Full-link Automated Testing Method and System Based on Internet Finance

Abstract

The full-link automated testing method and system based on Internet finance provided by the embodiments of the present application, when identifying and improving the trading link disturbance data clusters, simultaneously take into account the description knowledge of multiple dimensions of the trading link disturbance data clusters and the influence of the description knowledge of multiple dimensions of other trading link data clusters related to the trading link disturbance data clusters. Based on the description knowledge after integrating the description knowledge of multiple dimensions of multiple trading link data clusters, re-integrate according to the relevance with the trading link disturbance data clusters, so that the obtained integrated description knowledge covers the description knowledge of multiple dimensions of multiple trading link data clusters, thereby adapting to different disturbance ranges of the trading link data clusters. Identify and improve the trading link disturbance data clusters according to the integrated description knowledge, thereby alleviating and narrowing the disturbance range of the trading link data clusters, and making the completeness of the identified and improved trading link data clusters better.

Description

Technical Field

[0001] The present application relates to the field of data processing, and more particularly, to a full-link automated testing method and system based on Internet finance. Background Art

[0002] In the financial field, complex functional divisions involve a large number of business systems for business processing. Different links involve transaction systems to complete corresponding business execution content. For each business link, the background needs to conduct orderly monitoring, which poses challenges to the stability and automatic error correction ability of the entire system. Before the system is put into operation, in order to verify the system performance, full-link testing needs to be carried out to optimize the system according to the test results. Many problems are often found during the testing process. For example, the disturbed link data contained in the link test data is not conducive to subsequent test analysis. How to effectively process these disturbed data is a technical problem that needs to be solved. Summary of the Invention

[0003] The purpose of the present invention is to provide a full-link automated testing method and system based on Internet finance to improve the above-mentioned technical problems.

[0004] The technical solution of the embodiment of the present application is implemented as follows:

[0005] In a first aspect, the embodiment of the present application provides a full-link automated testing method based on Internet finance, which is applied to a financial full-link automated testing system. The method includes:

[0006] In response to a test instruction, obtain a full-link test report of financial transactions; wherein, the full-link test report of financial transactions includes a plurality of transaction link data clusters collected according to a preset period, and the plurality of transaction link data clusters include transaction link disturbed data clusters;

[0007] Mine first description knowledge of multiple dimensions in each of the transaction link data clusters, and respectively integrate the multiple first description knowledge corresponding to the same transaction link data cluster to obtain second description knowledge;

[0008] According to the commonality measurement result between the second description knowledge of the plurality of transaction link data clusters and the second description knowledge of the transaction link disturbed data cluster, perform eccentric calculation and integration on the second description knowledge of the plurality of transaction link data clusters to obtain first integrated description knowledge corresponding to the transaction link disturbed data cluster;

[0009] Identify and improve the transaction link disturbed data cluster according to the first integrated description knowledge;

[0010] Among them, the integrating the multiple first description knowledges corresponding to the same transaction link data cluster to obtain the second description knowledge includes: for the multiple first description knowledges corresponding to any one of the transaction link data clusters, converting the dimensions of the first description knowledges other than the first description knowledge of the target dimension to the target dimension; wherein, the target dimension is a preset dimension among the multiple dimensions; integrating the converted multiple first description knowledges to obtain the second description knowledge of the transaction link data cluster.

[0011] As an implementation manner, the integrating the converted multiple first description knowledges to obtain the second description knowledge of the transaction link data cluster includes:

[0012] Performing knowledge splicing on the converted multiple first description knowledges to obtain spliced description knowledge;

[0013] Performing a linear transformation operation on the spliced description knowledge to obtain the second description knowledge of the transaction link data cluster.

[0014] As an implementation manner, the mining the multiple dimensions of the first description knowledge in each transaction link data cluster includes:

[0015] For each transaction link data cluster among the multiple transaction link data clusters, performing description knowledge mining on the transaction link data cluster to obtain the third description knowledge of the transaction link data cluster;

[0016] Quantifying the third description knowledge to obtain the quantified description knowledge of the first dimension;

[0017] Performing dimension embedding mapping on the quantified description knowledge of the first dimension to obtain the quantified description knowledge of the second dimension;

[0018] Performing information restoration on the quantified description knowledge of the second dimension to obtain the first description knowledge of the second dimension;

[0019] Performing dimension extension mapping on the first description knowledge of the second dimension to obtain the extended description knowledge of the first dimension, and performing information restoration on the extended description knowledge of the first dimension and the quantified description knowledge of the first dimension to obtain the first description knowledge of the first dimension.

[0020] As an implementation manner, the multiple dimensions include M, where M > 2, and the mining the multiple dimensions of the first description knowledge in each transaction link data cluster includes:

[0021] For each transaction link data cluster among the multiple transaction link data clusters, performing description knowledge mining on the transaction link data cluster to obtain the third description knowledge of the transaction link data cluster;

[0022] Quantify the third described knowledge to obtain the quantified description knowledge of the first dimension;

[0023] Perform dimension embedding mapping and quantization on the quantified description knowledge of the first dimension to obtain the quantified description knowledge of the second dimension until the quantified description knowledge of the Nth dimension is obtained, where N = M - 1;

[0024] Perform dimension embedding mapping on the quantified description knowledge of the Nth dimension to obtain the quantified description knowledge of the Mth dimension;

[0025] Perform information restoration on the quantified description knowledge of the Mth dimension to obtain the first description knowledge of the Mth dimension;

[0026] Perform dimension expansion mapping on the first description knowledge of the Mth dimension to obtain the expanded description knowledge of the Nth dimension, and perform information restoration on the expanded description knowledge of the Nth dimension and the quantified description knowledge of the Nth dimension to obtain the first description knowledge of the Nth dimension until the first description knowledge of the first dimension is obtained.

[0027] As an implementation, the quantified description knowledge of the first dimension is a tensor; performing dimension embedding mapping and quantization on the quantified description knowledge of the first dimension to obtain the quantified description knowledge of the second dimension includes:

[0028] Perform dimension embedding mapping on the quantified description knowledge of the first dimension to obtain the embedded description knowledge of the second dimension;

[0029] Decompose the embedded description knowledge to obtain multiple first tensor blocks, and the first tensor block includes the description knowledge on multiple coordinates;

[0030] For each description knowledge, optimize and adjust the description knowledge according to the multiple description knowledge in the first tensor block to which the description knowledge belongs and the coordinate description knowledge of the multiple description knowledge, where the coordinate description knowledge represents the coordinate of the corresponding description knowledge in the first tensor block;

[0031] Establish a second tensor block with the optimized and adjusted multiple description knowledge of the same first tensor block;

[0032] Stitch multiple second tensor blocks according to the coordinates of the multiple first tensor blocks in the embedded description knowledge to obtain a first stitched tensor;

[0033] Determine the quantified description knowledge of the second dimension according to the first stitched tensor.

[0034] As an implementation, for each piece of described knowledge, based on the multiple pieces of described knowledge in the first tensor block to which the described knowledge belongs and the coordinate description knowledge of the multiple pieces of described knowledge, optimizing and adjusting the described knowledge includes:

[0035] For each first tensor block, integrating the multiple pieces of described knowledge in the first tensor block with their corresponding coordinate description knowledge respectively to obtain multiple first integrated description knowledge;

[0036] For each piece of described knowledge in the first tensor block, based on the commonality measurement result between the first integrated description knowledge corresponding to the described knowledge and the multiple first integrated description knowledge, performing an eccentric calculation and integration on the multiple first integrated description knowledge, and determining the described knowledge obtained after the eccentric calculation and integration as the optimized and adjusted described knowledge of the described knowledge;

[0037] The determining the quantization description knowledge of the second dimension based on the first spliced tensor includes: determining the first spliced tensor as the quantization description knowledge of the second dimension;

[0038] Before disassembling the embedded description knowledge to obtain multiple first tensor blocks, the method further includes: performing coordinate transformation on the description knowledge in the embedded description knowledge according to the first coordinate transformation tensor to obtain the optimized and adjusted embedded description knowledge;

[0039] The determining the quantization description knowledge of the second dimension based on the first spliced tensor includes:

[0040] Performing coordinate transformation on the description knowledge in the first spliced tensor according to the second coordinate transformation tensor, and determining the obtained description knowledge tensor after transformation as the quantization description knowledge of the second dimension, where the second coordinate transformation tensor is the inverse transformation tensor of the first coordinate transformation tensor.

[0041] As an implementation, the information restoration of the extended description knowledge of the Nth dimension and the quantization description knowledge of the Nth dimension to obtain the first description knowledge of the Nth dimension includes:

[0042] Integrating the extended description knowledge of the Nth dimension with the quantization description knowledge of the Nth dimension to obtain the integrated description knowledge of the Nth dimension;

[0043] Performing an information restoration operation on the integrated description knowledge of the Nth dimension to obtain the first description knowledge of the Nth dimension.

[0044] As an implementation manner, the full-link automated testing method based on Internet finance is executed according to the full-link automated testing network, and the method further includes the debugging steps of the full-link automated testing network, including:

[0045] Obtain a sample of the full-link test report for financial transactions. The sample of the full-link test report for financial transactions includes multiple transaction link data cluster samples collected according to a preset period. The multiple transaction link data cluster samples include transaction link perturbation data cluster samples, and obtain the transaction link data cluster indication information corresponding to the transaction link perturbation data cluster samples;

[0046] Process the transaction link data cluster samples in the full-link test report sample for financial transactions according to the full-link automated testing network to obtain the inferred transaction link data clusters corresponding to the transaction link perturbation data cluster samples;

[0047] According to the data cluster comparison network, perform descriptive knowledge mining on the transaction link data cluster indication information and the inferred transaction link data clusters respectively to obtain the fourth descriptive knowledge of the transaction link data cluster indication information and the fifth descriptive knowledge of the inferred transaction link data clusters;

[0048] According to the data cluster comparison network, obtain the descriptive knowledge error between the fourth descriptive knowledge and the fifth descriptive knowledge, and determine the descriptive knowledge error as the first quality evaluation factor;

[0049] Debug the full-link automated testing network according to the first quality evaluation factor.

[0050] As an implementation manner, after identifying and improving the transaction link perturbation data clusters according to the first integrated descriptive knowledge, the method further includes:

[0051] Obtain the full-link test report for financial transactions after identification and improvement and the report template library. The report template library includes report templates corresponding to at least one transaction type;

[0052] Perform descriptive knowledge mining on the full-link test report for financial transactions after identification and improvement to obtain the descriptive knowledge of the full-link test report for financial transactions, and perform descriptive knowledge mining on the report template to obtain the report template descriptive knowledge of the report template;

[0053] According to the descriptive knowledge of the full-link test report for financial transactions and the report template descriptive knowledge, classify the full-link test report for financial transactions after identification and improvement based on a preset method to obtain the classification results of the full-link test report for financial transactions after identification and improvement relative to each transaction type under each preset method;

[0054] Determine the transaction type of the fully refined financial transaction full-link test report based on the classification results of the fully refined financial transaction full-link test report for each transaction type under each preset method.

[0055] In a second aspect, an embodiment of the present application provides a financial full-link automated test system, including a processor and a memory. The memory stores a computer program, and when the computer program is executed by the processor, the above method is implemented.

[0056] In the method and system for full-link automated testing based on Internet finance provided by the embodiments of the present application, when identifying and refining the transaction link perturbation data clusters, the influence of the description knowledge of the multi-dimensions of the transaction link perturbation data clusters and the description knowledge of multiple dimensions of other transaction link data clusters related to the transaction link perturbation data clusters is taken into consideration at the same time. Based on the description knowledge after integrating the description knowledge of multiple dimensions of multiple transaction link data clusters, and re-integrating according to the relevance with the transaction link perturbation data clusters, the obtained integrated description knowledge covers the description knowledge of multiple dimensions of multiple transaction link data clusters, so as to adapt to different perturbation ranges of the transaction link data clusters. Identify and refine the transaction link perturbation data clusters based on the integrated description knowledge, thereby alleviating and narrowing the perturbation range of the transaction link data clusters, and making the completeness of the identified and refined transaction link data clusters better.

[0057] In the following description, other features will be partly stated. When examining the following content and the drawings, those skilled in the art will partly discover these features, or may learn these features through production or application. Through practicing or using various aspects of the methods, tools, and combinations listed in the detailed examples described later, the features in the current application can be implemented and obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The drawings here are incorporated into the specification and form a part of this specification. These drawings show embodiments consistent with the present disclosure and are used together with the specification to explain the technical solutions of the present disclosure.

[0059] Figure 1 It is a schematic diagram of the application scenario of the full-link automated testing method based on Internet finance provided by the embodiments of the present application.

[0060] Figure 2 It is a flowchart of a full-link automated testing method based on Internet finance provided by the embodiments of the present application.

[0061] Figure 3 It is a schematic diagram of the functional module architecture of the full-link automated testing device provided by the embodiments of the present application.

[0062] Figure 4It is a schematic diagram of the composition of a financial full-link automated testing system provided by an embodiment of the present application. Specific embodiments

[0063] In order to make the objectives, technical solutions, and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be construed as limiting the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present application.

[0064] In the following descriptions, reference is made to "some embodiments", "as an implementation manner / scheme", "in an implementation manner". These describe subsets of all possible embodiments. However, it can be understood that "some embodiments", "as an implementation manner / scheme", "in an implementation manner" can be the same subsets or different subsets of all possible embodiments, and can be combined with each other without conflict.

[0065] In the following descriptions, the terms "first / second / third" and other similar terms are only used to distinguish similar objects and do not represent a specific order for the objects. It can be understood that "first / second / third" can be interchanged with a specific order or sequence when permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0066] The full-link automated testing method based on Internet finance provided by the embodiments of the present application can be executed by electronic devices such as a financial full-link automated testing system. The electronic device can be various types of terminals such as a laptop computer, a tablet computer, a desktop computer, etc., or can be implemented as a server. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or can also be a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, Content Delivery Network (CDN), and big data and artificial intelligence platforms.

[0067] Next, an exemplary application will be described when the financial full-link automated testing system device is implemented as a server. The technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application.

[0068] Figure 1It is a schematic diagram of the application scenario of the full-link automated testing method based on Internet finance provided by an embodiment of the present application. Communication connections are established between multiple business systems 100 and the financial full-link automated testing system 300 through the network 200. The financial full-link automated testing system 300 is used to execute the method provided by the embodiment of the present application. Specifically, the embodiment of the present application provides a full-link automated testing method based on Internet finance, and this method is applied to the financial full-link automated testing system 300, as Figure 2 shown, this method includes:

[0069] Step S10: In response to a test instruction, obtain a financial transaction full-link test report.

[0070] In the embodiment of the present application, the financial transaction full-link test report includes multiple transaction link data clusters collected according to a preset period, and the multiple transaction link data clusters include transaction link perturbation data clusters. This financial transaction full-link test report is a financial transaction full-link test report obtained by collecting data during the process of conducting a financial transaction full-link test. The financial transaction full-link test can be a virtual financial transaction test during the test process. Each link forms a link topology. The data in the transaction link data cluster can include, but is not limited to, transaction objects, transaction categories, transaction times, transaction amounts, etc. After the virtual transaction is completed, a financial transaction full-link test report is obtained according to the test instruction. In this financial transaction full-link test report, the multiple transaction link data clusters are sorted according to the order of transaction behaviors occurring in the transaction link data cluster, and one transaction link data cluster corresponds to one financial transaction link. The data contained in the transaction link perturbation data cluster has perturbations, such as missing. The transaction link perturbation data cluster is any one of the multiple transaction link data clusters. 4

[0071] Step S20: Mine first description knowledge of multiple dimensions in each transaction link data cluster, and integrate the multiple first description knowledge corresponding to the same transaction link data cluster to obtain second description knowledge.

[0072] The multiple dimensions are used to represent the dimensional values of the description knowledge, such as two-dimensional, three-dimensional. The first description knowledge can be a tensor, such as a second-order tensor, a third-order tensor. The description knowledge can be a vector representation of data mined by an artificial intelligence expert system. Among them, each transaction link data cluster corresponds to first description knowledge of multiple dimensions. The first description knowledge of each dimension represents the corresponding transaction link data cluster, and the first description knowledge of different dimensions contains various characteristic information in the transaction link data cluster.

[0073] For each transaction link data cluster, multiple first description knowledges corresponding to the transaction link data cluster are integrated to obtain the second description knowledge of the transaction link data cluster. In this way, the second description knowledge of each transaction link data cluster is obtained. Since each second description knowledge is determined by integrating the first description knowledges of multiple dimensions of the corresponding transaction link data cluster, and the second description knowledge covers the multi-dimensional description knowledge of the corresponding transaction link data cluster, the information content of the second description knowledge is more abundant.

[0074] Step S30: According to the commonality measurement result between the second description knowledges of multiple transaction link data clusters and the second description knowledge of the transaction link perturbation data cluster, perform eccentric calculation and integration on the second description knowledges of multiple transaction link data clusters to obtain the first integrated description knowledge corresponding to the transaction link perturbation data cluster.

[0075] In the embodiment of the present application, the commonality measurement result between the second description knowledge of any transaction link data cluster and the second description knowledge of the transaction link perturbation data cluster represents the correlation degree between the transaction link data cluster and the transaction link perturbation data cluster. In the present application, there are involvement relationships among multiple transaction link data clusters collected according to a preset period in the same report. According to the correlation degrees between each transaction link data cluster and the transaction link perturbation data cluster, perform eccentric calculation and integration on the second description knowledges of multiple transaction link data clusters (that is, assign different weights to the corresponding second description knowledges according to the correlation degrees, and then perform integration operations on multiple second description knowledges after weight calculation, such as addition or splicing fusion) to obtain the first integrated description knowledge corresponding to the transaction link perturbation data cluster. In this way, the first integrated description knowledge can cover the characteristic information of the transaction link perturbation data cluster and the characteristic information of other transaction link data clusters involved with the transaction link perturbation data cluster at the same time. At the same time, according to the correlation degrees between the remaining transaction link data clusters and the transaction link perturbation data cluster, integrate the characteristic information of the remaining transaction link data clusters to make the characteristic information of the first integrated description knowledge more substantial.

[0076] Step S40: Identify and improve the transaction link perturbation data cluster according to the first integrated description knowledge.

[0077] In the embodiment of the present application, since the first integrated description knowledge integrates the characteristic information of multiple transaction link data clusters, identifying and improving the transaction link perturbation data cluster based on the first integrated description knowledge comprehensively measures the influence of the remaining transaction link data clusters involved with the transaction link perturbation data cluster to identify and improve the transaction link perturbation data cluster, and obtain the transaction link data cluster after identification and improvement, eliminating perturbation information, such as filling in the missing and cleaning up the redundant, which is convenient for subsequent further analysis in the testing process, such as transaction data classification and analysis.

[0078] The method provided in the above steps S10 to S40 measures the influence of the descriptive knowledge of the multi-dimensions of the transaction link perturbation data cluster and the descriptive knowledge of multiple dimensions of other transaction link data clusters related to the transaction link perturbation data cluster when identifying and improving the transaction link perturbation data cluster. It adopts the descriptive knowledge after integrating the descriptive knowledge of the multi-dimensions of multiple transaction link data clusters, and re-integrates it according to the relevance with the transaction link perturbation data cluster, so that the obtained integrated descriptive knowledge covers the descriptive knowledge of the multi-dimensions of multiple transaction link data clusters, thereby adapting to different perturbation ranges of the transaction link data cluster. It identifies and improves the transaction link perturbation data cluster based on the integrated descriptive knowledge, thereby alleviating and narrowing the perturbation range of the transaction link data cluster, and making the integrity of the identified and improved transaction link data cluster better.

[0079] The following further analysis of the above-mentioned test process, such as the classification and analysis of transaction data, will be briefly described in the embodiments of the present application. Among them, for the classification and analysis of transaction data, it may include:

[0080] Step S51: Obtain the fully tested report of the financial transaction full link after identification and improvement and the report template library.

[0081] Among them, the report template library includes report templates corresponding to at least one transaction type.

[0082] Step S52: Mine the descriptive knowledge of the fully tested report of the financial transaction full link after identification and improvement to obtain the descriptive knowledge of the fully tested report of the financial transaction full link after identification and improvement, and mine the descriptive knowledge of the report template to obtain the report template descriptive knowledge of the report template.

[0083] The description knowledge of the fully - linked financial transaction test report after identification improvement and the description knowledge of the report template are both obtained by performing description knowledge mining based on a pre - set neural network model that has been debugged. The neural network model can be a neural network architecture such as CNN, RNN, or LSTM. Before performing description knowledge mining on the fully - linked financial transaction test report after identification improvement to obtain the description knowledge of the fully - linked financial transaction test report, and performing description knowledge mining on the report template to obtain the description knowledge of the report template, it may also include the debugging process of the neural network model, which specifically may include: obtaining a debugging report library, where the debugging report library includes a pre - estimated debugging report corresponding to the annotation transaction type information and debugging report templates corresponding to at least one transaction type; performing description knowledge mining on the pre - estimated debugging report based on the neural network model to be trained to obtain the description knowledge of the pre - estimated debugging report, and performing description knowledge mining on the debugging report template based on the neural network model to be trained to obtain the description knowledge of the debugging report template; according to the description knowledge of the pre - estimated debugging report and the description knowledge of the debugging report template, performing reasoning on the pre - estimated debugging report based on the neural network model to be trained in a preset manner to obtain a set of transaction types of the pre - estimated debugging report, where the set of transaction types includes the target transaction type corresponding to each preset manner; iterating the neural network model to be trained until convergence through the target transaction type corresponding to each preset manner and the transaction type information to obtain a trained neural network model. Among them, according to the description knowledge of the pre - estimated debugging report and the description knowledge of the debugging report template, performing reasoning on the pre - estimated debugging report based on the neural network model to be trained in a preset manner to obtain a set of transaction types of the pre - estimated debugging report, which specifically may include: based on the neural network model to be trained, calculating the matching score between the description knowledge of the pre - estimated debugging report and the description knowledge of the debugging report template in a preset manner to obtain the temporary matching score between the pre - estimated debugging report and the debugging report template under each preset manner; determining the target transaction type of the pre - estimated debugging report under each preset manner according to the temporary matching score; generating a set of transaction types of the pre - estimated debugging report according to the target transaction type of the pre - estimated debugging report under each preset manner.

[0084] Among them, according to the target transaction type corresponding to each preset manner and the transaction type information, iterating the neural network model to be trained until convergence to obtain a trained neural network model, which specifically may include: obtaining the first error result between the target transaction type corresponding to each preset manner and the transaction type information; performing an integration operation on the first error result to obtain a target error result; iterating the neural network model to be trained according to the target error result until convergence to obtain a trained neural network model.

[0085] Step S53: Describe the knowledge of the full-link test report of financial transactions and the knowledge of the report template, and classify the full-link test report of financial transactions based on a preset method to obtain the classification results of the full-link test report of financial transactions for each preset method and each transaction type.

[0086] Specifically, the matching score between the knowledge described in the full-link test report of financial transactions and the knowledge described in the report template can be determined based on a preset method to obtain the target matching score between the full-link test report of financial transactions and the report template corresponding to each transaction type under each preset method; the target matching score is determined as the classification result. There can be multiple preset methods, and the preset method is a strategy for classifying the full-link test report of financial transactions. For example, methods such as cosine distance and Euclidean distance are used to determine the target matching score (similarity).

[0087] Step S54: Determine the transaction type of the full-link test report of financial transactions according to the classification results of the full-link test report of financial transactions for each preset method and each transaction type.

[0088] Specifically, for each transaction type, the classification results corresponding to each preset method are integrated to obtain the target classification result corresponding to the transaction type. For example, the eccentricity coefficient corresponding to each preset method is obtained, and based on the eccentricity coefficient, the classification results corresponding to each preset method are integrated to obtain the target classification result corresponding to the transaction type. Then, according to the target classification result corresponding to the transaction type, the transaction type of the full-link test report of financial transactions is determined.

[0089] Based on the above steps S10 - S40, the second description knowledge of each transaction link data cluster can be optimized and adjusted, or updated, to strengthen the sequential relationship between multiple transaction link data clusters, and then the transaction link perturbation data cluster can be identified and improved.

[0090] As another feasible embodiment, the full-link automated test method based on Internet finance provided in this application may include:

[0091] Step S100: In response to a test instruction, obtain a full-link test report of financial transactions.

[0092] The full - link test report of financial transactions includes multiple clusters of transaction - link data collected according to a preset cycle. The multiple clusters of transaction - link data include a cluster of transaction - link perturbation data. As an implementation, the full - link test report of financial transactions includes multiple consecutive clusters of transaction - link data, and the cluster of transaction - link perturbation data is one of the clusters of transaction - link data among the multiple clusters. Among them, the clusters of transaction - link data included in the full - link test report of financial transactions are arranged in sequence in the clusters of transaction - link data. For any cluster of transaction - link data in the report, multiple clusters of transaction - link data before and after this cluster of transaction - link data are all related to this cluster of transaction - link data. The full - link test report of financial transactions obtained in the embodiments of the present application includes clusters of transaction - link data before and after the cluster of transaction - link perturbation data to accurately identify and improve this cluster of transaction - link data. After determining the cluster of transaction - link perturbation data, multiple clusters of transaction - link data collected according to the preset cycle are determined based on the cluster of transaction - link perturbation data to obtain the full - link test report of financial transactions.

[0093] As an implementation, determining the cluster of transaction - link perturbation data among the multiple clusters of transaction - link data can be to compare each cluster of transaction - link data with the corresponding preset data evaluation index to judge whether the fields are complete or whether there are redundant fields. Of course, it can also be identified manually.

[0094] Step S200: Mine first - description knowledge of multiple dimensions in each cluster of transaction - link data.

[0095] Step S300: For multiple pieces of first - description knowledge corresponding to any cluster of transaction - link data, convert the first - description knowledge other than that of the target dimension to the target dimension.

[0096] In the embodiments of the present application, the target dimension is a preset dimension among the multiple dimensions. For any cluster of transaction - link data, multiple pieces of first - description knowledge corresponding to the cluster of transaction - link data are of different multiple dimensions. Converting the first - description knowledge other than that of the target dimension among the multiple pieces of first - description knowledge to the target dimension, then the converted multiple pieces of first - description knowledge are all of the target dimension, and the converted multiple pieces of first - description knowledge are of the same dimension, which is conducive to integrating the converted multiple pieces of first - description knowledge. As an implementation, the way to convert the other first - description knowledge to the target dimension can be achieved through up - sampling.

[0097] Step S400: Integrate the converted multiple pieces of first - description knowledge of the cluster of transaction - link data to obtain the second - description knowledge of the cluster of transaction - link data.

[0098] Since the multiple first description knowledges after conversion are of the same dimension (both are the target dimension), the multiple first description knowledges after conversion can be directly integrated to obtain the second description knowledge of the transaction link data cluster. Based on the above steps S100 - S400, the second description knowledge of each transaction link data cluster is obtained after processing the multiple first description knowledges of each transaction link data cluster.

[0099] As an implementation manner, step S400 may specifically include: performing knowledge splicing on the multiple first description knowledges after conversion to obtain a spliced description knowledge, and performing a linear transformation operation (such as filtering and smoothing based on a convolution matrix) on the spliced description knowledge to obtain the second description knowledge of the transaction link data cluster. Based on this, by performing knowledge splicing and convolution to process the multiple first description knowledges after conversion of the transaction link data cluster, the description knowledges of multiple dimensions of the transaction link data cluster can be more fully integrated, and the accuracy of the second description knowledge is improved. Among them, the dimension of the second description knowledge is the target dimension.

[0100] Step S500: For the target description knowledge among the multiple second description knowledges, according to the commonality measurement result between the multiple second description knowledges and the target description knowledge, perform eccentric calculation and integration on the multiple second description knowledges to obtain a second integrated description knowledge.

[0101] The target description knowledge is any one of the multiple first description knowledges. The commonality measurement result between any second description knowledge and the target description knowledge represents the correlation degree between the second description knowledge and the target description knowledge, and at the same time represents the correlation degree between the transaction link data cluster corresponding to the second description knowledge and the transaction link data cluster corresponding to the target description knowledge. According to the commonality measurement result between the multiple second description knowledges and the target description knowledge, perform eccentric calculation and integration on the multiple second description knowledges. The second integrated description knowledge obtained by integration covers both the second description knowledge of the corresponding transaction link data cluster and the description knowledge of other transaction link data clusters involved in the transaction link data cluster.

[0102] Step S600: Integrate the second integrated description knowledge with the target description knowledge, and determine the integrated description knowledge as the description knowledge after optimized adjustment of the target description knowledge.

[0103] The optimized adjusted description knowledge is the optimized adjusted description knowledge of the transaction link data cluster corresponding to the target description knowledge. Based on steps S500 and S600, each second description knowledge is optimized and updated to obtain the optimized adjusted description knowledge of each second description knowledge, and then the optimized adjusted description knowledge of multiple transaction link data clusters is obtained. According to the second description knowledge of multiple transaction link data clusters, the second description knowledge of each transaction link data cluster is optimized one by one, strengthening the sequential connection between the description knowledge of multiple transaction link data clusters and the relevance between the optimized adjusted description knowledge of multiple transaction link data clusters. In addition, eccentric calculation and integration are performed on multiple second description knowledge to obtain the second integrated description knowledge corresponding to the target description knowledge, and then the integrated description knowledge is integrated with the target description knowledge to ensure that the obtained optimized adjusted description knowledge is adapted to the transaction link data cluster corresponding to the target description knowledge, and the optimized adjusted description knowledge is more accurate.

[0104] As an implementation, after obtaining the second description knowledge of multiple transaction link data clusters, each second description knowledge is optimized and adjusted based on steps S500 and S600 through an attention network (such as a multi-head attention mechanism) to obtain the optimized adjusted description knowledge of each second description knowledge.

[0105] Step S700: According to the commonality measurement result between the optimized adjusted description knowledge of multiple transaction link data clusters and the optimized adjusted description knowledge of the transaction link perturbation data cluster, eccentric calculation and integration are performed on the optimized adjusted description knowledge of multiple transaction link data clusters to obtain the first integrated description knowledge corresponding to the transaction link perturbation data cluster.

[0106] In the embodiment of the present application, the commonality measurement result between the optimized adjusted description knowledge of any transaction link data cluster and the optimized adjusted description knowledge of the transaction link perturbation data cluster represents the relevance between the transaction link data cluster and the transaction link perturbation data cluster.

[0107] Step S800: Identify and improve the transaction link perturbation data cluster according to the first integrated description knowledge.

[0108] Because the first integrated description knowledge integrates the description knowledge of multiple dimensions of multiple transaction link data clusters, the information content of the first integrated description knowledge is more substantial. Identifying and improving the transaction link perturbation data cluster according to the first integrated description knowledge can comprehensively consider the influence generated by other transaction link data clusters, thereby making the identified and improved transaction link data cluster more accurate. As an implementation, a linear transformation operation is performed on the first integrated description knowledge to obtain the identified and improved transaction link data cluster.

[0109] As an implementation manner, step S800 may specifically include: integrating the optimized and adjusted description knowledge of the first integrated description knowledge and the transaction link perturbation data cluster to obtain the third integrated description knowledge, and identifying and improving the transaction link perturbation data cluster based on the third integrated description knowledge. After obtaining the first integrated description knowledge corresponding to the transaction link perturbation data cluster, the first integrated description knowledge and the optimized and adjusted description knowledge of the transaction link perturbation data cluster are integrated again to make the obtained third integrated description knowledge adapt to the transaction link perturbation data cluster, ensuring that the accurately obtained third integrated description knowledge can be obtained, thereby ensuring the accuracy of the identified and improved transaction link data cluster. After obtaining the optimized and adjusted description knowledge of multiple second description knowledge, based on the attention network, the first integrated description knowledge is obtained according to step S700, and the first integrated description knowledge and the optimized and adjusted description knowledge of the transaction link perturbation data cluster are integrated to obtain accurate third integrated description knowledge. After obtaining the second description knowledge of each transaction link data cluster, first, the second description knowledge of each transaction link data cluster is adjusted and optimized according to the second description knowledge of multiple transaction link data clusters, and then the first integrated description knowledge is obtained by using the optimized and adjusted second description knowledge of multiple transaction link data clusters, or steps S500 and S600 are skipped, and directly, based on the commonality measurement result between the second description knowledge of multiple transaction link data clusters and the second description knowledge of the transaction link perturbation data cluster, eccentric calculation and integration are performed on the second description knowledge of multiple transaction link data clusters to obtain the first integrated description knowledge. As an implementation manner, when the first integrated description knowledge is obtained by using the second description knowledge of multiple transaction link data clusters, identifying and improving the transaction link perturbation data cluster includes: integrating the first integrated description knowledge and the second description knowledge of the transaction link perturbation data cluster to obtain the fourth integrated description knowledge, and identifying and improving the transaction link perturbation data cluster based on the fourth integrated description knowledge.

[0110] When identifying and improving the transaction link perturbation data cluster, the influence of the description knowledge of the multi-dimensions of the transaction link perturbation data cluster and the description knowledge of multiple dimensions of other transaction link data clusters involved in the transaction link perturbation data cluster is measured at the same time. Based on the description knowledge after integrating the description knowledge of the multi-dimensions of multiple transaction link data clusters, and re-integrating according to the correlation degree with the transaction link perturbation data cluster, the obtained integrated description knowledge covers the description knowledge of the multi-dimensions of multiple transaction link data clusters, so as to adapt to different perturbation ranges of the transaction link data cluster. Based on the integrated description knowledge, the transaction link perturbation data cluster is identified and improved, the perturbation range of the transaction link data cluster is alleviated, and the identified and improved transaction link data cluster is more accurate.

[0111] In addition, after obtaining the second description knowledge of each transaction link data cluster, the second description knowledge of each transaction link data cluster is adjusted and optimized according to the second description knowledge of multiple transaction link data clusters, so as to increase the sequential continuity between the description knowledge of multiple transaction link data clusters, improve the accuracy of the description knowledge of each optimized and adjusted transaction link data cluster, and further improve the accuracy of the identified and improved transaction link data cluster. In addition, after obtaining the first integrated description knowledge corresponding to the transaction link perturbation data cluster, the first integrated description knowledge is re-integrated with the optimized and adjusted description knowledge of the transaction link perturbation data cluster, so that the obtained third integrated description knowledge is adapted to the transaction link perturbation data cluster, ensuring the accuracy of the obtained third integrated description knowledge, and thus ensuring the accuracy of the identified and improved transaction link data cluster.

[0112] When there are M dimensions, where M>2, the first description knowledge of multiple dimensions is mined in the transaction link data cluster according to the increasing order of multiple dimensions. As an embodiment, the process of mining the first description knowledge of multiple dimensions in each transaction link data cluster includes:

[0113] (1) For each transaction link data cluster among multiple transaction link data clusters, description knowledge mining is performed on the transaction link data cluster to obtain the third description knowledge of the transaction link data cluster. Among them, the third description knowledge indicates the vector expression of the transaction link data cluster, such as a tensor (such as a second-order tensor, a matrix).

[0114] As an implementation method, in a large-scale cluster transaction link test scenario, each transaction link data cluster contains a large amount of data. Then, this step (1) may specifically include: disassembling the transaction link data cluster to obtain multiple data sub-clusters, performing description knowledge mining on each data sub-cluster respectively to obtain the description knowledge of each data sub-cluster; and splicing the description knowledge of multiple data sub-clusters according to the coordinates of multiple data sub-clusters in the transaction link data cluster to obtain the third description knowledge of the transaction link data cluster. Using disassembly and then integration to obtain the third description knowledge of the transaction link data cluster can avoid directly performing description knowledge mining on a large-scale transaction link data cluster and relieve the computational overhead. As an implementation method, the description knowledge of each data sub-cluster is a tensor (such as a second-order tensor), and the description knowledge of multiple data sub-clusters is spliced into a tensor according to the coordinates of multiple data sub-clusters in the transaction link data cluster, and the tensor obtained by knowledge splicing is determined as the third description knowledge of the transaction link data cluster.

[0115] (2) Quantify (such as encode) the third description knowledge to obtain the quantified description knowledge of the first dimension.

[0116] In step (2), the quantization description knowledge of the first dimension can be tensors of different orders such as second-order tensors and third-order tensors. Since the third description knowledge indicates the eigenvector of the transaction link data cluster, quantizing the third description knowledge can strengthen the involvement relationship between the description knowledge in the third description knowledge and enhance the expression effect of the quantization description knowledge on the transaction link data cluster. As an implementation manner, the third description knowledge is the first dimension. Quantizing the third description knowledge only enhances the expression effect of the description knowledge on the transaction link data cluster and does not convert the dimension of the description knowledge.

[0117] (3) Perform dimension embedding mapping and quantization on the quantization description knowledge of the first dimension to obtain the quantization description knowledge of the second dimension until the quantization description knowledge of the Nth dimension is obtained, where N = M - 1.

[0118] Among them, the quantization description knowledge of multiple dimensions is obtained one by one according to the order of decreasing multiple dimensions. The first dimension is the preset dimension among multiple dimensions. According to the dimension embedding mapping of the quantization description knowledge of the first dimension, the purpose of dimensionality reduction is completed, so as to obtain the description knowledge of the second dimension, and at the same time, re-quantization is performed to enhance the expression effect of the description knowledge of the second dimension on the transaction link data cluster again. Repeat the process of obtaining the quantization description knowledge of the second dimension to obtain the quantization description knowledge of the next dimension until the quantization description knowledge of the Nth dimension is obtained.

[0119] (4) Perform dimension embedding mapping on the quantization description knowledge of the Nth dimension to obtain the quantization description knowledge of the Mth dimension.

[0120] Adopt dimension embedding mapping on the quantization description knowledge of the Nth dimension to obtain the description knowledge of the Mth dimension, and determine the obtained description knowledge of the Mth dimension as the quantization description knowledge of the Mth dimension.

[0121] (5) Perform information restoration on the quantization description knowledge of the Mth dimension to obtain the first description knowledge of the Mth dimension.

[0122] Perform information restoration on the Mth quantization description knowledge to strengthen the involvement relationship between the description knowledge in the description knowledge and enhance the expression effect of the quantization description knowledge on the transaction link data cluster. Among them, when performing information restoration on the description knowledge, only the involvement between the description knowledge in the description knowledge is strengthened, and the dimension of the description knowledge is not converted.

[0123] (6) Perform dimension expansion mapping on the first description knowledge of the Mth dimension to obtain the expanded description knowledge of the Nth dimension; perform information restoration on the expanded description knowledge of the Nth dimension and the quantization description knowledge of the Nth dimension to obtain the first description knowledge of the Nth dimension until the first description knowledge of the first dimension is obtained.

[0124] In the above steps, the Mth dimension among the M dimensions is the smallest dimension among the M dimensions, and the first dimension is a preset dimension among the multiple dimensions. From the first dimension to the Nth dimension, the dimensions gradually decrease. Among them, the first description knowledge of the multiple dimensions is obtained one by one according to the increasing order of the multiple dimensions. At the same time, when obtaining the first description knowledge of the dimensions other than the first description knowledge of the Mth dimension, they are all obtained based on the first description knowledge of the adjacent smaller dimension and the quantization description knowledge of the current dimension, so as to strengthen the expression effect of the first description knowledge of the multiple dimensions on the transaction link data cluster. The first description knowledge of the Mth dimension is subjected to dimension expansion mapping, and at the same time, the first description knowledge of the Nth dimension is obtained based on the extended description knowledge of the Nth dimension and the quantization description knowledge of the Nth dimension. Based on this rule, the first description knowledge of the first dimension is obtained until the first description knowledge of the first dimension is obtained.

[0125] The above steps (1)-(6) are described by taking any transaction link data cluster as an example. However, for multiple transaction link data clusters, the first description knowledge of the M dimensions of each transaction link data cluster can be obtained according to the above steps. For each transaction link data cluster among the multiple transaction link data clusters, starting from the third description knowledge of the transaction link data cluster, the first description knowledge of the multiple dimensions of each transaction link data cluster is obtained through multiple quantizations and multiple information restorations, strengthening the expression effect of each first description knowledge on the transaction link data cluster, and increasing the accuracy of the first description knowledge of the multiple dimensions of each transaction link data cluster. In addition, when obtaining the first description knowledge of the dimensions other than the first description knowledge of the Mth dimension, they are all determined based on the first description knowledge of the adjacent smaller dimension and the quantization description knowledge of the current dimension, strengthening the expression effect of the first description knowledge of the multiple dimensions on the transaction link data cluster.

[0126] In addition, after performing dimension embedding mapping on the quantization description knowledge of the first dimension, the embedded description knowledge is first disassembled into multiple tensor blocks for optimization adjustment, and then integrated to obtain the second quantization description knowledge. Obtaining the second quantization description knowledge may specifically include:

[0127] (I) Perform dimension embedding mapping on the quantization description knowledge of the first dimension to obtain the embedded description knowledge of the second dimension.

[0128] In the embodiment of the present application, the second dimension is smaller than the first dimension. The dimension embedding mapping of the quantization description knowledge can be implemented according to any feasible dimension embedding mapping method, such as the embedding descent method. As an implementation manner, the second dimension is half of the first dimension.

[0129] (II) Disassemble the embedded description knowledge of the second dimension to obtain multiple first tensor blocks.

[0130] Among them, the embedded description knowledge of the second dimension is a tensor, and the embedded description knowledge is disassembled to obtain multiple first tensor blocks. Each first tensor block includes description knowledge on multiple coordinates. For example, if the first tensor block is a second-order tensor or a third-order tensor, and the multiple description knowledge covered by the first tensor block are all scalars, then the first tensor block is a second-order tensor; if the multiple description knowledge included in the first tensor block are all vectors, then the first tensor block is a third-order tensor. As an implementation, the data size (here referring to the dimension) of each first tensor block is the same. In other words, the number of description knowledge on multiple coordinates included in each first tensor block is the same. As an implementation, when the embedded description knowledge is a third-order tensor, the multiple first tensor blocks obtained by disassembling the embedded description knowledge are also third-order tensors; when the embedded description knowledge is a second-order tensor, the multiple first tensor blocks obtained by disassembling the embedded description knowledge are also second-order tensors.

[0131] (III) For each piece of description knowledge, according to the multiple pieces of description knowledge and the coordinate description knowledge of the multiple pieces of description knowledge in the first tensor block to which the description knowledge belongs, optimize and adjust the description knowledge.

[0132] Among them, the coordinate description knowledge represents the coordinates of the corresponding description knowledge in the first tensor block, and the coordinate description knowledge can be a first-order tensor (i.e., a vector) or a second-order tensor (i.e., a matrix). For each first tensor block, since the first tensor block includes multiple pieces of description knowledge, optimize and adjust each piece of description knowledge according to the multiple pieces of description knowledge and the coordinate description knowledge of the multiple pieces of description knowledge. In this way, each piece of description knowledge not only integrates its own coordinate description knowledge, but also integrates the other description knowledge in the same first tensor block and the coordinate description knowledge of the other description knowledge, thereby strengthening the involvement relationship between the description knowledge at different positions in the same first tensor block.

[0133] As an implementation manner, step (III) specifically includes: for each first tensor block, integrating multiple description knowledges in the first tensor block with the corresponding coordinate description knowledges respectively to obtain multiple first integrated description knowledges; for each description knowledge in the first tensor block, according to the commonality measurement result between the first integrated description knowledge corresponding to the description knowledge and the multiple first integrated description knowledges, performing eccentric calculation and integration on the multiple first integrated description knowledges, and determining the description knowledge obtained after the eccentric calculation and integration as the description knowledge after optimization and adjustment. Wherein, for any description knowledge in the first tensor block, the commonality measurement result between the first integrated description knowledge corresponding to the description knowledge and the multiple first integrated description knowledges represents the correlation degree between the first integrated description knowledges corresponding to the multiple description knowledges in the first tensor block and the first integrated description knowledge corresponding to this description knowledge. According to the commonality measurement result between the multiple first integrated description knowledges and the first integrated description knowledge corresponding to the description knowledge, performing eccentric calculation and integration on the multiple first integrated description knowledges, and determining the description knowledge obtained after the eccentric calculation and integration as the description knowledge after optimization and adjustment, so that the description knowledges at the remaining positions integrated in the description knowledge after optimization and adjustment are integrated according to the correlation degree with the remaining description knowledges, thus increasing the accuracy of the description knowledge after optimization and adjustment.

[0134] (IV) Establish a second tensor block with the multiple description knowledges after optimization and adjustment of the same first tensor block.

[0135] For the same first tensor block, obtain the description knowledges after optimization and adjustment of the multiple description knowledges in the first tensor block, that is, obtain the multiple description knowledges after optimization and adjustment of the first tensor block, and establish a second tensor block with the multiple description knowledges after optimization and adjustment according to the coordinates of the multiple description knowledges in the first tensor block.

[0136] (V) Stitch multiple second tensor blocks according to the coordinates of the multiple first tensor blocks in the embedded description knowledge in the second dimension to obtain a first stitched tensor.

[0137] The multiple first tensor blocks are disassembled from the embedded description knowledge, and the coordinates of different first tensor blocks in the embedded description knowledge are different. After obtaining the second tensor blocks corresponding to each first tensor block, stitch the multiple second tensor blocks according to the coordinates of the multiple first tensor blocks in the embedded description knowledge to ensure that the coordinates of the multiple second tensor blocks in the first stitched tensor are the same as the coordinates of the corresponding first tensor blocks in the embedded description knowledge, so as to ensure that the obtained first stitched tensor is accurate.

[0138] As an implementation manner, step (V) may specifically include: performing a description knowledge mapping transformation on each second tensor block to obtain an optimized and adjusted second tensor block, and splicing the optimized and adjusted second tensor blocks according to the coordinates of the multiple first tensor blocks in the embedding description knowledge of the second dimension to obtain a first spliced tensor. After obtaining each second tensor block, perform a description knowledge mapping transformation on each second tensor block (for example, through linear mapping) to project each second tensor block onto the target description knowledge value range, so as to obtain multiple optimized and adjusted second tensor blocks. In other words, the multiple optimized and adjusted second tensor blocks are located in the target description knowledge value range. In the embodiment of the present application, by first disassembling the embedding description knowledge into multiple tensor blocks, optimizing and adjusting each of the multiple tensor blocks respectively, and then splicing the optimized and adjusted tensor blocks, the description knowledge (first spliced tensor) after optimizing and adjusting the embedding description knowledge is obtained. Since the number of description knowledge covered by each tensor block is less than the number of description knowledge covered by the embedding description knowledge, disassembling it into multiple tensor blocks and then optimizing and adjusting the description knowledge in each tensor block respectively can make the number of description knowledge less when optimizing and debugging each tensor block, alleviating the computational overhead.

[0139] (VI) Determine the first spliced tensor as the quantization description knowledge of the second dimension.

[0140] Since the embedding description knowledge is the second dimension, and the optimized and adjusted first spliced description knowledge is also the second dimension, the first spliced tensor is the quantization description knowledge of the second dimension. In the embodiment of the present application, by directly determining the first spliced tensor as the quantization description knowledge of the second dimension, other embodiments may skip step (VI) and determine the quantization description knowledge of the second dimension according to the first spliced tensor. As an implementation manner, after obtaining the first spliced tensor, perform a description knowledge mapping transformation on the first spliced tensor to obtain the quantization description knowledge of the second dimension. Perform a description knowledge mapping transformation on the first spliced tensor, so as to project the first spliced tensor onto the target description knowledge value range and obtain the quantization description knowledge of the second dimension.

[0141] After obtaining the first integrated description knowledge, the first integrated description knowledge may also be adjusted multiple times according to the process of optimizing the embedding description knowledge of the second dimension in steps (II) to (V), and the transaction link perturbation data cluster may be identified and improved according to the optimized and adjusted first integrated description knowledge. After obtaining the first integrated description knowledge, adjust the first integrated description knowledge multiple times to enhance the expression effect of the first integrated description knowledge on the transaction link data cluster.

[0142] When quantifying the dimensional embedding mapping information of the obtained second dimension, coordinate transformation is performed on the descriptive knowledge in the embedded descriptive knowledge, and then the transformed embedded descriptive knowledge is optimized by disassembling. Then, inverse coordinate transformation is performed on the descriptive knowledge in the optimized and adjusted embedded descriptive knowledge to increase the involvement relationship between the descriptive knowledge in the finally obtained descriptive knowledge.

[0143] The specific steps for obtaining the second quantified descriptive knowledge may specifically include:

[0144] (a)Perform dimensional embedding mapping on the quantified descriptive knowledge of the first dimension to obtain the embedded descriptive knowledge of the second dimension.

[0145] (b)According to the first coordinate transformation tensor, perform coordinate transformation on the descriptive knowledge in the embedded descriptive knowledge to obtain the optimized and adjusted embedded descriptive knowledge.

[0146] The first coordinate transformation tensor is used to transform the coordinates of the descriptive knowledge contained in the tensor. As an implementation manner, if the embedded descriptive knowledge of the second dimension is a second-order tensor, then the tensor obtained by multiplying the first coordinate transformation tensor and the embedded descriptive knowledge is the optimized and adjusted embedded descriptive knowledge.

[0147] (c)Disassemble the optimized and adjusted embedded descriptive knowledge to obtain multiple first tensor blocks.

[0148] (d)For each descriptive knowledge, optimize and adjust the descriptive knowledge according to the multiple descriptive knowledge and the coordinate descriptive knowledge of the multiple descriptive knowledge in the first tensor block to which the descriptive knowledge belongs.

[0149] (e)Establish a second tensor block for the multiple descriptive knowledge after optimization and adjustment of the same first tensor block.

[0150] (f)According to the coordinates of the multiple first tensor blocks in the optimized and adjusted embedded descriptive knowledge, splice the multiple second tensor blocks to obtain a first spliced tensor.

[0151] (g)According to the second coordinate transformation tensor, perform coordinate transformation on the descriptive knowledge in the first spliced tensor, and determine the descriptive knowledge tensor obtained after transformation as the quantified descriptive knowledge of the second dimension.

[0152] The second coordinate transformation tensor is the inverse transformation tensor of the first coordinate transformation tensor. After obtaining the first spliced tensor, perform coordinate transformation on the descriptive knowledge in the first spliced tensor according to the second coordinate transformation tensor, so that the coordinates of each descriptive knowledge in the obtained descriptive knowledge tensor are the same as the coordinates of each descriptive knowledge before adjustment in the embedded descriptive knowledge, completing the reduction of the descriptive knowledge coordinates and ensuring the accuracy of the obtained quantified descriptive knowledge.

[0153] As an implementation manner, step (g) specifically includes: performing coordinate transformation on the description knowledge in the first splicing tensor, and then performing description knowledge mapping transformation on the obtained description knowledge tensor after transformation to obtain the quantified description knowledge of the second dimension.

[0154] In the above process, when quantifying the embedded description knowledge, first perform coordinate transformation on the description knowledge in the embedded description knowledge to obtain the optimized and adjusted embedded description knowledge, and then disassemble the optimized and adjusted embedded description knowledge into multiple first tensor blocks for optimization. It is possible that each description knowledge after optimization and adjustment integrates the description knowledge on other coordinates in different local ranges, strengthening the correlation between the description knowledge on different coordinates in the embedded description knowledge and improving the accuracy of the quantified description knowledge.

[0155] In addition, the specific steps for information restoration of the extended description knowledge of the Nth dimension and the quantified description knowledge of the Nth dimension may include:

[0156] (H1) Integrate the extended description knowledge of the Nth dimension with the quantified description knowledge of the Nth dimension to obtain the integrated description knowledge of the Nth dimension.

[0157] The extended description knowledge of the Nth dimension and the quantified description knowledge of the Nth dimension can both exist in any form. For example, if both the extended description knowledge of the Nth dimension and the quantified description knowledge of the Nth dimension are second-order tensors, then add the extended description knowledge of the Nth dimension and the quantified description knowledge of the Nth dimension, and the obtained tensor is the integrated description knowledge of the Nth dimension.

[0158] (H2) Disassemble the integrated description knowledge of the Nth dimension to obtain multiple third tensor blocks.

[0159] (H3) For each description knowledge, optimize and adjust the description knowledge according to the multiple description knowledge in the third tensor block to which the description knowledge belongs and the coordinate description knowledge of the multiple description knowledge.

[0160] (H4) Establish a fourth tensor block for the multiple description knowledge after optimization and adjustment in the same third tensor block.

[0161] (H5) According to the coordinates of the multiple third tensor blocks in the integrated description knowledge of the Nth dimension, splice the multiple fourth tensor blocks to obtain the second splicing tensor.

[0162] (H6) Determine the second splicing tensor as the first description knowledge of the Nth dimension.

[0163] As an implementation, after obtaining the second spliced tensor, a description knowledge mapping transformation is performed on the second spliced tensor to obtain the quantified description knowledge of the Nth dimension. By performing a description knowledge mapping transformation on the second spliced tensor, the second spliced tensor is projected onto the target description knowledge value range to obtain the quantified description knowledge of the Nth dimension.

[0164] Based on the above steps, before disassembling the integrated description knowledge of the Nth dimension, first perform a coordinate transformation on the description knowledge in the integrated description knowledge, then disassemble it into tensor blocks for quantization, and then perform an inverse coordinate transformation on the tensor obtained by splicing multiple tensor block knowledge, so as to improve the correlation between the description knowledge in the finally obtained description knowledge. The information reduction of the extended description knowledge of the Nth dimension and the quantified description knowledge of the Nth dimension includes:

[0165] (K1) Integrate the extended description knowledge of the Nth dimension and the quantified description knowledge of the Nth dimension to obtain the integrated description knowledge of the Nth dimension.

[0166] (K2) According to the third coordinate transformation tensor, perform a coordinate transformation on the description knowledge in the integrated description knowledge to obtain the optimized and adjusted integrated description knowledge.

[0167] (K3) Disassemble the optimized and adjusted integrated description knowledge to obtain multiple third tensor blocks.

[0168] (K4) For each description knowledge, according to the multiple description knowledge and the coordinate description knowledge of the multiple description knowledge in the third tensor block to which the description knowledge belongs, optimize and adjust the description knowledge.

[0169] (K5) The multiple description knowledge after optimized adjustment of the same third tensor block form the fourth tensor block.

[0170] (K6) According to the coordinates of the multiple third tensor blocks in the optimized and adjusted integrated description knowledge, splice the multiple fourth tensor blocks to obtain the second spliced tensor.

[0171] (K7) According to the fourth coordinate transformation tensor, perform a coordinate transformation on the description knowledge in the second spliced tensor, and determine the obtained description knowledge tensor after transformation as the first description knowledge of the Nth dimension.

[0172] Among them, the fourth coordinate transformation tensor is the inverse transformation tensor of the third coordinate transformation tensor. After obtaining the second spliced tensor, according to the fourth coordinate transformation tensor, perform a coordinate transformation on the description knowledge in the second spliced tensor, so that the coordinates of each description knowledge in the obtained description knowledge tensor after transformation are the same as the coordinates of each description knowledge before optimization in the extended description knowledge of the Nth dimension, and complete the reduction of the coordinates where the description knowledge is located.

[0173] As an implementation manner, step (K7) includes: after performing coordinate transformation on the description knowledge in the second splicing tensor, performing a description knowledge mapping transformation on the obtained description knowledge tensor after transformation to obtain the quantified description knowledge of the Nth dimension.

[0174] The description of the above embodiments is based on M > 2. In another embodiment, the number of dimensions is equal to 2. Then, mining the first description knowledge of multiple dimensions in each transaction link data cluster may specifically include:

[0175] (X1) For each transaction link data cluster among multiple transaction link data clusters, perform description knowledge mining on the transaction link data cluster to obtain the third description knowledge of the transaction link data cluster.

[0176] (X2) Quantify the third description knowledge to obtain the quantified description knowledge of the first dimension.

[0177] (X3) Perform a dimension embedding mapping on the quantified description knowledge of the first dimension to obtain the quantified description knowledge of the second dimension.

[0178] (X4) Perform information restoration on the quantified description knowledge of the second dimension to obtain the first description knowledge of the second dimension.

[0179] (X5) Perform a dimension expansion mapping on the first description knowledge of the second dimension to obtain the expanded description knowledge of the first dimension, and perform information restoration on the expanded description knowledge of the first dimension and the quantified description knowledge of the first dimension to obtain the first description knowledge of the first dimension.

[0180] Processing the transaction link data cluster is executed according to the full-link automated test network. Before that, it is necessary to debug the full-link automated test network. The specific process of debugging may include:

[0181] (Y1) Obtain a sample of the full-link test report of financial transactions.

[0182] The sample of the full-link test report of financial transactions includes multiple samples of transaction link data clusters collected according to a preset period. The multiple samples of transaction link data clusters include samples of transaction link perturbation data clusters, and obtain the transaction link data cluster indication information corresponding to the samples of transaction link perturbation data clusters. Among them, the sample of the full-link test report of financial transactions is a virtual financial transaction test during the test process. Each link forms a link topology, and the sample of the full-link test report of financial transactions includes samples of transaction link perturbation data clusters. The data integrity of the transaction link data cluster indication information is greater than that of the sample of the transaction link perturbation data cluster. For example, the transaction link data cluster indication information is a transaction link data cluster obtained by improving the sample of the transaction link perturbation data cluster. For example, the transaction link data cluster indication information is obtained by manually improving the sample of the transaction link perturbation data cluster.

[0183] (Y2) Process the transaction link data cluster samples in the full-link automated test report sample of financial transactions according to the full-link automated test network to obtain the inferred transaction link data cluster corresponding to the transaction link perturbation data cluster sample.

[0184] (Y3) Conduct descriptive knowledge mining on the transaction link data cluster indication information and the inferred transaction link data cluster respectively according to the data cluster comparison network to obtain the fourth descriptive knowledge of the transaction link data cluster indication information and the fifth descriptive knowledge of the inferred transaction link data cluster.

[0185] The data cluster comparison network is used to compare the descriptive knowledge of data. The fourth descriptive knowledge is used to indicate the transaction link data cluster indication information, and the fifth descriptive knowledge is used to indicate the inferred transaction link data cluster.

[0186] (Y4) Determine the descriptive knowledge error between the fourth descriptive knowledge and the fifth descriptive knowledge according to the data cluster comparison network, and determine the descriptive knowledge error as the first quality evaluation factor.

[0187] The descriptive knowledge error represents the cost between the fourth descriptive knowledge and the fifth descriptive knowledge. According to the descriptive knowledge error, the accuracy of the full-link automated test network can be reflected. The descriptive knowledge error is determined as the quality evaluation factor for debugging the full-link automated test network (a coefficient for evaluating accuracy, such as loss).

[0188] (Y5) Debug the full-link automated test network according to the first quality evaluation factor.

[0189] Debug the full-link automated test network according to the first quality evaluation factor to increase the accuracy of the full-link automated test network. As an implementation, debugging the full-link automated test network may specifically include: obtaining the second quality evaluation factor according to the data difference between the transaction link data cluster indication information and the inferred transaction link data cluster; debugging the full-link automated test network according to the first quality evaluation factor and the second quality evaluation factor. Among them, the data difference represents the data difference situation between the transaction link data cluster indication information and the inferred transaction link data cluster. Since both the first quality evaluation factor and the second quality evaluation factor can reflect the accuracy of the full-link automated test network, debugging the full-link automated test network according to the first quality evaluation factor and the second quality evaluation factor improves the accuracy of the full-link automated test network.

[0190] In the embodiment of the present application, by combining a data cluster comparison network, the description knowledge error between the inference transaction link data cluster output by the full-link automated test network and the description knowledge of the transaction link data cluster indication information is used to debug the full-link automated test network, and the accuracy of the full-link automated test network is enhanced. In addition, by synthesizing the data differences between the inference transaction link data cluster and the transaction link data cluster indication information, and debugging the full-link automated test network by synthesizing the data differences and the description knowledge error, the reference elements are more comprehensive and the accuracy of the full-link automated test network is higher.

[0191] Based on the above embodiment, the embodiment of the present application provides a full-link automated test device Figure 3 which is a full-link automated test device 340 provided by the embodiment of the present application, as Figure 3 shown. The device 340 includes:

[0192] A report acquisition module 341, configured to acquire a full-link test report of financial transactions in response to a test instruction; wherein, the full-link test report of financial transactions includes a plurality of transaction link data clusters collected according to a preset period, and the plurality of transaction link data clusters include transaction link perturbation data clusters;

[0193] A knowledge acquisition module 342, configured to mine first description knowledge of multiple dimensions in each of the transaction link data clusters, and respectively integrate the multiple first description knowledge corresponding to the same transaction link data cluster to obtain second description knowledge;

[0194] A knowledge integration module 343, configured to perform eccentric calculation and integration on the second description knowledge of the plurality of transaction link data clusters according to the commonality measurement result between the second description knowledge of the plurality of transaction link data clusters and the second description knowledge of the transaction link perturbation data cluster, to obtain first integrated description knowledge corresponding to the transaction link perturbation data cluster;

[0195] An identification improvement module 344, configured to identify and improve the transaction link perturbation data cluster according to the first integrated description knowledge.

[0196] Wherein, the step of respectively integrating the multiple first description knowledge corresponding to the same transaction link data cluster to obtain second description knowledge includes: for the multiple first description knowledge corresponding to any one of the transaction link data clusters, converting the dimensions of the other first description knowledge except the first description knowledge of the target dimension to the target dimension; wherein, the target dimension is a preset dimension among the multiple dimensions; and integrating the converted multiple first description knowledge to obtain the second description knowledge of the transaction link data cluster.

[0197] The description of the above device embodiment is similar to the description of the above method embodiment, and has similar beneficial effects as the method embodiment. For technical details not disclosed in the device embodiment of the present application, please refer to the description of the method embodiment of the present application for understanding.

[0198] If the technical solution of this application involves personal or private information, the product using the technical solution of this application has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing personal information. If the technical solution of this application involves sensitive personal information, the product using the technical solution of this application has obtained the individual's separate consent before processing sensitive personal information, and at the same time meets the requirements of "explicit consent", and collects within the scope of laws and regulations. For example, on personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that he or she agrees to collect his or her personal information; or on the device that processes personal information, the personal information processing rules are notified by obvious signs / information, and the individual's authorization is obtained through pop-up information or by asking the individual to upload his or her personal information; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

[0199] It should be noted that in the embodiment of the present application, if the above-mentioned alarm processing method is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, which is stored in a storage medium, including a number of instructions to enable an electronic device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM), a disk or an optical disk. In this way, the embodiment of the present application is not limited to any specific combination of hardware and software.

[0200] An embodiment of the present application provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and the processor implements the above-mentioned alarm processing method when executing the computer program.

[0201] The embodiment of the present application provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned alarm processing method is implemented. The computer-readable storage medium can be transient or non-transient.

[0202] An embodiment of the present application provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program. When the computer program is read and executed by a computer, some or all of the steps in the above method are implemented. This computer program product can be specifically implemented in a manner of hardware, software, or a combination thereof. In an alternative embodiment, the computer program product is specifically embodied as a computer storage medium. In another alternative embodiment, the computer program product is specifically embodied as a software product, such as a Software Development Kit (SDK), etc.

[0203] It should be noted that Figure 4 is a schematic diagram of the hardware entity of a financial full-link automated test system provided by an embodiment of the present application. As Figure 4 shown, the hardware entity of the financial full-link automated test system 300 includes: a processor 310, a communication interface 320, and a memory 330, where: The processor 310 generally controls the overall operation of the financial full-link automated test system 300. The communication interface 320 can enable the electronic device to communicate with other terminals or servers through a network. The memory 330 is configured to store instructions and applications executable by the processor 310, and can also cache data to be processed or already processed by the processor 310 and each module in the financial full-link automated test system 300 (for example, image data, audio data, voice communication data, and video communication data), and can be implemented by flash memory (FLASH) or random access memory (Random Access Memory, RAM). Data transmission can be performed between the processor 310, the communication interface 320, and the memory 330 through a bus 340. It should be pointed out here that: The descriptions of the above storage medium and device embodiments are similar to the descriptions of the above method embodiments, and have beneficial effects similar to those of the method embodiments. For the technical details not disclosed in the storage medium and device embodiments of the present application, please refer to the descriptions of the method embodiments of the present application for understanding.

[0204] It should be understood that the term "one embodiment" or "an embodiment" mentioned throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of the present application. Therefore, the appearances of the phrase "in one embodiment" or "in an embodiment" throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in various embodiments of the present application, the order numbers of the above processes do not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application. The serial numbers of the embodiments of the present application above are only for description and do not represent the superiority or inferiority of the embodiments.

[0205] It should be noted that in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the existence of additional identical elements in the process, method, article or device comprising such element.

[0206] In several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the units is only a logical functional division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed with each other can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0207] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units; they can be located in one place or distributed to multiple network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0208] In addition, each functional unit in the embodiments of this application can be all integrated in a processing unit, or each unit can be separately a unit alone, or two or more units can be integrated in a unit; the above integrated units can be implemented in the form of hardware, or in the form of a combination of hardware and software functional units.

[0209] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above method embodiments; and the foregoing storage medium includes: various media such as removable storage devices, read-only memory (ROM), magnetic disks or optical discs that can store program codes.

[0210] Alternatively, if the above integrated units of the present application are implemented in the form of software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence or the part that contributes to the related technology, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing an electronic device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: various media that can store program codes, such as removable storage devices, ROMs, magnetic disks, or optical discs.

[0211] As described above, the above are only the implementation manners of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application.

Claims

1. A full-link automated testing method based on Internet finance, characterized in that, Applied to a financial full-link automated testing system, the method includes: Responding to a test instruction to obtain a financial transaction full-link test report; wherein, the financial transaction full-link test report includes a plurality of transaction link data clusters collected according to a preset period, and the plurality of transaction link data clusters include transaction link perturbation data clusters; Mining first description knowledge of multiple dimensions in each of the transaction link data clusters, and respectively integrating the multiple first description knowledge corresponding to the same transaction link data cluster to obtain second description knowledge; According to the commonality measurement result between the second description knowledge of the plurality of transaction link data clusters and the second description knowledge of the transaction link perturbation data cluster, performing eccentric calculation and integration on the second description knowledge of the plurality of transaction link data clusters to obtain the first integrated description knowledge corresponding to the transaction link perturbation data cluster; wherein, the commonality measurement result between the second description knowledge of any transaction link data cluster and the second description knowledge of the transaction link perturbation data cluster represents the correlation degree between the transaction link data cluster and the transaction link perturbation data cluster; eccentric calculation and integration means assigning different weights to the corresponding second description knowledge according to the correlation degree, and then integrating the multiple second description knowledge after the weight calculation; Identifying and improving the transaction link perturbation data cluster according to the first integrated description knowledge; Wherein, the integrating the multiple first description knowledge corresponding to the same transaction link data cluster to obtain second description knowledge includes: for the multiple first description knowledge corresponding to any one of the transaction link data clusters, converting the dimensions of the other first description knowledge except the first description knowledge of the target dimension to the target dimension; wherein, the target dimension is a preset dimension among the multiple dimensions; integrating the converted multiple first description knowledge to obtain the second description knowledge of the transaction link data cluster; The multiple dimensions include M, and M>2. Mining first description knowledge of multiple dimensions in each transaction link data cluster includes: For each transaction link data cluster among the plurality of transaction link data clusters, performing description knowledge mining on the transaction link data cluster to obtain the third description knowledge of the transaction link data cluster; Quantifying the third description knowledge to obtain the quantified description knowledge of the first dimension; Performing dimension embedding mapping and quantization on the quantified description knowledge of the first dimension to obtain the quantified description knowledge of the second dimension until the quantified description knowledge of the Nth dimension is obtained, wherein N = M - 1; Performing dimension embedding mapping on the quantified description knowledge of the Nth dimension to obtain the quantified description knowledge of the Mth dimension; Restoring the information of the quantified description knowledge of the Mth dimension to obtain the first description knowledge of the Mth dimension; Performing dimension expansion mapping on the first description knowledge of the Mth dimension to obtain the expanded description knowledge of the Nth dimension, and restoring the information of the expanded description knowledge of the Nth dimension and the quantified description knowledge of the Nth dimension to obtain the first description knowledge of the Nth dimension until the first description knowledge of the first dimension is obtained.

2. The method according to claim 1, wherein Integrating the converted multiple first description knowledges to obtain the second description knowledge of the transaction link data cluster includes: Performing knowledge splicing on the converted multiple first description knowledges to obtain spliced description knowledge; Performing a linear transformation operation on the spliced description knowledge to obtain the second description knowledge of the transaction link data cluster.

3. The method according to claim 1, characterized in that, The quantization description knowledge of the first dimension is a tensor; Performing dimension embedding mapping and quantization on the quantization description knowledge of the first dimension to obtain the quantization description knowledge of the second dimension includes: Performing dimension embedding mapping on the quantization description knowledge of the first dimension to obtain the embedding description knowledge of the second dimension; Decomposing the embedding description knowledge to obtain multiple first tensor blocks, and the first tensor block includes description knowledges on multiple coordinates; For each description knowledge, according to the multiple description knowledges in the first tensor block to which the description knowledge belongs and the coordinate description knowledges of the multiple description knowledges, optimizing and adjusting the description knowledge, and the coordinate description knowledge represents the coordinate of the corresponding description knowledge in the first tensor block; Establishing a second tensor block for the multiple description knowledges after optimizing and adjusting the same first tensor block; Splicing the multiple second tensor blocks according to the coordinates of the multiple first tensor blocks in the embedding description knowledge to obtain a first spliced tensor; Determining the quantization description knowledge of the second dimension according to the first spliced tensor.

4. The method according to claim 3, wherein For each description knowledge, according to the multiple description knowledges in the first tensor block to which the description knowledge belongs and the coordinate description knowledges of the multiple description knowledges, optimizing and adjusting the description knowledge includes: For each first tensor block, integrating the multiple description knowledges in the first tensor block with their corresponding coordinate description knowledges respectively to obtain multiple first integrated description knowledges; For each description knowledge in the first tensor block, according to the commonality measurement result between the first integrated description knowledge corresponding to the description knowledge and the multiple first integrated description knowledges, performing eccentric calculation and integration on the multiple first integrated description knowledges, and determining the description knowledge obtained after eccentric calculation and integration as the description knowledge after optimizing and adjusting the description knowledge; Determining the quantization description knowledge of the second dimension according to the first spliced tensor includes: determining the first spliced tensor as the quantization description knowledge of the second dimension; Before decomposing the embedding description knowledge to obtain multiple first tensor blocks, the method further includes: performing coordinate transformation on the description knowledge in the embedding description knowledge according to a first coordinate transformation tensor to obtain an optimized and adjusted embedding description knowledge; Determining the quantization description knowledge of the second dimension according to the first spliced tensor includes: Performing coordinate transformation on the description knowledge in the first spliced tensor according to a second coordinate transformation tensor, and determining the obtained description knowledge tensor after transformation as the quantization description knowledge of the second dimension, and the second coordinate transformation tensor is the inverse transformation tensor of the first coordinate transformation tensor.

5. The method according to claim 1, characterized in that, Restoring information from the extended description knowledge of the Nth dimension and the quantization description knowledge of the Nth dimension to obtain the first description knowledge of the Nth dimension, including: Integrating the extended description knowledge of the Nth dimension with the quantization description knowledge of the Nth dimension to obtain the integrated description knowledge of the Nth dimension; Performing an information restoration operation on the integrated description knowledge of the Nth dimension to obtain the first description knowledge of the Nth dimension.

6. The method according to any one of claims 1 to 5, characterized in that The full-link automated testing method based on Internet finance is executed according to the full-link automated testing network. The method further includes the debugging steps of the full-link automated testing network, including: Obtaining a sample of the full-link test report for financial transactions, where the sample of the full-link test report for financial transactions includes multiple transaction link data cluster samples collected according to a preset period. The multiple transaction link data cluster samples include transaction link perturbation data cluster samples, and obtaining the transaction link data cluster indication information corresponding to the transaction link perturbation data cluster samples; Processing the transaction link data cluster samples in the full-link test report sample for financial transactions according to the full-link automated testing network to obtain the inferred transaction link data cluster corresponding to the transaction link perturbation data cluster samples; According to the data cluster comparison network, respectively performing description knowledge mining on the transaction link data cluster indication information and the inferred transaction link data cluster to obtain the fourth description knowledge of the transaction link data cluster indication information and the fifth description knowledge of the inferred transaction link data cluster; According to the data cluster comparison network, obtaining the description knowledge error between the fourth description knowledge and the fifth description knowledge, and determining the description knowledge error as the first quality evaluation factor; Debugging the full-link automated testing network according to the first quality evaluation factor.

7. The method according to claim 6, wherein After identifying and improving the transaction link perturbation data cluster according to the first integrated description knowledge, the method further includes: Obtaining the full-link test report for financial transactions after identification and improvement and a report template library, where the report template library includes report templates corresponding to at least one transaction type; Performing description knowledge mining on the full-link test report for financial transactions after identification and improvement to obtain the full-link test report description knowledge for financial transactions, and performing description knowledge mining on the report template to obtain the report template description knowledge of the report template; Classifying the full-link test report for financial transactions after identification and improvement based on a preset method according to the full-link test report description knowledge for financial transactions and the report template description knowledge to obtain the classification results of the full-link test report for financial transactions after identification and improvement relative to each transaction type in each preset method; Determining the transaction type of the full-link test report for financial transactions after identification and improvement according to the classification results of the full-link test report for financial transactions after identification and improvement relative to each transaction type in each preset method.

8. A full-link automated testing method based on Internet finance, characterized in that, Applied to a financial full-link automated testing system, the method includes: In response to a test instruction, obtain a full-link test report for financial transactions; wherein, the full-link test report for financial transactions includes a plurality of transaction link data clusters collected according to a preset period, and the plurality of transaction link data clusters include transaction link perturbation data clusters. For each transaction link data cluster among the plurality of transaction link data clusters, perform descriptive knowledge mining on the transaction link data cluster to obtain third descriptive knowledge of the transaction link data cluster; quantify the third descriptive knowledge to obtain quantified descriptive knowledge of the first dimension; perform dimension embedding mapping on the quantified descriptive knowledge of the first dimension to obtain quantified descriptive knowledge of the second dimension; perform information restoration on the quantified descriptive knowledge of the second dimension to obtain first descriptive knowledge of the second dimension; perform dimension expansion mapping on the first descriptive knowledge of the second dimension to obtain extended descriptive knowledge of the first dimension, perform information restoration on the extended descriptive knowledge of the first dimension and the quantified descriptive knowledge of the first dimension to obtain first descriptive knowledge of the first dimension, and respectively integrate the multiple first descriptive knowledge corresponding to the same transaction link data cluster to obtain second descriptive knowledge. According to the commonality measurement result between the second descriptive knowledge of the plurality of transaction link data clusters and the second descriptive knowledge of the transaction link perturbation data cluster, perform eccentric calculation and integration on the second descriptive knowledge of the plurality of transaction link data clusters to obtain first integrated descriptive knowledge corresponding to the transaction link perturbation data cluster; wherein, the commonality measurement result between the second descriptive knowledge of any transaction link data cluster and the second descriptive knowledge of the transaction link perturbation data cluster represents the correlation degree between the transaction link data cluster and the transaction link perturbation data cluster; eccentric calculation and integration means assigning different weights to the corresponding second descriptive knowledge according to the correlation degree, and after the weight calculation, performing an integration operation on the multiple second descriptive knowledge. Identify and improve the transaction link perturbation data cluster according to the first integrated descriptive knowledge. Among them, the step of respectively integrating the multiple first descriptive knowledge corresponding to the same transaction link data cluster to obtain second descriptive knowledge includes: for the multiple first descriptive knowledge corresponding to any one of the transaction link data clusters, convert the dimensions of the other first descriptive knowledge except the first descriptive knowledge of the target dimension to the target dimension; wherein, the target dimension is a preset dimension among the multiple dimensions; integrate the converted multiple first descriptive knowledge to obtain second descriptive knowledge of the transaction link data cluster.

9. A financial full-link automated testing system, characterized in that, It includes a processor and a memory, and the memory stores a computer program, and when the computer program is executed by the processor, the method according to any one of claims 1 to 7 or claim 8 is implemented.

Images