WO2022041866A1

WO2022041866A1 - Method, apparatus and device for determining causal relationship, and readable storage medium

Info

Publication number: WO2022041866A1
Application number: PCT/CN2021/094934
Authority: WO
Inventors: 张天豫; 范力欣; 吴锦和
Original assignee: 深圳前海微众银行股份有限公司
Priority date: 2020-08-27
Filing date: 2021-05-20
Publication date: 2022-03-03
Also published as: CN112000856A

Abstract

A method, apparatus and device for determining a causal relationship, and a readable storage medium. The method comprises: acquiring first data and second data, and converting the second data on the basis of a preset neural network and the first data, so as to obtain third data matching the first data in terms of the divergence of the preset neural network (S10); converting the first data on the basis of the preset neural network and the second data, so as to obtain fourth data matching the second data in terms of the divergence of the preset neural network (S20); and acquiring a first conversion time for conversion between the second data and the third data and a second conversion time for conversion between the first data and the fourth data, and determining a causal relationship between the first data and the second data according to a first conversion difficulty level coefficient corresponding to the first conversion time and a second conversion difficulty level coefficient corresponding to the second conversion time (S30).

Description

Method, apparatus, device and readable storage medium for determining causal relationship

This application claims the priority of the Chinese patent application filed on August 27, 2020, the application number is 202010891985.7, and the title is "Method, Apparatus, Equipment and Readable Storage Medium for Determining Causal Relationship", which is hereby incorporated in its entirety as refer to.

technical field

The present application relates to the technical field of financial technology (Fintech), and in particular, to a method, apparatus, device and readable storage medium for determining a causal relationship.

Background technique

With the continuous development of financial technology (Fintech), especially Internet technology finance, more and more technologies (such as artificial intelligence, big data analysis, cloud storage, etc.) are applied in the financial field, but the financial field also proposes various technologies. higher requirements, such as the requirement to improve the accuracy of data analysis, etc.

At present, machine learning algorithms are widely used in various fields as an effective data analysis tool, but the ability of machine learning algorithms to infer causal relationships between data is slightly lacking, which makes the accuracy of data analysis not high, and limits machine learning. The popularization and use of the algorithm in the fields that require high data analysis accuracy.

Therefore, how to infer the causal relationship between data and improve the accuracy of data analysis is a technical problem that needs to be solved urgently.

technical problem

The main purpose of this application is to provide a method, device, device and readable storage medium for determining a causal relationship, which aims to solve the technical problem of how to infer the causal relationship between data and improve the accuracy of data analysis in the prior art.

technical solutions

In order to achieve the above object, the present application provides a method for determining a causal relationship, and the method for determining a causal relationship includes the following steps:

Obtain the first data and the second data, and convert the second data based on the preset neural network and the first data to obtain the first data matching the divergence of the preset neural network with the first data. three data;

Based on the preset neural network and the second data, converting the first data to obtain fourth data matching the divergence of the second data in the preset neural network;

Acquiring a first conversion time for conversion between the second data and the third data, and a second conversion time for conversion between the first data and the fourth data, and according to the first conversion time The corresponding first conversion difficulty degree coefficient and the second conversion difficulty degree coefficient corresponding to the second conversion time determine the causal relationship between the first data and the second data.

Optionally, according to the first conversion difficulty degree coefficient corresponding to the first conversion time and the second conversion difficulty degree coefficient corresponding to the second conversion time, determine the relationship between the first data and the first data Two steps of causality between data include:

Comparing the first conversion difficulty degree coefficient with the second conversion difficulty degree coefficient, and determining the magnitude relationship between the first conversion difficulty degree coefficient and the second conversion difficulty degree coefficient;

If the magnitude relationship is that the first conversion difficulty degree coefficient is greater than the second conversion difficulty degree coefficient, determine the reason why the first data constitutes the causal relationship, and the second data constitutes the result in the causal relationship ;

If the magnitude relationship is that the first conversion difficulty degree coefficient is smaller than the second conversion difficulty degree coefficient, determine the reason why the second data constitutes the causal relationship, and the first data constitutes the result in the causal relationship .

Optionally, the step of converting the second data based on the preset neural network and the first data to obtain third data matching the divergence of the preset neural network with the first data include:

Using the first data as a guide, convert the second data based on the generator in the preset neural network to obtain intermediate data;

Based on the discriminator in the preset neural network, the intermediate data is discriminated, and the divergence value between the first data and the intermediate data is determined;

According to the divergence value, third data matching the divergence of the preset neural network with the first data is determined.

Optionally, the step of discriminating the intermediate data based on a discriminator in a preset neural network, and determining a divergence value between the first data and the intermediate data includes:

based on the discriminator in the preset neural network, discriminate the intermediate data, and generate a first probability distribution of the intermediate data;

A second probability distribution of the first data is acquired, and a divergence value between the first data and the intermediate data is generated according to the first probability distribution and the second probability distribution.

Optionally, the step of determining, according to the divergence value, third data that matches the first data on the divergence of the preset neural network includes:

Determine whether the divergence value is less than a preset threshold, and if it is less than a preset threshold, determine the intermediate data as the third data;

If the divergence value is not less than the preset threshold, perform the step of converting the second data based on the generator in the preset neural network according to the divergence value, until the divergence The value is less than the preset threshold.

Optionally, after the step of determining the intermediate data as the third data, the method includes:

Find the first time point of the first conversion of the second data, and determine the second time point of the third data;

According to the first time point and the second time point, a first conversion time corresponding to the third data is generated.

Optionally, before the step of acquiring the first data and the second data, the method further includes:

acquiring a first data set and a second data set with a preset causal relationship, and processing the first data set and the second data set based on a preset neural network to generate a processing result;

It is judged whether the causal relationship in the processing result is consistent with the preset causal relationship, and if so, the first data and the second data are processed based on the preset neural network.

Further, in order to achieve the above object, the present application also provides a device for determining a causal relationship, the device for determining a causal relationship includes:

The acquisition module is used to acquire the first data and the second data, and based on the preset neural network and the first data, convert the second data to obtain the dispersion of the first data in the preset neural network. degree of matching third data;

a conversion module, configured to convert the first data based on the preset neural network and the second data to obtain fourth data matching the second data in the divergence of the preset neural network;

A determination module, configured to obtain a first conversion time for conversion between the second data and the third data, and a second conversion time for conversion between the first data and the fourth data, and according to the A first conversion difficulty degree coefficient corresponding to the first conversion time and a second conversion difficulty degree coefficient corresponding to the second conversion time are used to determine a causal relationship between the first data and the second data.

Further, in order to achieve the above object, the present application also provides a causal relationship determination device, the causal relationship determination device includes a memory, a processor, and a causal relationship stored on the memory and can be run on the processor. A program for determining a relationship, when the program for determining a causal relationship is executed by the processor implements the steps of the method for determining a causal relationship as described above.

Further, in order to achieve the above purpose, the present application also provides a readable storage medium, on which a program for determining a causal relationship is stored, and the program for determining a causal relationship is executed by a processor to achieve the above-mentioned The steps of the method for determining the causal relationship.

beneficial effect

Compared with the method, device, device and readable storage medium for determining the causal relationship of the present application, which is difficult to infer the causal relationship between data in the prior art, the method adopted in the present application is: acquiring the first data and the second data, and using the first data One data is used as a reference, and the second data is converted through a preset neural network to obtain third data matching the divergence of the first data in the preset neural network; Convert the first data to obtain fourth data matching the second data on the divergence of the preset neural network; and then obtain the first conversion time of the conversion between the second data and the third data, and the first data and the The second conversion time for conversion between the fourth data is determined according to the first conversion difficulty degree coefficient corresponding to the first conversion time and the second conversion difficulty degree coefficient corresponding to the second conversion time. The causal relationship between the second data. The first conversion time is the length of time for forming the third data, the corresponding first conversion difficulty coefficient represents the difficulty of converting the second data into third data matching the first data, and the second conversion time is The duration of forming the fourth data, and the corresponding second conversion difficulty coefficient represents the difficulty of converting the first data into fourth data matching the second data. The shorter the time period and the smaller the corresponding coefficient, the easier the conversion is, and the cause in the causal relationship is formed; the longer the time period and the larger the corresponding coefficient, the more difficult the conversion is, and the effect in the causal relationship is formed. The magnitude relationship between the first conversion difficulty degree coefficient and the second conversion difficulty degree coefficient corresponding to the conversion time and the second conversion time respectively is used to accurately determine the causal relationship between the first data and the second data. It overcomes the defect that it is difficult to infer the causal relationship between data in the prior art, and is beneficial to the improvement of the accuracy of data analysis.

Description of drawings

1 is a schematic structural diagram of a device hardware operating environment involved in a device embodiment solution for determining a causal relationship of the present application;

2 is a schematic flowchart of a first embodiment of a method for determining a causal relationship of the present application;

3 is a schematic flowchart of a third embodiment of a method for determining a causal relationship of the present application;

FIG. 4 is a schematic diagram of functional modules of a preferred embodiment of an apparatus for determining a causal relationship of the present application.

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

Embodiments of the present invention

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

The present application provides a device for determining a causal relationship. Referring to FIG. 1 , FIG. 1 is a schematic structural diagram of a device hardware operating environment involved in an embodiment solution of the device for determining a causal relationship in the present application.

As shown in FIG. 1 , the device for determining the causal relationship may include: a processor 1001 , such as a CPU, a communication bus 1002 , a user interface 1003 , a network interface 1004 , and a memory 1005 . Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. Optionally, the network interface 1004 may include a standard wired interface and a wireless interface (eg, a WI-FI interface). The memory 1005 may be high-speed RAM memory, or may be non-volatile memory, such as disk memory. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001 .

Those skilled in the art can understand that the hardware structure of the device for determining the causal relationship shown in FIG. 1 does not constitute a limitation on the device for determining the causal relationship, and may include more or less components than those shown in the figure, or combine some components, or a different arrangement of components.

As shown in FIG. 1 , the memory 1005 as a readable storage medium may include an operating system, a network communication module, a user interface module and a program for determining a causal relationship. Among them, the operating system is a program that manages and controls causal relationship determination equipment and software resources, and supports the operation of network communication modules, user interface modules, causal relationship determination programs, and other programs or software; network communication modules are used to manage and control the network. Interface 1004 ; the user interface module is used to manage and control the user interface 1003 .

In the hardware structure of the causal relationship determination device shown in FIG. 1 , the network interface 1004 is mainly used to connect to the background server and perform data communication with the background server; the user interface 1003 is mainly used to connect the client (client) and perform data communication with the client Data communication; the processor 1001 can call the causal relationship determination program stored in the memory 1005, and perform the following operations:

Further, according to the first conversion difficulty degree coefficient corresponding to the first conversion time and the second conversion difficulty degree coefficient corresponding to the second conversion time, determine the first data and the second conversion difficulty degree coefficient. The steps for causality between data include:

Further, the step of converting the second data based on the preset neural network and the first data to obtain third data matching the divergence of the preset neural network with the first data includes: :

Further, the step of discriminating the intermediate data based on the discriminator in the preset neural network, and determining the divergence value between the first data and the intermediate data includes:

Further, the step of determining, according to the divergence value, third data matching the divergence of the preset neural network with the first data includes:

Further, after the step of determining the intermediate data as the third data, the processor 1001 may call the causal relationship determination program stored in the memory 1005, and perform the following operations:

Further, before the step of acquiring the first data and the second data, the processor 1001 may call the causal relationship determination program stored in the memory 1005, and perform the following operations:

The specific implementation of the device for determining the causal relationship of the present application is basically the same as the following embodiments of the method for determining the causal relationship, and details are not repeated here.

The present application also provides a method for determining a causal relationship.

Referring to FIG. 2 , FIG. 2 is a schematic flowchart of a first embodiment of a method for determining a causal relationship of the present application.

This embodiment of the present application provides an embodiment of a method for determining a causal relationship. It should be noted that although a logical sequence is shown in the flowchart, in some cases, the sequence shown here may be performed in a different order. or the described steps. Specifically, the method for determining the causal relationship in this embodiment includes:

Step S10, acquiring the first data and the second data, and converting the second data based on the preset neural network and the first data to obtain the divergence of the preset neural network with the first data matching third data;

The method for determining a causal relationship in this embodiment is applied to a determining device, and the determining device may be a server or a client. As for the server, it communicates and connects with a plurality of clients that have a causal relationship determination requirement, and this embodiment takes clients as an example for description. Wherein, the client is deployed with a preset neural network, and the preset neural network is preferably an adversarial neural network (GAN, Generative Neural Network). Adversarial Network), the adversarial neural network includes a generator and a discriminator. The generator generates simulated information to deceive the discriminator, and the discriminator is used to distinguish the simulated information from the real information; in the end, the generator generates enough information to be "real", and the discriminator is difficult to judge the authenticity of the information generated by the generator. . This implementation analyzes the causal relationship between the data by means of a preset neural network.

Specifically, the data with the causal relationship explanation requirement is used as the first data and the second data, and the second data is converted through a preset neural network and the first data. Taking the first data as a reference, the generator in the preset neural network converts the second data to generate simulation data similar to the first data. In addition, the generated simulation data is discriminated by the discriminator in the preset neural network, and the third data matching the divergence of the preset neural network with the first data is generated according to the difficulty of the discrimination. The divergence is used to compare the closeness of two probability distributions. In this embodiment, the divergence of the preset neural network is used to represent the similarity between the simulated data processed by the preset neural network and the real data. The third data matched in divergence indicates that the similarity between the third data generated by the generator and the first data is high, and it is difficult to be discriminated by the discriminator.

Step S20, converting the first data based on the preset neural network and the second data to obtain fourth data matching the second data on the divergence of the preset neural network;

Further, for the first data, the preset neural network and the second data are used for conversion. Using the second data as a reference, the generator in the preset neural network converts the first data to generate simulation data similar to the second data. Similarly, the generated simulation data is discriminated by the discriminator in the preset neural network to generate fourth data that matches the second data in the divergence of the preset neural network, reflecting the high similarity with the second data sex.

Step S30: Acquire a first conversion time for conversion between the second data and the third data, and a second conversion time for conversion between the first data and the fourth data, and perform the conversion according to the first conversion time. A first conversion difficulty degree coefficient corresponding to a conversion time and a second conversion difficulty degree coefficient corresponding to the second conversion time determine a causal relationship between the first data and the second data.

Furthermore, in the process of converting the second data, the conversion duration is recorded to obtain the time duration spent generating the third data as the first conversion time between the second data and the third data. At the same time, for the conversion of the first data, the duration is also recorded to obtain the duration of generating the fourth data as the second conversion time between the first data and the fourth data. The difference in duration reflects the difficulty of the conversion. The shorter the conversion time, the easier the conversion, and vice versa. The transformation process reflects the derivation process of the logical relationship between the cause and the result. The easier the transformation, the easier the derivation, reflecting the derivation from the cause to the result. The more difficult the conversion, the more difficult the derivation, reflecting the derivation from the result to the cause. Therefore, the causal relationship between the data can be reflected by the degree of difficulty represented by the time spent in the conversion. That is, the causal relationship between the first data and the second data is determined by the magnitude relationship between the first conversion time and the second conversion time. Specifically, according to the first conversion difficulty degree coefficient corresponding to the first conversion time, and the second conversion difficulty degree coefficient corresponding to the second conversion time, the step of determining the causal relationship between the first data and the second data includes:

Step S31, comparing the first conversion difficulty coefficient with the second conversion difficulty coefficient to determine the magnitude relationship between the first conversion difficulty coefficient and the second conversion difficulty coefficient ;

Step S32, if the magnitude relationship is that the first conversion difficulty degree coefficient is greater than the second conversion difficulty degree coefficient, determine the reason why the first data constitutes the causal relationship, and the second data constitutes the causal relationship result in;

Step S33, if the magnitude relationship is that the first conversion difficulty degree coefficient is smaller than the second conversion difficulty degree coefficient, determine the reason why the second data constitutes the causal relationship, and the first data constitutes the causal relationship results in .

Further, the first conversion time and the second conversion time are compared to determine the magnitude relationship between the two. If it is determined by comparison that the magnitude relationship is that the first conversion time is greater than the second conversion time, it means that the conversion time of the second data to the first data is long, and the conversion time of the first data to the second data is short, so the first data can be determined. The second data constitutes the cause in the causal relationship, and the second data constitutes the effect in the used relationship. Conversely, if it is determined by comparison that the size relationship is that the first conversion time is less than the second conversion time, it means that the conversion time of the second data to the first data is short, and the conversion time of the first data to the second data is long, so it can be determined that the second data is converted to the second data. The data constitute the cause in the causal relationship, and the first data constitute the effect in the used relationship.

It should be noted that, for the case where the magnitude relationship between the first conversion time and the second conversion time is the same, it is difficult to distinguish the causal relationship between the first data and the second data, and there is no output between the first data and the second data. Hint information about causality. In addition, in order to prevent the conversion time from being too long, a preset time threshold is preset to indicate that the conversion time exceeds the normal conversion time. When both the first conversion time and the second conversion time are greater than the preset time threshold, it means that the first data is difficult to Converting to the second data, and the second data is also difficult to convert to the first data, there is no logical relationship between the cause and the result between the first data and the second data, so the prompt information that there is no causal relationship is output. When any one of the first conversion time and the second conversion time is greater than the preset time threshold, the causal relationship between the first data and the second data is determined according to the conversion time not greater than the preset time threshold. If the first transition time is greater than the preset time threshold and the second transition time is not greater than the preset interval threshold, the causal relationship between the first data and the second data is determined according to the second transition time. Since the second conversion time is the time when the first data is converted to the second data, and the second conversion time is greater than the preset time threshold, it means that the conversion time of the first data to the second data is short, so the first data constitutes a causal relationship. cause, while the second data constitutes the effect in a causal relationship. In this way, according to the magnitude relationship between the first conversion time and the second conversion time, the causal relationship between the first data and the second data is determined.

Compared with the method for determining the causal relationship of the present application, which is difficult to infer the causal relationship between the data in the prior art, the method adopted in the present application is: obtaining the first data and the second data, and using the first data as a reference, through the preset neural network The network converts the second data to obtain third data that matches the first data on the divergence of the preset neural network; at the same time, taking the second data as a reference, converts the first data through the preset neural network to obtain the same The second data matches the fourth data on the divergence of the preset neural network; and then obtains the first conversion time of the conversion between the second data and the third data, and the second conversion time of the conversion between the first data and the fourth data. The conversion time is determined, and the causal relationship between the first data and the second data is determined according to the first conversion difficulty degree coefficient corresponding to the first conversion time and the second conversion difficulty degree coefficient corresponding to the second conversion time. The first conversion time is the length of time for forming the third data, the corresponding first conversion difficulty coefficient represents the difficulty of converting the second data into third data matching the first data, and the second conversion time is The duration of forming the fourth data, and the corresponding second conversion difficulty coefficient represents the difficulty of converting the first data into fourth data matching the second data. The shorter the time period and the smaller the corresponding coefficient, the easier the conversion is, and the cause in the causal relationship is formed; the longer the time period and the larger the corresponding coefficient, the more difficult the conversion is, and the effect in the causal relationship is formed. The magnitude relationship between the first conversion difficulty degree coefficient and the second conversion difficulty degree coefficient corresponding to the conversion time and the second conversion time respectively is used to accurately determine the causal relationship between the first data and the second data. It overcomes the defect that it is difficult to infer the causal relationship between data in the prior art, and is beneficial to the improvement of the accuracy of data analysis.

Further, based on the first embodiment of the method for determining a causal relationship of the present application, a second embodiment of the method for determining a causal relationship of the present application is proposed.

The difference between the second embodiment of the method for determining a causal relationship and the first embodiment of the method for determining a causal relationship is that the second data is converted based on the preset neural network and the first data to obtain the same The step of matching the third data with the first data on the divergence of the preset neural network includes:

Step S14, using the first data as a guide, converting the second data based on the generator in the preset neural network to obtain intermediate data;

Step S15, based on a discriminator in a preset neural network, discriminate the intermediate data, and determine a divergence value between the first data and the intermediate data;

In this embodiment, the second data is converted by the preset neural network and the first data, so as to obtain third data matching the divergence of the first data in the preset neural network. Specifically, using the first data as a guide, the generator in the preset neural network converts the second data to obtain the converted intermediate data. Thereafter, the converted intermediate data is discriminated by a discriminator in a preset neural network, the intermediate data is distinguished from the first data, and a divergence value between the first data and the intermediate data is generated. Specifically, based on the discriminator in the preset neural network, the intermediate data is discriminated, and the step of determining the divergence value between the first data and the intermediate data includes:

Step S151, based on a discriminator in a preset neural network, discriminate the intermediate data, and generate a first probability distribution of the intermediate data;

Step S152: Acquire a second probability distribution of the first data, and generate a divergence value between the first data and the intermediate data according to the first probability distribution and the second probability distribution.

Further, the discriminator in the preset neural network is invoked to discriminate the intermediate data, generate a first probability distribution of the intermediate data, and describe the information contained in the intermediate data through the first probability distribution. At the same time, a second probability distribution is generated for the first data, and the information contained in the first data is described by the second probability distribution. Then, the first probability distribution and the second probability distribution are calculated to generate a divergence value between them; wherein, the calculation may be to calculate the expected value of the logarithmic difference between the two, and the obtained expected value result is the divergence value.

Step S16, according to the divergence value, determine third data matching the divergence of the preset neural network with the first data.

Furthermore, since the divergence value is used to represent the similarity between the data converted by the preset neural network and the real data, it is possible to determine the divergence between the first data and the first data by the similarity represented by the divergence value. The matching third data, that is, the third data with high similarity with the first data. Specifically, according to the divergence value, the step of determining the third data matching the divergence of the preset neural network with the first data includes:

Step S161, judging whether the divergence value is less than a preset threshold, and if it is less than a preset threshold, then determining the intermediate data as the third data;

Step S162, if the divergence value is not less than the preset threshold, then according to the divergence value, perform the step of converting the second data based on the generator in the preset neural network, until the predetermined threshold is reached. The divergence value is less than the preset threshold.

Further, a preset threshold representing the level of similarity is preset, and the generated divergence value is compared with the preset threshold to determine whether the divergence value is smaller than the preset threshold. If it is determined by comparison that the divergence value is smaller than the preset threshold, it means that the probability distribution of the first data and the probability distribution of the intermediate data are close to each other, so that the intermediate data is determined as the first data having a high similarity with the first data. Three data. On the contrary, if it is determined by comparison that the divergence value is not less than the preset threshold, it means that the intermediate data obtained by the preset neural network processing is quite different from the first data, so it is necessary to continue to use the first data through the preset neural network. The data is used as a reference, and the second data is iteratively transformed.

Further, in the iterative conversion process of the second data, the divergence value is used as a reference, the preset neural network converts the second data in the direction of reducing the difference according to the first data, and generates intermediate data again to determine the divergence. value to determine whether it is less than the preset threshold. If it is less than the preset threshold, the intermediate data is determined as the third data. If it is still not less than the preset threshold, continue to convert to obtain new intermediate data; this cycle until the generated divergence value is less than the preset threshold, and the obtained intermediate data is determined to be on the divergence of the preset neural network with the first data. matching third data.

Further, after the step of determining the intermediate data as the third data, the method includes:

Step a1, find the first time point of the first conversion to the second data, and determine the second time point of the third data;

Step a2, generating a first conversion time corresponding to the third data according to the first time point and the second time point.

In this embodiment, when the second data is converted for the first time, the time point of the conversion is recorded as the first time point. After the third data is obtained, the recorded first time point is searched, and the current time point at which the third data is obtained is recorded as the second time point for generating the third data. Compare between the first time point and the second time point to get the time interval between the two. The time interval is the time it takes to convert the third data, which is used as the first conversion time corresponding to the third data to be used to determine the difference between the first data and the third data in combination with the second conversion time corresponding to the fourth data. The causal relationship between the two data.

It should be noted that when generating the fourth data, the second data is used as a guide, and the first data is converted through a preset neural network to obtain the intermediate data and the second data for discrimination, and generate the divergence between the two. value to determine the fourth data. The specific generation process of the fourth data is similar to the above-mentioned generation process of the third data, and details are not described here.

In this embodiment, a cyclic update conversion mechanism is set, and for the intermediate data obtained by each conversion, the divergence value calculation is performed with the first data, until the calculated divergence value is greater than the preset threshold, the intermediate data is regarded as close to the first data. The third data of the data. Further, the degree of difficulty of conversion is reflected by the length of time spent in the entire process of obtaining the third data, and the accuracy of determining the causal relationship is improved by accurately determining the third data.

Further, referring to FIG. 3 , based on the first or second embodiment of the method for determining the causal relationship of the present application, a third embodiment of the method for determining the causal relationship of the present application is proposed.

The difference between the third embodiment of the method for determining a causal relationship and the first or second embodiment of the method for determining a causal relationship is that, before the step of acquiring the first data and the second data, the method further includes:

Step S40, obtaining a first data set and a second data set with a preset causal relationship, and processing the first data set and the second data set based on a preset neural network to generate a processing result;

Step S50, judging whether the causal relationship in the processing result is consistent with the preset causal relationship, and if so, the first data and the second data are processed based on the preset neural network.

In this embodiment, the validity of the preset neural network is determined, that is, it is determined whether the preset neural network can effectively determine the causal relationship between data. Specifically, a first data set and a second data set having a preset causal relationship are acquired, and both the first data set and the second data set contain multiple items of data. The preset causal relationship indicates that a clear causal logical relationship is set between the data in the first data set and the data in the second data set. For example, the data in the first data set is the cause in the causal relationship, and the data in the second data set is the result in the causal relationship.

Further, through the preset neural network, the first data set and the second data set are processed, and a processing result is generated, and the processing result indicates that the first data set and the second data set are judged, and the generated items in the two are generated. The causal relationship between cause and effect between data. Then, compare the causal relationship represented by the processing result with the preset causal relationship, and judge whether the two are consistent. If they are consistent, it means that the causal relationship obtained through the preset neural network processing is consistent with the original causal relationship, and the processed results are obtained. Precisely, the preset neural network can be used to determine the causal relationship, and used to process the first data and the second data to determine the causal relationship between the two.

Conversely, if the causal relationship obtained by processing is inconsistent with the original causal relationship, it means that the processing result is inaccurate, and the preset neural network cannot be used to determine the causal relationship, and the update parameters need to be reset for determination. If the preset neural network after updating the parameters can accurately obtain the processing result, the preset neural network after updating the parameters is used to process the first data and the second data. Otherwise, continue to update the parameters, and repeat this cycle until an accurate processing result is obtained, which is used to process the first data and the second data to determine the causal relationship between the two.

It should be noted that both the first data set and the second data set contain multiple pieces of data with preset causal relationships, and the generated processing results contain causal relationships between multiple pieces of data. When the relationship is consistent with the preset causal relationship, its essence is to determine whether the causal relationship between the various data is consistent with the preset causal relationship. Considering the influence of external environmental factors in the processing process, it is difficult to reach a situation where each causal relationship is consistent with the preset causal relationship, so a preset threshold, such as 95%, is set in advance according to the needs. When the proportion of each causal relationship that is consistent with the preset causal relationship is greater than the preset threshold, it is determined to be consistent; otherwise, it is determined to be inconsistent and parameters need to be updated for reprocessing until the proportion is greater than the preset threshold.

This embodiment ensures the accuracy of determining the causal relationship between the first data and the second data by determining whether the preset neural network can effectively determine the causal relationship between the data. At the same time, a proportional mechanism is set up to improve the efficiency of judgment while ensuring accurate judgment.

The present application also provides a device for determining a causal relationship.

Referring to FIG. 4 , FIG. 4 is a schematic diagram of functional modules of a first embodiment of an apparatus for determining a causal relationship of the present application. The means for determining the causal relationship includes:

Further, the determining module includes:

a first determining unit, configured to compare the first conversion difficulty level coefficient with the second conversion difficulty level coefficient, and determine the difference between the first conversion difficulty level coefficient and the second conversion difficulty level coefficient the size relationship between

The first constituting unit is configured to determine the reason why the first data constitutes the causal relationship if the magnitude relationship is that the first conversion difficulty degree coefficient is greater than the second conversion difficulty degree coefficient, and the second data constitutes the causal relationship. the outcome of the causal relationship;

The second structure unit is configured to determine the reason why the second data constitutes the causal relationship if the magnitude relationship is that the first conversion difficulty degree coefficient is smaller than the second conversion difficulty degree coefficient, and the first data constitutes the causal relationship. result in the causal relationship.

Further, the acquisition module includes:

a conversion unit, configured to use the first data as a guide, and convert the second data based on the generator in the preset neural network to obtain intermediate data;

A discriminating unit for discriminating the intermediate data based on a discriminator in a preset neural network, and determining a divergence value between the first data and the intermediate data;

A determining unit, configured to determine, according to the divergence value, third data matching the divergence of the preset neural network with the first data.

Further, the discrimination unit is also used for:

Further, the determining unit is also used for:

Further, the device for determining the causal relationship also includes:

a generating module, configured to obtain a first data set and a second data set with a preset causal relationship, and based on a preset neural network, process the first data set and the second data set to generate a processing result;

The judgment module is configured to judge whether the causal relationship in the processing result is consistent with the preset causal relationship, and if so, process the first data and the second data based on the preset neural network.

The specific implementation manner of the device for determining a causal relationship of the present application is basically the same as the embodiments of the above-mentioned methods for determining a causal relationship, and details are not repeated here.

In addition, an embodiment of the present application also provides a readable storage medium.

A program for determining a causal relationship is stored on the readable storage medium, and when the program for determining a causal relationship is executed by a processor, the steps of the method for determining a causal relationship as described above are implemented.

The readable storage medium of the present application may be a computer-readable storage medium, and its specific implementation is basically the same as that of the foregoing embodiments of the causal relationship determination method, and details are not described herein again.

The embodiments of the present application have been described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific embodiments, which are merely illustrative rather than restrictive. Under the inspiration of this application, without departing from the scope of protection of the purpose of this application and the claims, many forms can be made. Directly or indirectly applied in other related technical fields, these all fall within the protection of this application.

Claims

A method for determining a causal relationship, wherein the method for determining a causal relationship comprises the following steps:

Obtain the first data and the second data, and convert the second data based on the preset neural network and the first data to obtain the first data matching the divergence of the preset neural network with the first data. three data;

Based on the preset neural network and the second data, converting the first data to obtain fourth data matching the divergence of the second data in the preset neural network;

Acquiring a first conversion time for conversion between the second data and the third data, and a second conversion time for conversion between the first data and the fourth data, and according to the first conversion time The corresponding first conversion difficulty degree coefficient and the second conversion difficulty degree coefficient corresponding to the second conversion time determine the causal relationship between the first data and the second data.
The method for determining a causal relationship according to claim 1, wherein the first conversion difficulty degree coefficient corresponding to the first conversion time is based on the second conversion difficulty degree coefficient corresponding to the second conversion time. , the step of determining the causal relationship between the first data and the second data includes:

Comparing the first conversion difficulty degree coefficient with the second conversion difficulty degree coefficient, and determining the magnitude relationship between the first conversion difficulty degree coefficient and the second conversion difficulty degree coefficient;

If the magnitude relationship is that the first conversion difficulty degree coefficient is greater than the second conversion difficulty degree coefficient, determine the reason why the first data constitutes the causal relationship, and the second data constitutes the result in the causal relationship ;

If the magnitude relationship is that the first conversion difficulty degree coefficient is smaller than the second conversion difficulty degree coefficient, determine the reason why the second data constitutes the causal relationship, and the first data constitutes the result in the causal relationship .
The method for determining a causal relationship according to claim 1, wherein the second data is converted based on a preset neural network and the first data to obtain the first data in the preset neural network. The steps of matching the third data on the divergence include:

Using the first data as a guide, convert the second data based on the generator in the preset neural network to obtain intermediate data;

Based on the discriminator in the preset neural network, the intermediate data is discriminated, and the divergence value between the first data and the intermediate data is determined;

According to the divergence value, third data matching the divergence of the preset neural network with the first data is determined.
The method for determining a causal relationship according to claim 3, wherein the intermediate data is discriminated based on a discriminator in a preset neural network, and the dispersion between the first data and the intermediate data is determined. The steps of the degree value include:

based on the discriminator in the preset neural network, discriminate the intermediate data, and generate a first probability distribution of the intermediate data;

A second probability distribution of the first data is acquired, and a divergence value between the first data and the intermediate data is generated according to the first probability distribution and the second probability distribution.
The method for determining a causal relationship according to claim 3, wherein the step of determining, according to the divergence value, third data matching the divergence of the preset neural network with the first data comprises the following steps: :

Determine whether the divergence value is less than a preset threshold, and if it is less than a preset threshold, determine the intermediate data as the third data;

If the divergence value is not less than the preset threshold, perform the step of converting the second data based on the generator in the preset neural network according to the divergence value, until the divergence The value is less than the preset threshold.
The method for determining a causal relationship according to claim 5, wherein after the step of determining the intermediate data as the third data, the method comprises:

Find the first time point of the first conversion of the second data, and determine the second time point of the third data;

According to the first time point and the second time point, a first conversion time corresponding to the third data is generated.
The method for determining a causal relationship according to any one of claims 1-6, wherein, before the step of acquiring the first data and the second data, the method further comprises:

acquiring a first data set and a second data set with a preset causal relationship, and processing the first data set and the second data set based on a preset neural network to generate a processing result;

It is judged whether the causal relationship in the processing result is consistent with the preset causal relationship, and if so, the first data and the second data are processed based on the preset neural network.
A device for determining a causal relationship, wherein the device for determining a causal relationship includes:

The acquisition module is used to acquire the first data and the second data, and based on the preset neural network and the first data, convert the second data to obtain the dispersion of the first data in the preset neural network. degree of matching third data;

a conversion module, configured to convert the first data based on the preset neural network and the second data to obtain fourth data matching the second data in the divergence of the preset neural network;

A determination module, configured to obtain a first conversion time for conversion between the second data and the third data, and a second conversion time for conversion between the first data and the fourth data, and according to the A first conversion difficulty degree coefficient corresponding to the first conversion time and a second conversion difficulty degree coefficient corresponding to the second conversion time are used to determine a causal relationship between the first data and the second data.
A device for determining a causal relationship, wherein the device for determining a causal relationship includes a memory, a processor, and a program for determining a causal relationship that is stored on the memory and can be run on the processor. When the determination program is executed by the processor, the steps of implementing the method for determining a causal relationship according to claim 1 are implemented.
A device for determining a causal relationship, wherein the device for determining a causal relationship includes a memory, a processor, and a program for determining a causal relationship that is stored on the memory and can be run on the processor. When the determination program is executed by the processor, the steps of implementing the method for determining a causal relationship as claimed in claim 2 are implemented.
A device for determining a causal relationship, wherein the device for determining a causal relationship includes a memory, a processor, and a program for determining a causal relationship that is stored on the memory and can be run on the processor. When the determination program is executed by the processor, the steps of implementing the method for determining a causal relationship as claimed in claim 3 are implemented.
A device for determining a causal relationship, wherein the device for determining a causal relationship includes a memory, a processor, and a program for determining a causal relationship that is stored on the memory and can be run on the processor. When the determination program is executed by the processor, the steps of implementing the method for determining a causal relationship as claimed in claim 4 are implemented.
A device for determining a causal relationship, wherein the device for determining a causal relationship includes a memory, a processor, and a program for determining a causal relationship that is stored on the memory and can be run on the processor. When the determination program is executed by the processor, the steps of implementing the method for determining a causal relationship as claimed in claim 5 are implemented.
A device for determining a causal relationship, wherein the device for determining a causal relationship includes a memory, a processor, and a program for determining a causal relationship that is stored on the memory and can be run on the processor. When the determination program is executed by the processor, the steps of implementing the method for determining a causal relationship as claimed in claim 6 are implemented.
A readable storage medium, wherein a program for determining a causal relationship is stored on the readable storage medium, and when the program for determining a causal relationship is executed by a processor, the steps of the method for determining a causal relationship according to claim 1 are implemented .
A readable storage medium, wherein a program for determining a causal relationship is stored on the readable storage medium, and when the program for determining a causal relationship is executed by a processor, the steps of the method for determining a causal relationship according to claim 2 are implemented .
A readable storage medium, wherein a program for determining a causal relationship is stored on the readable storage medium, and when the program for determining a causal relationship is executed by a processor, the steps of the method for determining a causal relationship according to claim 3 are implemented .
A readable storage medium, wherein a program for determining a causal relationship is stored on the readable storage medium, and when the program for determining a causal relationship is executed by a processor, the steps of the method for determining a causal relationship according to claim 4 are implemented .
A readable storage medium, wherein a program for determining a causal relationship is stored on the readable storage medium, and when the program for determining a causal relationship is executed by a processor, the steps of the method for determining a causal relationship according to claim 5 are implemented .
A readable storage medium, wherein a program for determining a causal relationship is stored on the readable storage medium, and when the program for determining a causal relationship is executed by a processor, the steps of the method for determining a causal relationship according to claim 6 are implemented .