WO2021151306A1

WO2021151306A1 - Method and apparatus for smart analysis of question and answer linguistic material, electronic device, and readable storage medium

Info

Publication number: WO2021151306A1
Application number: PCT/CN2020/119102
Authority: WO
Inventors: 陈桢博; 郑立颖; 徐亮
Original assignee: 平安科技（深圳）有限公司
Priority date: 2020-06-19
Filing date: 2020-09-29
Publication date: 2021-08-05
Also published as: CN111563152A

Abstract

A method for smart analysis of question and answer linguistic material and an analysis apparatus, an electronic device, and a computer storage medium, relating to artificial intelligence technology, comprising: inputting a historical question and answer corpus into a pre-constructed data enhancement model to obtain an enhanced corpus (S1); using the historical question and answer corpus to train a pre-constructed convolutional neural network model to obtain an initial classification prediction model (S2); using the initial classification prediction model to perform analysis and filtering on the enhanced corpus to obtain an enhanced sub-corpus (S3); using the enhanced sub-corpus and the historical question and answer corpus to train the convolutional neural network model to obtain an analysis prediction model (S4); and using the analysis prediction model to analyze a user linguistic material text to be analyzed to obtain a final analysis result (S5). The present invention further relates to blockchain technology, and the data used for model training can be stored in a blockchain. The method described above can improve the accuracy of smart analysis of question and answer linguistic material.

Description

Intelligent question and answer corpus analysis method, device, electronic equipment and readable storage medium

This application requires the priority of a Chinese patent application filed with the Chinese Patent Office on June 19, 2020, the application number is CN202010564937.7, and the invention title is "Intelligent Question and Answer Corpus Analysis Method, Device, Electronic Equipment, and Readable Storage Medium". The entire content is incorporated into this application by reference.

Technical field

This application relates to the field of artificial intelligence, and in particular to a method, device, electronic device, and readable storage medium for analyzing intelligent question and answer corpus.

Background technique

With the development of artificial intelligence, machines are becoming more and more intelligent. The analysis of human behavior through human language is no longer a human patent. Machines can also analyze and judge human behavior through corpus. Reduced labor costs.

At present, due to the limitation of computing power, the machine cannot analyze any corpus. It can only analyze and judge the user’s question and answer corpus. The so-called question and answer corpus refers to the corpus that the user answers to the preset questions. The intelligent analysis of the question and answer corpus is applied in Many aspects of life, such as: mobile phone manufacturers perform mobile phone system version evaluation by analyzing user question and answer corpus, doctors performing preliminary screening for mental illness by analyzing patient’s question and answer corpus, and human resources by analyzing interviewer’s question and answer corpus Interview analysis.

However, the inventor realizes that the user's question and answer corpus data is small and difficult to obtain, resulting in a small amount of training data for the deep learning model of the existing intelligent question answering corpus analysis system, and the accuracy of the model is not high.

Summary of the invention

This application provides an intelligent question-and-answer corpus analysis method, device, electronic equipment, and computer-readable storage medium.

An intelligent question-and-answer corpus analysis method provided by this application includes:

Obtain a historical question and answer corpus, and input the historical question and answer corpus into a pre-built data enhancement model to obtain an enhanced corpus;

Use the historical question and answer corpus to train a pre-built convolutional neural network model to obtain an initial classification prediction model;

Using the initial classification prediction model to analyze and screen the enhanced corpus to obtain an enhanced corpus subset;

Training the convolutional neural network model by using the enhanced corpus subset and the historical question and answer corpus to obtain an analysis and prediction model;

The user corpus text to be analyzed is obtained, and the analysis prediction model is used to analyze the user corpus text to be analyzed to obtain a final analysis result.

This application also provides an intelligent question-and-answer corpus analysis device, which includes:

The corpus enhancement module is used to obtain a historical question and answer corpus, and input the historical question and answer corpus into a pre-built data enhancement model to obtain an enhanced corpus;

The enhanced corpus screening module is used to train a pre-built convolutional neural network model using the historical question and answer corpus to obtain an initial classification prediction model; use the initial classification prediction model to analyze and screen the enhanced corpus to obtain an enhanced corpus Subset;

A model training module for training the convolutional neural network model by using the enhanced corpus subset and the historical question and answer corpus to obtain an analysis and prediction model;

The model analysis module is used to obtain the user corpus text to be analyzed, use the analysis prediction model to analyze the user corpus text to be analyzed, and obtain the final analysis result.

This application also provides an electronic device, which includes:

Memory, storing at least one instruction; and

The processor executes the instructions stored in the memory to implement the following steps:

This application also provides a computer-readable storage medium, including a storage data area and a storage program area. The storage data area stores data created according to the use of blockchain nodes, and the storage program area stores a computer program, which is readable by the computer. At least one instruction is stored in the storage medium, and the at least one instruction is executed by the processor in the electronic device to implement the following steps:

Description of the drawings

FIG. 1 is a schematic flowchart of an intelligent question-and-answer corpus analysis method provided by an embodiment of this application;

2 is a schematic diagram of modules of an intelligent question and answer corpus analysis device provided by an embodiment of the application;

FIG. 3 is a schematic diagram of the internal structure of an electronic device that implements an intelligent question-and-answer corpus analysis method provided by an embodiment of the application;

The realization, functional characteristics, and advantages of the purpose of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings.

Detailed ways

It should be understood that the specific embodiments described here are only used to explain the present application, and are not used to limit the present application.

This application provides an intelligent question-and-answer corpus analysis method. Referring to FIG. 1, it is a schematic flowchart of a method for analyzing intelligent question and answer corpus provided by an embodiment of this application. The method can be executed by a device, and the device can be implemented by software and/or hardware.

In this embodiment, the intelligent question answering corpus analysis method includes:

S1. Obtain a historical question and answer corpus, and input the historical question and answer corpus into a pre-built data enhancement model to obtain an enhanced corpus;

In this embodiment of the application, the question and answer corpus in the historical question and answer corpus is a collection of answer texts for candidates to answer preset recruitment questions. For example, the recruitment question can be "What is your career plan?", "What is your career plan?" What do you think of the company's prospects?". The historical question and answer corpus can be obtained from the company's human resources department database.

Further, since the data of the historical question and answer corpus is small and not easy to obtain, it is necessary to expand the data of the historical question and answer corpus. Expand.

Preferably, the data enhancement model can be constructed using the currently known Seq2Seq algorithm and the variational autoencoder, using the question and answer corpus as the training set, and the marked question and answer corpus as the label set, to complete the data The training of the enhanced model, wherein the question and answer corpus is different from the historical question and answer corpus.

Further, the historical question and answer corpus is input to the data enhancement model, and the enhanced corpus is output, and the data expansion of the historical question and answer corpus is completed.

S2. Use the historical question and answer corpus to train a pre-built convolutional neural network model to obtain an initial classification prediction model;

In the embodiment of the application, the historical question and answer corpus is determined as the first training set, and the preset analysis position mark is performed on the historical question and answer corpus to obtain the first label set. Preferably, the analysis position There are five types: excellent, good, medium, pass and fail.

Preferably, the convolutional neural network model described in the preferred embodiment of the present application can be constructed using a deeply separable convolutional network model.

Further, training the convolutional neural network model by using the first training set and the first label set in the embodiment of the present application includes:

S21: Perform a depth-separable convolution pooling operation on the first training set according to the preset number of depth-separable convolution pooling to obtain a dimensionality reduction data set;

S22: Calculate the dimensionality reduction data set by using a preset activation function to obtain a predicted value, and calculate a loss value by using a pre-built loss function according to the predicted value and the label value contained in the first label set.

S23: Compare the magnitude of the loss value with a preset loss threshold, and if the loss value is greater than or equal to the loss threshold, return to S21; if the loss value is less than the loss threshold, obtain the classification prediction model.

In detail, the depth separable convolution pooling operation includes: performing a grouped convolution operation on the first training set to obtain a deep convolution data set, and then performing a point-wise convolution operation on the deep convolution data set to obtain A point-by-point convolution data set is performed, and an average pooling operation is performed on the point-by-point convolution data set to obtain the dimensionality reduction data set.

In a preferred embodiment of the present application, the activation function can be calculated using the following formula:

f(x)=max(0,x)

Where f(x) is the predicted value, and x is the data in the dimensionality reduction data set.

In a preferred embodiment of the present application, the loss function can be calculated using the following formula:

Wherein, N is the number of samples included in the training data, i is a positive integer, h _i to the tag value, m _i is the predicted value.

S3. Use the initial classification prediction model to analyze and screen the enhanced corpus to obtain an enhanced corpus subset;

In the embodiment of the present application, the data of the enhanced corpus is generated by a model. In order to improve the data authenticity of the enhanced corpus, thereby improving the training accuracy of subsequent models, the enhanced corpus is screened.

In detail, the initial classification prediction model is used to analyze the enhanced corpus to obtain a classification analysis result set of each piece of data in the enhanced corpus. For example: the classification prediction model has five classification predictions: excellent, good, medium, pass and fail. When the classification prediction model predicts the data, the result will output the classification analysis result set, and include the classification of each level. Classification analysis results, such as excellent 0.6, good 0.4, medium 0.45, pass 0.6 and fail 0.5.

Further, in order to ensure the credibility of the prediction score, the data corresponding to the classification analysis result set in which the classification analysis result in the enhanced corpus set is less than the preset threshold is deleted to obtain the enhanced corpus subset. Preferably, the threshold can be set For example, the classification prediction model predicts the data A of the enhanced corpus, and the classification analysis result set is: excellent 0.3, good 0.4, medium 0.45, pass 0.2 and fail 0.2, the classification of each level The analysis results are all less than 0.5, indicating that the authenticity of data A is low, and data A is deleted.

S4. Train the convolutional neural network model by using the enhanced corpus subset and the historical question and answer corpus to obtain an analysis and prediction model;

In this embodiment of the application, in order to make the training data of the model more real and comprehensive, the enhanced corpus subset and the historical question and answer corpus are used as the second training set, and the second training set is subjected to a preset analysis file The position mark obtains the second label set. Preferably, the analysis gear can be five kinds of excellent, good, medium, pass and fail.

Further, the embodiment of the present application uses the second training set and the second label set to train the convolutional neural network model to obtain the analysis and prediction model.

S5. Obtain a user corpus text to be analyzed, and analyze the user corpus text to be analyzed by using the analysis prediction model to obtain a final analysis result;

In this embodiment of the application, the user corpus text to be analyzed is the answer text that the user answers according to preset recruitment questions. For example, the recruitment question can be "What is your career plan?" What is the outlook?".

In another embodiment of the present application, the data used for training the analysis and prediction model may be stored in the blockchain.

Further, using the analysis prediction model to analyze the user corpus text to be analyzed to obtain a classification analysis result set, sort the classification analysis results in the classification analysis result set, and select the classification analysis result with the largest value as the final analysis result . For example: the analysis of each category in the analysis result set is excellent 0.3, good 0.4, medium 0.45, pass 0.2 and fail 0.2. The 0.45 analysis is the largest, so medium 0.45 is used as the final analysis result.

In the embodiment of the present application, the historical question and answer corpus is input to a pre-built data enhancement model to obtain an enhanced corpus, and the historical question and answer corpus is data augmented; the historical question and answer corpus is used to train the pre-built convolution The neural network model obtains the initial classification prediction model, the initial classification prediction model is used to analyze and screen the enhanced corpus, and the data whose classification analysis results are greater than the preset threshold in the enhanced corpus are summarized to obtain the enhanced corpus subset, The enhanced corpus is screened to improve the data authenticity of the enhanced corpus; the enhanced corpus subset and the historical question and answer corpus are used to train the convolutional neural network model to obtain an analysis and prediction model, which improves the model’s performance Training accuracy; obtain the user corpus text to be analyzed, use the analysis prediction model to analyze the user corpus text to be analyzed to obtain the final analysis result; enhance the sample capacity by performing data enhancement on the existing small sample training data, and further The enhanced data is screened, and the accuracy of model prediction is improved, which solves the problem of inaccurate model training caused by fewer training data samples.

As shown in Figure 2, it is a functional module diagram of the intelligent question and answer corpus analysis device of the present application.

The intelligent question answering corpus analysis device 100 described in this application can be installed in an electronic device. According to the realized functions, the intelligent question and answer corpus analysis device may include a corpus enhancement module 101, an enhanced corpus screening module 102, a model training module 103, and a model analysis module 104. The module described in the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can complete fixed functions, and are stored in the memory of the electronic device.

In this embodiment, the functions of each module/unit are as follows:

The corpus enhancement module 101 is used to obtain a historical question and answer corpus, and input the historical question and answer corpus into a pre-built data enhancement model to obtain an enhanced corpus.

Preferably, the data enhancement model can be constructed using the currently known Seq2Seq algorithm and a variational autoencoder. The corpus enhancement module 101 uses the question and answer corpus as the training set and the marked question and answer corpus as the label. And then complete the training of the data enhancement model, wherein the question and answer corpus is different from the historical question and answer corpus.

Further, the corpus enhancement module 101 inputs the historical question and answer corpus into the pre-built data enhancement model, outputs the enhanced corpus, and completes the data expansion of the historical question and answer corpus.

The enhanced corpus screening module 102 is configured to use the historical question and answer corpus to train a pre-built convolutional neural network model to obtain an initial classification prediction model; use the initial classification prediction model to analyze and screen the enhanced corpus to obtain Enhanced corpus subset.

In the embodiment of this application, the enhanced corpus screening module 102 determines the historical question and answer corpus as the first training set, and performs a preset analysis position mark on the historical question and answer corpus to obtain the first label set, preferably In particular, the analysis gears can be classified into five types: excellent, good, medium, pass and fail.

Further, the enhanced corpus screening module 102 of the embodiment of the present application obtains the classification prediction model by the following means:

A: According to the preset depth separable convolution pooling times, perform the depth separable convolution pooling operation on the first training set to obtain a dimensionality reduction data set;

B: Use a preset activation function to calculate the dimensionality reduction data set to obtain a predicted value, and use a pre-built loss function to calculate a loss value based on the predicted value and the label value contained in the first label set.

C: Compare the magnitude of the loss value with the preset loss threshold, if the loss value is greater than or equal to the loss threshold, return to A; if the loss value is less than the loss threshold, obtain the classification prediction model.

In detail, the enhanced corpus screening module 102 obtains the dimensionality reduction data set by using the following method: performing a grouped convolution operation on the first training set to obtain a deep convolution data set, and then performing a calculation on the deep convolution data set Performing a point-by-point convolution operation to obtain a point-by-point convolution data set, and performing an average pooling operation on the point-by-point convolution data set to obtain the dimensionality reduction data set.

f(x)=max(0,x)

In the embodiment of the present application, the data of the enhanced corpus is generated by a model. In order to improve the data authenticity of the enhanced corpus, thereby further improving the training accuracy of subsequent models, the enhanced corpus is screened.

In detail, the enhanced corpus screening module 102 uses the initial classification prediction model to analyze the enhanced corpus set to obtain a classification analysis result set of each piece of data in the enhanced corpus set. For example: the classification prediction model has five classification predictions: excellent, good, medium, pass and fail. When the classification prediction model predicts the data, the result will output the classification analysis result set, and include the classification of each level. Classification analysis results, such as excellent 0.6, good 0.4, medium 0.45, pass 0.6 and fail 0.5.

Further, in order to ensure the credibility of the predicted score, the enhanced corpus screening module 102 deletes the data corresponding to the classification analysis result set in which the classification analysis result in the enhanced corpus set is less than a preset threshold to obtain an enhanced corpus subset, preferably Preferably, the threshold may be set to 0.5. For example, the classification prediction model predicts the data A of the enhanced corpus, and the result set of classification analysis is: excellent 0.3, good 0.4, medium 0.45, pass 0.2 and fail 0.2, the classification analysis results of each gear are all less than 0.5, indicating that the authenticity of data A is low, and data A is deleted.

The model training module 103 is configured to train the convolutional neural network model by using the enhanced corpus subset and the historical question and answer corpus to obtain an analysis and prediction model.

In the embodiment of the present application, in order to make the training data of the model more realistic and comprehensive, the model training module 103 uses the enhanced corpus subset and the historical question and answer corpus as the second training set, and the second training set Performing the preset analysis gear mark to obtain the second label set, preferably, the analysis gear can be five kinds of excellent, good, medium, pass and fail.

Further, the model training module 103 of the embodiment of the present application trains the convolutional neural network model by using the second training set and the second label set to obtain the analysis and prediction model.

The model analysis module 104 obtains the user corpus text to be analyzed, uses the analysis prediction model to analyze the user corpus text to be analyzed, and obtains the final analysis result.

In another embodiment of the present application, the data used for training the analysis and prediction model training can be stored in the blockchain.

Further, the model analysis module 104 uses the analysis prediction model to analyze the user corpus text to be analyzed to obtain a classification analysis result set, sort the classification analysis results in the classification analysis result set, and select the classification with the largest value The analysis result is regarded as the final analysis result. For example: the analysis of each category in the analysis result set is excellent 0.3, good 0.4, medium 0.45, pass 0.2 and fail 0.2. The 0.45 analysis is the largest, so medium 0.45 is used as the final analysis result.

As shown in FIG. 3, it is a schematic diagram of the structure of an electronic device implementing the intelligent question answering corpus analysis method in this application.

The electronic device 1 may include a processor 10, a memory 11, and a bus, and may also include a computer program stored in the memory 11 and running on the processor 10, such as an intelligent question-and-answer corpus analysis program.

The memory 11 includes at least one type of readable storage medium. The readable storage medium may be non-volatile or volatile. The readable storage medium includes flash memory, mobile hard disk, and multimedia card. , Card-type memory (for example: SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. The memory 11 may be an internal storage unit of the electronic device 1 in some embodiments, for example, a mobile hard disk of the electronic device 1. In other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart media card (SMC), and a secure digital (Secure Digital) equipped on the electronic device 1. , SD) card, flash card (Flash Card), etc. Further, the memory 11 may also include both an internal storage unit of the electronic device 1 and an external storage device. The memory 11 can be used not only to store application software and various data installed in the electronic device 1, such as codes of a smart question and answer corpus analysis program, etc., but also to temporarily store data that has been output or will be output.

The processor 10 may be composed of integrated circuits in some embodiments, for example, may be composed of a single packaged integrated circuit, or may be composed of multiple integrated circuits with the same function or different functions, including one or more Combinations of central processing unit (CPU), microprocessor, digital processing chip, graphics processor, and various control chips, etc. The processor 10 is the control unit of the electronic device, which uses various interfaces and lines to connect the various components of the entire electronic device, and runs or executes programs or modules (such as smart devices) stored in the memory 11 Question and answer corpus analysis program, etc.), and call the data stored in the memory 11 to execute various functions of the electronic device 1 and process data.

The bus may be a peripheral component interconnect standard (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into address bus, data bus, control bus and so on. The bus is configured to implement connection and communication between the memory 11 and at least one processor 10 and the like.

FIG. 3 only shows an electronic device with components. Those skilled in the art can understand that the structure shown in FIG. 3 does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure. Components, or combinations of certain components, or different component arrangements.

For example, although not shown, the electronic device 1 may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, thereby controlling power The device implements functions such as charge management, discharge management, and power consumption management. The power supply may also include any components such as one or more DC or AC power supplies, recharging devices, power failure detection circuits, power converters or inverters, and power status indicators. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

Further, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and/or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is usually used in the electronic device 1 Establish a communication connection with other electronic devices.

Optionally, the electronic device 1 may also include a user interface. The user interface may be a display (Display) and an input unit (such as a keyboard (Keyboard)). Optionally, the user interface may also be a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode, organic light-emitting diode) touch device, etc. Among them, the display can also be appropriately called a display screen or a display unit, which is used to display the information processed in the electronic device 1 and to display a visualized user interface.

It should be understood that the embodiments are only for illustrative purposes, and are not limited by this structure in the scope of the patent application.

The intelligent question and answer corpus analysis program 12 stored in the memory 11 in the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can realize:

Specifically, for the specific implementation method of the above-mentioned instructions by the processor 10, reference may be made to the description of the relevant steps in the embodiment corresponding to FIG. 1, which will not be repeated here.

Further, if the integrated module/unit of the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory) .

Further, the computer usable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function, etc.; the storage data area may store a block chain node Use the created data, etc.

In the several embodiments provided in this application, it should be understood that the disclosed equipment, device, and method may be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the objectives of the solutions of the embodiments.

In addition, the functional modules in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware, or may be implemented in the form of hardware plus software functional modules.

For those skilled in the art, it is obvious that the present application is not limited to the details of the foregoing exemplary embodiments, and the present application can be implemented in other specific forms without departing from the spirit or basic characteristics of the present application.

Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of this application is defined by the appended claims rather than the above description, and therefore it is intended to fall into the claims. All changes in the meaning and scope of the equivalent elements of are included in this application. Any reference signs in the claims should not be regarded as limiting the claims involved.

The blockchain referred to in this application is a new application mode of computer technology such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm. Blockchain, essentially a decentralized database, is a series of data blocks associated with cryptographic methods. Each data block contains a batch of network transaction information for verification. The validity of the information (anti-counterfeiting) and the generation of the next block. The blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.

In addition, it is obvious that the word "including" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in the system claims can also be implemented by one unit or device through software or hardware. The second class words are used to indicate names, and do not indicate any specific order.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the application and not to limit them. Although the application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the application can be Make modifications or equivalent replacements without departing from the spirit and scope of the technical solution of the present application.

Claims

An intelligent question answering corpus analysis method, wherein the method includes:

Obtain a historical question and answer corpus, and input the historical question and answer corpus into a pre-built data enhancement model to obtain an enhanced corpus;

Use the historical question and answer corpus to train a pre-built convolutional neural network model to obtain an initial classification prediction model;

Using the initial classification prediction model to analyze and screen the enhanced corpus to obtain an enhanced corpus subset;

Training the convolutional neural network model by using the enhanced corpus subset and the historical question and answer corpus to obtain an analysis and prediction model;

The user corpus text to be analyzed is obtained, and the analysis prediction model is used to analyze the user corpus text to be analyzed to obtain a final analysis result.
The intelligent question answering corpus analysis method of claim 1, wherein the training of the pre-built convolutional neural network model using the historical question answering corpus to obtain the initial classification prediction model comprises:

Determining the historical question and answer corpus as the first training set;

Mark the preset analysis positions on the historical question and answer corpus to obtain a first label set;

The first training set and the first label set are used to train the convolutional neural network model to obtain the initial classification prediction model.
3. The intelligent question answering corpus analysis method of claim 2, wherein said training said convolutional neural network model using said first training set and said first label set comprises:

A: According to the preset depth separable convolution pooling times, perform the depth separable convolution pooling operation on the first training set to obtain a dimensionality reduction data set;

B: Use a preset activation function to calculate the dimensionality reduction data set to obtain a predicted value, and use a pre-built loss function to calculate according to the predicted value and the label value contained in the first label set to obtain a loss value ；

C: Compare the size of the loss value with the preset loss threshold, if the loss value is greater than or equal to the loss threshold, return to A; if the loss value is less than the loss threshold, obtain the initial classification prediction model .
5. The intelligent question answering corpus analysis method of claim 1, wherein the use of the initial classification prediction model to analyze and screen the enhanced corpus to obtain an enhanced corpus subset comprises:

Using the initial classification prediction model to analyze the enhanced corpus to obtain a classification analysis result set of each piece of data in the enhanced corpus;

The data corresponding to the classification analysis result set whose classification analysis results are all less than the preset threshold is deleted from the enhanced corpus set to obtain the enhanced corpus subset.
5. The intelligent question and answer corpus analysis method of claim 1, wherein the obtaining the user corpus text to be analyzed, and using the analysis prediction model to analyze the user corpus text to be analyzed to obtain the final analysis result comprises:

Analyze the user corpus text to be analyzed by using the analysis and prediction model to obtain a classification analysis result set;

Sorting the classification analysis results in the classification analysis result set;

The classification analysis result with the largest numerical value is selected as the final analysis result.
The intelligent question answering corpus analysis method according to claim 3, wherein said performing a deep separable convolution pooling operation on said first training set to obtain a dimensionality reduction data set comprises:

Performing a grouped convolution operation on the first training set to obtain a deep convolution data set;

Performing a point-by-point convolution operation on the deep convolution data set to obtain a point-by-point convolution data set;

Perform an average pooling operation on the point-by-point convolution data set to obtain the dimensionality reduction data set.
The intelligent question answering corpus analysis method according to claim 2, wherein the analysis grades are excellent, good, medium, pass and fail.
An intelligent question and answer corpus analysis device, wherein the device includes:

The corpus enhancement module is used to obtain a historical question and answer corpus, and input the historical question and answer corpus into a pre-built data enhancement model to obtain an enhanced corpus;

The enhanced corpus screening module is used to train a pre-built convolutional neural network model using the historical question and answer corpus to obtain an initial classification prediction model; use the initial classification prediction model to analyze and screen the enhanced corpus to obtain an enhanced corpus Subset;

A model training module for training the convolutional neural network model by using the enhanced corpus subset and the historical question and answer corpus to obtain an analysis and prediction model;

The model analysis module is used to obtain the user corpus text to be analyzed, use the analysis prediction model to analyze the user corpus text to be analyzed, and obtain the final analysis result.
An electronic device, wherein the electronic device includes:

At least one processor; and,

A memory communicatively connected with the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can execute the following steps:

Obtain a historical question and answer corpus, and input the historical question and answer corpus into a pre-built data enhancement model to obtain an enhanced corpus;

Use the historical question and answer corpus to train a pre-built convolutional neural network model to obtain an initial classification prediction model;

Using the initial classification prediction model to analyze and screen the enhanced corpus to obtain an enhanced corpus subset;

Training the convolutional neural network model by using the enhanced corpus subset and the historical question and answer corpus to obtain an analysis and prediction model;

The user corpus text to be analyzed is obtained, and the analysis prediction model is used to analyze the user corpus text to be analyzed to obtain a final analysis result.
9. The electronic device according to claim 9, wherein the training a pre-built convolutional neural network model using the historical question and answer corpus to obtain an initial classification prediction model comprises:

Determining the historical question and answer corpus as the first training set;

Mark the preset analysis positions on the historical question and answer corpus to obtain a first label set;

The first training set and the first label set are used to train the convolutional neural network model to obtain the initial classification prediction model.
11. The electronic device according to claim 10, wherein said training said convolutional neural network model using said first training set and said first label set comprises:

A: According to the preset depth separable convolution pooling times, perform the depth separable convolution pooling operation on the first training set to obtain a dimensionality reduction data set;

B: Use a preset activation function to calculate the dimensionality reduction data set to obtain a predicted value, and use a pre-built loss function to calculate according to the predicted value and the label value contained in the first label set to obtain a loss value ；

C: Compare the size of the loss value with the preset loss threshold, if the loss value is greater than or equal to the loss threshold, return to A; if the loss value is less than the loss threshold, obtain the initial classification prediction model .
9. The electronic device according to claim 9, wherein said using said initial classification prediction model to analyze and screen said enhanced corpus to obtain an enhanced corpus subset, comprising:

Using the initial classification prediction model to analyze the enhanced corpus to obtain a classification analysis result set of each piece of data in the enhanced corpus;

The data corresponding to the classification analysis result set whose classification analysis results are all less than the preset threshold is deleted from the enhanced corpus set to obtain the enhanced corpus subset.
9. The electronic device according to claim 9, wherein said acquiring the user corpus text to be analyzed, using the analysis prediction model to analyze the user corpus text to be analyzed to obtain the final analysis result, comprises:

Analyze the user corpus text to be analyzed by using the analysis and prediction model to obtain a classification analysis result set;

Sorting the classification analysis results in the classification analysis result set;

The classification analysis result with the largest numerical value is selected as the final analysis result.
The electronic device according to claim 11, wherein said performing a depth-separable convolution pooling operation on said first training set to obtain a dimensionality reduction data set comprises:

Performing a grouped convolution operation on the first training set to obtain a deep convolution data set;

Performing a point-by-point convolution operation on the deep convolution data set to obtain a point-by-point convolution data set;

Perform an average pooling operation on the point-by-point convolution data set to obtain the dimensionality reduction data set.
A computer-readable storage medium includes a storage data area and a storage program area. The storage data area stores data created according to the use of a blockchain node. The storage program area stores a computer program, wherein the computer program is stored by a processor. The following steps are implemented during execution:

Obtain a historical question and answer corpus, and input the historical question and answer corpus into a pre-built data enhancement model to obtain an enhanced corpus;

Use the historical question and answer corpus to train a pre-built convolutional neural network model to obtain an initial classification prediction model;

Using the initial classification prediction model to analyze and screen the enhanced corpus to obtain an enhanced corpus subset;

Training the convolutional neural network model by using the enhanced corpus subset and the historical question and answer corpus to obtain an analysis and prediction model;

The user corpus text to be analyzed is obtained, and the analysis prediction model is used to analyze the user corpus text to be analyzed to obtain a final analysis result.
15. The computer-readable storage medium of claim 15, wherein the training a pre-built convolutional neural network model using the historical question and answer corpus to obtain an initial classification prediction model comprises:

Determining the historical question and answer corpus as the first training set;

Mark the preset analysis positions on the historical question and answer corpus to obtain a first label set;

The first training set and the first label set are used to train the convolutional neural network model to obtain the initial classification prediction model.
15. The computer-readable storage medium of claim 16, wherein said training said convolutional neural network model using said first training set and said first label set comprises:

A: According to the preset depth separable convolution pooling times, perform the depth separable convolution pooling operation on the first training set to obtain a dimensionality reduction data set;

B: Use a preset activation function to calculate the dimensionality reduction data set to obtain a predicted value, and use a pre-built loss function to calculate according to the predicted value and the label value contained in the first label set to obtain a loss value ；

C: Compare the size of the loss value with the preset loss threshold, if the loss value is greater than or equal to the loss threshold, return to A; if the loss value is less than the loss threshold, obtain the initial classification prediction model .
15. The computer-readable storage medium according to claim 15, wherein said using said initial classification prediction model to analyze and filter said enhanced corpus to obtain an enhanced subset of corpus comprises:

Using the initial classification prediction model to analyze the enhanced corpus to obtain a classification analysis result set of each piece of data in the enhanced corpus;

The data corresponding to the classification analysis result set whose classification analysis results are all less than the preset threshold is deleted from the enhanced corpus set to obtain the enhanced corpus subset.
15. The computer-readable storage medium according to claim 15, wherein the obtaining the user corpus text to be analyzed, and using the analysis prediction model to analyze the user corpus text to be analyzed to obtain the final analysis result, comprises:

Analyze the user corpus text to be analyzed by using the analysis and prediction model to obtain a classification analysis result set;

Sorting the classification analysis results in the classification analysis result set;

The classification analysis result with the largest numerical value is selected as the final analysis result.
17. The computer-readable storage medium according to claim 17, wherein said performing a deeply separable convolutional pooling operation on said first training set to obtain a dimensionality reduction data set comprises:

Performing a grouped convolution operation on the first training set to obtain a deep convolution data set;

Performing a point-by-point convolution operation on the deep convolution data set to obtain a point-by-point convolution data set;

Perform an average pooling operation on the point-by-point convolution data set to obtain the dimensionality reduction data set.