audio Layer 4), a notebook computer, or a desktop computer. Terminal 1100 may also be referred to by other names such as user equipment, portable terminal, laptop terminal, desktop terminal, and so forth.

The terminal 1100 includes: a processor 1101 and a memory 1102.

Processor 1101 may include one or more processing cores, such as a 4-core processor, an 8-core processor, or the like. The processor 1101 may be implemented in at least one hardware form of a DSP (Digital Signal Processing), an FPGA (Field-Programmable Gate Array), and a PLA (Programmable Logic Array). The processor 1101 may also include a main processor and a coprocessor, where the main processor is a processor for Processing data in an awake state, and is also called a Central Processing Unit (CPU); a coprocessor is a low power processor for processing data in a standby state. In some embodiments, the processor 1101 may be integrated with a GPU (Graphics Processing Unit) that is responsible for rendering and rendering content that the display screen needs to display. In some embodiments, the processor 1101 may further include an AI (Artificial Intelligence) processor for processing computing operations related to machine learning.

Memory 1102 may include one or more computer-readable storage media, which may be non-transitory. Memory 1102 can also include high-speed random access memory, as well as non-volatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In some embodiments, a non-transitory computer readable storage medium in memory 1102 is used to store at least one computer program for execution by processor 1101 to implement the processing methods of the text mapping models provided by the method embodiments herein.

In some embodiments, the terminal 1100 may further include: a peripheral interface 1103 and at least one peripheral. The processor 1101, memory 1102 and peripheral interface 1103 may be connected by a bus or signal lines. Various peripheral devices may be connected to the peripheral interface 1103 by buses, signal lines, or circuit boards. Specifically, the peripheral device includes: at least one of radio frequency circuitry 1104, display screen 1105, camera assembly 1106, audio circuitry 1107, positioning assembly 1108, and power supply 1109.

The peripheral interface 1103 may be used to connect at least one peripheral associated with I/O (Input/Output) to the processor 1101 and the memory 1102. In some embodiments, the processor 1101, memory 1102, and peripheral interface 1103 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 1101, the memory 1102 and the peripheral device interface 1103 may be implemented on separate chips or circuit boards, which is not limited by this embodiment.

The Radio Frequency circuit 1104 is used to receive and transmit RF (Radio Frequency) signals, also called electromagnetic signals. The radio frequency circuit 1104 communicates with communication networks and other communication devices via electromagnetic signals. The radio frequency circuit 1104 converts an electric signal into an electromagnetic signal to transmit, or converts a received electromagnetic signal into an electric signal. Optionally, the radio frequency circuit 1104 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a subscriber identity module card, and so forth. The radio frequency circuit 1104 may communicate with other terminals via at least one wireless communication protocol. The wireless communication protocols include, but are not limited to: the world wide web, metropolitan area networks, intranets, generations of mobile communication networks (2G, 3G, 4G, and 5G), Wireless local area networks, and/or WiFi (Wireless Fidelity) networks. In some embodiments, the rf circuit 1104 may further include NFC (Near Field Communication) related circuits, which are not limited in this application.

The display screen 1105 is used to display a UI (User Interface). The UI may include graphics, text, icons, video, and any combination thereof. When the display screen 1105 is a touch display screen, the display screen 1105 also has the ability to capture touch signals on or over the surface of the display screen 1105. The touch signal may be input to the processor 1101 as a control signal for processing. At this point, the display screen 1105 may also be used to provide virtual buttons and/or a virtual keyboard, also referred to as soft buttons and/or a soft keyboard. In some embodiments, display 1105 may be one, disposed on a front panel of terminal 1100; in other embodiments, the display screens 1105 can be at least two, respectively disposed on different surfaces of the terminal 1100 or in a folded design; in other embodiments, display 1105 can be a flexible display disposed on a curved surface or on a folded surface of terminal 1100. Even further, the display screen 1105 may be arranged in a non-rectangular irregular pattern, i.e., a shaped screen. The Display screen 1105 may be made of LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode), and the like.

Camera assembly 1106 is used to capture images or video. Optionally, camera assembly 1106 includes a front camera and a rear camera. The front camera is arranged on the front panel of the terminal, and the rear camera is arranged on the back of the terminal. In some embodiments, the number of the rear cameras is at least two, and each rear camera is any one of a main camera, a depth-of-field camera, a wide-angle camera and a telephoto camera, so that the main camera and the depth-of-field camera are fused to realize a background blurring function, and the main camera and the wide-angle camera are fused to realize panoramic shooting and VR (Virtual Reality) shooting functions or other fusion shooting functions. In some embodiments, camera assembly 1106 may also include a flash. The flash lamp can be a monochrome temperature flash lamp or a bicolor temperature flash lamp. The double-color-temperature flash lamp is a combination of a warm-light flash lamp and a cold-light flash lamp, and can be used for light compensation at different color temperatures.

The audio circuitry 1107 may include a microphone and a speaker. The microphone is used for collecting sound waves of a user and the environment, converting the sound waves into electric signals, and inputting the electric signals to the processor 1101 for processing or inputting the electric signals to the radio frequency circuit 1104 to achieve voice communication. For stereo capture or noise reduction purposes, multiple microphones may be provided, each at a different location of terminal 1100. The microphone may also be an array microphone or an omni-directional pick-up microphone. The speaker is used to convert electrical signals from the processor 1101 or the radio frequency circuit 1104 into sound waves. The loudspeaker can be a traditional film loudspeaker or a piezoelectric ceramic loudspeaker. When the speaker is a piezoelectric ceramic speaker, the speaker can be used for purposes such as converting an electric signal into a sound wave audible to a human being, or converting an electric signal into a sound wave inaudible to a human being to measure a distance. In some embodiments, the audio circuitry 1107 may also include a headphone jack.

Positioning component 1108 is used to locate the current geographic position of terminal 1100 for purposes of navigation or LBS (Location Based Service). The Positioning component 1108 may be a Positioning component based on the Global Positioning System (GPS) in the united states, the beidou System in china, or the galileo System in russia.

Power supply 1109 is configured to provide power to various components within terminal 1100. The power supply 1109 may be alternating current, direct current, disposable or rechargeable. When the power supply 1109 includes a rechargeable battery, the rechargeable battery may be a wired rechargeable battery or a wireless rechargeable battery. The wired rechargeable battery is a battery charged through a wired line, and the wireless rechargeable battery is a battery charged through a wireless coil. The rechargeable battery may also be used to support fast charge technology.

In some embodiments, terminal 1100 can also include one or more sensors 1110. The one or more sensors 1110 include, but are not limited to: acceleration sensor 1111, gyro sensor 1112, pressure sensor 1113, fingerprint sensor 1114, optical sensor 1115, and proximity sensor 1116.

Acceleration sensor 1111 may detect acceleration levels in three coordinate axes of a coordinate system established with terminal 1100. For example, the acceleration sensor 1111 may be configured to detect components of the gravitational acceleration in three coordinate axes. The processor 1101 may control the display screen 1105 to display the user interface in a landscape view or a portrait view according to the gravitational acceleration signal collected by the acceleration sensor 1111. The acceleration sensor 1111 may also be used for acquisition of motion data of a game or a user.

The gyro sensor 1112 may detect a body direction and a rotation angle of the terminal 1100, and the gyro sensor 1112 may cooperate with the acceleration sensor 1111 to acquire a 3D motion of the user with respect to the terminal 1100. From the data collected by gyroscope sensor 1112, processor 1101 may implement the following functions: motion sensing (such as changing the UI according to a user's tilting operation), image stabilization at the time of photographing, game control, and inertial navigation.

Pressure sensor 1113 may be disposed on a side bezel of terminal 1100 and/or underlying display screen 1105. When the pressure sensor 1113 is disposed on the side frame of the terminal 1100, the holding signal of the terminal 1100 from the user can be detected, and the processor 1101 performs left-right hand recognition or shortcut operation according to the holding signal collected by the pressure sensor 1113. When the pressure sensor 1113 is disposed at the lower layer of the display screen 1105, the processor 1101 controls the operability control on the UI interface according to the pressure operation of the user on the display screen 1105. The operability control comprises at least one of a button control, a scroll bar control, an icon control and a menu control.

The fingerprint sensor 1114 is configured to collect a fingerprint of the user, and the processor 1101 identifies the user according to the fingerprint collected by the fingerprint sensor 1114, or the fingerprint sensor 1114 identifies the user according to the collected fingerprint. Upon recognizing that the user's identity is a trusted identity, the user is authorized by the processor 1101 to perform relevant sensitive operations including unlocking the screen, viewing encrypted information, downloading software, paying for and changing settings, etc. Fingerprint sensor 1114 may be disposed on the front, back, or side of terminal 1100. When a physical button or vendor Logo is provided on the terminal 1100, the fingerprint sensor 1114 may be integrated with the physical button or vendor Logo.

Optical sensor 1115 is used to collect ambient light intensity. In one embodiment, the processor 1101 may control the display brightness of the display screen 1105 based on the ambient light intensity collected by the optical sensor 1115. Specifically, when the ambient light intensity is high, the display brightness of the display screen 1105 is increased; when the ambient light intensity is low, the display brightness of the display screen 1105 is reduced. In another embodiment, processor 1101 may also dynamically adjust the shooting parameters of camera assembly 1106 based on the ambient light intensity collected by optical sensor 1115.

A proximity sensor 1116, also referred to as a distance sensor, is provided on the front panel of terminal 1100. Proximity sensor 1116 is used to capture the distance between the user and the front face of terminal 1100. In one embodiment, when the proximity sensor 1116 detects that the distance between the user and the front face of the terminal 1100 is gradually decreased, the display screen 1105 is controlled by the processor 1101 to switch from a bright screen state to a dark screen state; when the proximity sensor 1116 detects that the distance between the user and the front face of the terminal 1100 becomes progressively larger, the display screen 1105 is controlled by the processor 1101 to switch from a breath-screen state to a light-screen state.

Those skilled in the art will appreciate that the configuration shown in fig. 11 does not constitute a limitation of terminal 1100, and may include more or fewer components than those shown, or may combine certain components, or may employ a different arrangement of components.

Optionally, the computer device is provided as a server. Fig. 12 is a schematic structural diagram of a server 1200 according to an embodiment of the present application, where the server 1200 may generate a relatively large difference due to different configurations or performances, and may include one or more processors (CPUs) 1201 and one or more memories 1202, where the memory 1202 stores at least one computer program, and the at least one computer program is loaded and executed by the processors 1201 to implement the methods provided by the foregoing method embodiments. Of course, the server may also have components such as a wired or wireless network interface, a keyboard, and an input/output interface, so as to perform input/output, and the server may also include other components for implementing the functions of the device, which are not described herein again.

The embodiment of the present application further provides a computer-readable storage medium, in which at least one computer program is stored, and the at least one computer program is loaded and executed by a processor to implement the operations performed by the processing method of the text mapping model of the above embodiment.

The embodiment of the present application further provides a computer program product or a computer program, and provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the computer program implements the operations performed by the processing method of the text mapping model according to the above embodiment.

It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.

The above description is only an alternative embodiment of the present application and should not be construed as limiting the present application, and any modification, equivalent replacement, or improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for processing a text mapping model, the method comprising:

mapping the sample text information based on a text mapping model to obtain predicted text information;

2. The method of claim 1, wherein before determining second label information based on a first similarity between the predicted text information and the sample text information and a second similarity between the first label information and the sample text information, the method further comprises:

acquiring a third similarity and a fourth similarity between the predicted text information and the sample text information, wherein the third similarity indicates a difference condition of words contained in the predicted text information and the sample text information, and the fourth similarity indicates a semantic similarity condition between the predicted text information and the sample text information;

and performing weighted fusion on the third similarity and the fourth similarity to obtain the first similarity.

3. The method of claim 2, wherein obtaining a third similarity between the predicted text information and the sample text information comprises:

dividing the predicted text information respectively based on the number of at least one character to obtain at least one first word set, wherein the words belonging to the same first word set contain the same number of characters;

based on the number of at least one character, dividing the sample text information respectively to obtain at least one second word set, wherein the words belonging to the same second word set contain the same number of characters;

determining a first number and a second number, the first number indicating a sum of a number of different words in the first set of words and the second set of words for each of the character numbers, the second number indicating a total number of words in at least one of the first set of words and at least one of the second set of words;

determining a ratio of the first number to the second number as the third degree of similarity between the predicted text information and the sample text information.

4. The method of claim 2, wherein obtaining a fourth similarity between the predicted text information and the sample text information comprises:

semantic extraction is respectively carried out on the predicted text information and the sample text information to obtain a first semantic feature of the predicted text information and a second semantic feature of the sample text information;

determining a similarity between the first semantic feature and the second semantic feature as the fourth similarity.

5. The method of claim 2, wherein obtaining a fourth similarity between the predicted text information and the sample text information comprises:

splicing the predicted text information and the sample text information to obtain spliced text information;

semantic extraction is carried out on the spliced text information to obtain a third semantic feature corresponding to the spliced text information;

classifying the third semantic features to obtain a classification result;

determining the classification result as the fourth similarity.

6. The method of claim 1, wherein training the text mapping model based on the second label information, the predicted text information, and the first similarity corresponding to the predicted text information comprises:

acquiring a first loss value corresponding to the predicted text information based on the second label information and the predicted text information;

and training the text mapping model based on the first similarity and the first loss value corresponding to the predicted text information.

7. The method of claim 6, wherein training the text mapping model based on the first similarity and the first loss value corresponding to the predicted text information comprises:

respectively acquiring a weight parameter corresponding to each predicted text message based on a difference value between the target similarity and the first similarity corresponding to each predicted text message;

based on the weight parameter corresponding to each predicted text message, carrying out weighted average on the first loss values corresponding to the plurality of predicted text messages to obtain a second loss value;

determining an average value of the first loss values corresponding to the plurality of predicted text information as a third loss value;

training the text mapping model based on the second loss value and the third loss value.

8. The method of claim 1, wherein the obtaining sample text information and first label information comprises:

acquiring a first text information set, wherein the first text information set comprises a plurality of first text information and label information corresponding to each first text information;

determining similarity between each first text message and corresponding label information;

and based on the similarity corresponding to each piece of first text information, screening out the sample text information and the first label information with the similarity larger than the similarity threshold value from the first text information set.

9. The method according to claim 8, wherein the step of screening out the sample text information and the first label information with a similarity greater than the similarity threshold from the first text information set based on the similarity corresponding to each of the first text information comprises:

screening at least one piece of second text information with the similarity larger than the similarity threshold value from the first text information set based on the similarity corresponding to each piece of first text information, and enabling the screened second text information and corresponding label information to form a second text information set;

and acquiring the sample text information and the first label information from the second text information set, wherein the sample text information is any one of the second text information.

10. The method of claim 8, wherein before mapping the sample text information based on the text mapping model to obtain predicted text information, the method further comprises:

and training the text mapping model based on the first text information in the first text information set and the corresponding label information.

11. The method according to any one of claims 1-10, wherein determining second label information based on a first similarity between the predicted text information and the sample text information and a second similarity between the first label information and the sample text information comprises:

determining predicted text information corresponding to the largest first similarity as the second label information in response to the largest first similarity being larger than the second similarity among the plurality of first similarities, the plurality of first similarities being similarities between the plurality of predicted text information and the sample text information; alternatively, the first and second electrodes may be,

and determining the first label information as the second label information in response to the plurality of first similarities not being greater than the second similarity.

12. The method according to any one of claims 1-10, wherein after the training of the text mapping model based on the second label information, the predicted text information, and the first similarity corresponding to the predicted text information, the method further comprises:

and mapping the target text information based on the text mapping model to obtain similar text information of the target text information.

13. An apparatus for processing a text mapping model, the apparatus comprising:

14. A computer device comprising a processor and a memory, the memory having stored therein at least one computer program, the at least one computer program being loaded and executed by the processor to perform the operations performed by the method of processing a text mapping model according to any of claims 1 to 12.

15. A computer-readable storage medium, having stored thereon at least one computer program, which is loaded and executed by a processor to perform the operations performed by the method for processing a text mapping model according to any of claims 1 to 12.

16. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, is adapted to carry out the operations of the method of processing a text mapping model according to any of the claims 1 to 12.