A non-functional testing method and system for financial systems based on artificial intelligence

Abstract

The present invention relates to a non-functional testing method and system for a financial system based on artificial intelligence. The problem of analyzing the results of rapid implementation of non-functional testing is abstracted into an artificial intelligence image recognition problem, and an artificial intelligence image recognition algorithm is used to automatically judge and analyze the test results of rapid implementation of non-functional testing. Artificial intelligence graphic recognition is used to perform image model transfer learning training using historical rapid implementation test result data to generate a rapid implementation result judgment model. Subsequent rapid implementation results are judged by the model to intelligently judge the system with risk problems. The use of transfer learning can avoid the disadvantage of high hardware requirements for model training and improve the training speed; result judgment does not rely on experience, which improves the accuracy; only test results with risk problems prompted by the system require manual intervention by professional non-functional testers, which greatly reduces labor costs.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based financial system non-functional testing method and system. Background Art

[0002] With the development of Internet technology, financial software systems have also developed rapidly, and the speed of version iteration is also very fast. On average, the number of production versions of banking systems reaches hundreds every month. Production systems are divided into two categories, one is the newly launched system, and the other is the transformation of existing systems. For newly launched systems, detailed non-functional testing needs to be implemented, including single transaction benchmarks, load, capacity, stability, high availability, etc. For the transformation of existing systems, it is not necessary to conduct complete non-functional testing every time, but for important transaction systems, rapid implementation testing is important and necessary. The judgment of the results of rapid implementation (memory anomalies, response time anomalies, etc.) needs to rely on manual work, and requires extremely high professional background and experience knowledge. In order to improve the ability of rapid implementation and the correctness and efficiency of result analysis. Summary of the invention

[0003] In order to solve the deficiencies of the prior art, the present invention proposes a non-functional test rapid implementation analysis method and system based on artificial intelligence. The problem of non-functional test rapid implementation result analysis is abstracted into an artificial intelligence image recognition problem, and an artificial intelligence image recognition algorithm is used to automatically judge and analyze the test results of non-functional test rapid implementation. Artificial intelligence graphic recognition is used to perform image model transfer learning training with historical rapid implementation test result data to generate a rapid implementation result judgment model. Subsequent rapid implementation results are judged by this model, and a system with risk problems is intelligently judged.

[0004] To achieve the above objectives, the technical solutions adopted by the present invention include:

[0005] A non-functional test rapid implementation analysis method based on artificial intelligence, characterized by comprising:

[0006] S1. Obtain image data of historical non-functional tests;

[0007] S2, dividing the image data of the historical non-functional test into a training set and a verification set;

[0008] S3, classify and label the image data in the training set according to the classification labels;

[0009] S4, inputting the classified training set into the image classification model, performing iterative training, and obtaining a fast implementation result judgment model to be determined;

[0010] S5, inputting the verification set into the rapid implementation result judgment model to be determined, and evaluating the judged rapid implementation result;

[0011] S6. When the evaluation accuracy reaches the first threshold, stop training and obtain a fast implementation result judgment model;

[0012] S7, obtaining image data of non-functional testing of the financial system to be tested;

[0013] S8, inputting the image data of the non-functional test into the rapid implementation result judgment model;

[0014] S9. Output the rapid implementation judgment results of the financial system;

[0015] S10, saving the outputted rapid implementation judgment result into the optimized training data set;

[0016] S11. When the number of judgment results in the optimized training data set reaches a second threshold, the optimized training data set is divided into a new training set and a verification set, and S3 to S6 are repeatedly executed to obtain an optimized rapid implementation result judgment model.

[0017] Furthermore, the classification labels include primary classification labels and secondary classification labels.

[0018] The first-level classification labels include: rapid implementation result is normal, rapid implementation result is abnormal;

[0019] The secondary classification labels include: a CPU usage trend label, a memory usage trend label, a TPS trend label, and a response time trend label.

[0020] Furthermore, the image data of the non-functional test includes: a CPU usage trend graph, a memory usage trend graph, a TPS trend graph, and a response time trend graph.

[0021] Furthermore, the rapid implementation result judgment model includes: a CPU usage model, a memory usage model, a TPS model and a response time model.

[0022] Further, when the judgment results of the rapid implementation result judgment model are all passed, the rapid implementation judgment result of the financial system is passed;

[0023] When the judgment result of any one of the rapid implementation result judgment models is failure, the rapid implementation judgment result of the financial system is failure.

[0024] Furthermore, the first threshold is 95%.

[0025] Furthermore, the image classification model is an Inception V3 image classification model.

[0026] The present invention also relates to a financial system non-functional testing system based on artificial intelligence, characterized in that it includes:

[0027] A data acquisition module, used for acquiring image data of non-functional testing;

[0028] A classification module is used to classify and label the image data in the training set according to the classification labels;

[0029] A training module is used to input the classified training set into the image classification model, perform iterative training, and obtain a fast implementation result judgment model to be determined;

[0030] A verification module, used for inputting the verification set into the rapid implementation result judgment model to be determined, and evaluating the judged rapid implementation result;

[0031] A determination module, when the evaluation accuracy of the rapid implementation result judgment model to be determined reaches a first threshold, stops training and determines the rapid implementation result judgment model;

[0032] A detection module, inputting the image data of the non-functional test into a rapid implementation result judgment model to detect the result of the non-functional rapid implementation;

[0033] An output module, outputting the rapid implementation judgment result of the financial system;

[0034] The optimization training module divides the optimization training data set into a new training set and a validation set when the number of judgment results in the optimization training data set reaches a second threshold, and optimizes the training of the rapid implementation result judgment model;

[0035] The prompt module is used to prompt manual analysis after learning that the rapid implementation judgment result of the financial system is not passed.

[0036] The present invention also relates to a computer-readable storage medium, characterized in that a computer program is stored on the storage medium, and when the computer program is executed by a processor, the above-mentioned artificial intelligence-based non-functional testing method for a financial system is implemented.

[0037] The present invention also relates to an electronic device, characterized in that it comprises a processor and a memory;

[0038] The memory is used to store the acquired image data of the non-functional test;

[0039] The processor is used to call the above-mentioned artificial intelligence-based non-functional testing method for the financial system by calling the memory.

[0040] The present invention also relates to a computer program product, comprising a computer program and / or instructions, characterized in that when the computer program and / or instructions are executed by a processor, the above-mentioned artificial intelligence-based non-functional testing method for a financial system is implemented.

[0041] The beneficial effects of the present invention are:

[0042] The artificial intelligence-based non-functional testing method and system for a financial system of the present invention utilizes artificial intelligence graphic recognition, uses historical rapid implementation test result data to perform image model transfer learning training, generates a rapid implementation result judgment model, and uses the model to judge subsequent rapid implementation results. The system intelligently judges whether there are risk problems. The transfer learning method is adopted, and transfer learning is performed on the existing inceptionV3 model, which can avoid the disadvantage of high hardware requirements for model training and improve the training speed. The result judgment does not rely on experience, which improves the accuracy. Only when the system prompts that there are risk problems in the test results, professional non-functional testers need to intervene manually, which greatly reduces the labor cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 The present invention is a flowchart of a non-functional testing method for a financial system based on artificial intelligence.

[0044] Figure 2 The present invention is a schematic diagram of the structure of a non-functional testing system for a financial system based on artificial intelligence. DETAILED DESCRIPTION

[0045] In order to more clearly understand the content of the present invention, it will be described in detail with reference to the drawings and embodiments.

[0046] The present invention abstracts the problem of rapid implementation result analysis of non-functional tests into an artificial intelligence image recognition problem, and uses an artificial intelligence image recognition algorithm to automatically judge and analyze the rapid implementation test results of non-functional tests. By using artificial intelligence graphic recognition, image model transfer learning training is performed using historical rapid implementation test result data to generate a rapid implementation result judgment model, and subsequent rapid implementation results are judged by the model, so as to intelligently judge the system with risk problems.

[0047] The first aspect of the present invention relates to a process of steps as follows Figure 1 The method shown comprises:

[0048] S1. Obtain image data of historical non-functional tests;

[0049] Obtain the test results of non-functional rapid implementation that requires manual participation in historical tests. Non-functional testing refers to testing other than functional testing, such as performance capacity, availability, maintainability, and security testing. The test results of non-functional rapid implementation mainly include CPU usage trend charts, memory trend charts, TPS trend charts, response time trend charts, etc., as a set of candidate image inputs for artificial intelligence image recognition training.

[0050] S2, dividing the image data of the historical non-functional test into a training set and a verification set;

[0051] The training set is used to train the model, and the validation set is used to verify the accuracy of the model. When the model accuracy reaches a certain level, the training is stopped.

[0052] S3, classify and label the image data in the training set according to the classification labels;

[0053] Classification labels include primary classification labels and secondary classification labels.

[0054] The first-level classification labels include: a rapid implementation result of normal (a result of a system with a passed test result) and a rapid implementation result of abnormal (a result of a system with a failed test result);

[0055] The secondary classification labels include: CPU usage trend label, memory usage trend label, TPS trend label, response time trend label, and the image uniformly captures the result of the stability test for 24 hours.

[0056] S4, inputting the classified training set into the image classification model, performing iterative training, and obtaining a fast implementation result judgment model to be determined;

[0057] Preferably, the image classification model is an Inception V3 image classification model; the Inception V3 image classification model is an image classification model trained by Google on the large image database ImageNet, which can classify images of various categories.

[0058] TensorFlow is an open source, Python-based deep learning framework for high-performance numerical computing. With its flexible architecture, users can easily deploy computing work to a variety of platforms (CPU, GPU, TPU) and devices (desktop devices, server clusters, mobile devices, edge devices, etc.). Developed by Google, it has a wide range of applications in scenarios such as graph classification, audio processing, recommendation systems, and natural language processing.

[0059] In deep learning, transfer learning is the process of migrating the weights of each node in a layer-by-layer network from a trained network to a new network, rather than starting from scratch and training a neural network for each specific task. For example, training a graphical neural network requires a lot of resources, and transfer learning can reduce the consumption of resources for training, such as graphics cards, training time, and a large amount of training data, with the advantages of shorter training time, faster convergence speed, and more accurate weight parameters.

[0060] Through the TensorFlow open source framework and with the help of Google's open source inception V3 network for transfer learning, the classified images are used to train an image classification model for non-functional test result scenarios, which is applied to the prediction of non-functional rapid implementation results.

[0061] S5, inputting the verification set into the rapid implementation result judgment model to be determined, and evaluating the judged rapid implementation result;

[0062] The training set and test set each take 100 pictures each time, and each iteration is 10 times, and the test set is used to evaluate the results.

[0063] S6. When the evaluation accuracy reaches 95%, stop training and obtain the rapid implementation result judgment model;

[0064] S7, obtaining image data of non-functional testing of the financial system to be tested;

[0065] S8, inputting the image data of the non-functional test into a rapid implementation result judgment model;

[0066] Use the trained model to test the results of non-functional rapid implementation and obtain the test results of non-functional rapid implementation, including the CPU usage trend chart, memory trend chart, TPS trend chart, and response time trend chart during the test, as the input of the model built in the above steps, and judge whether the test results pass or not based on the output results of the model.

[0067] When the judgment results of the rapid implementation result judgment model are all passed, the rapid implementation judgment result of the financial system is passed;

[0068] When the judgment result of any one of the rapid implementation result judgment models is failure, the rapid implementation judgment result of the financial system is failure, prompting manual analysis;

[0069] S9, outputting the rapid implementation judgment result of the financial system;

[0070] In order to improve the accuracy of model prediction, test result images are regularly added to retrain the model, and the threshold of model judgment correctness can be gradually increased.

[0071] S10, saving the outputted rapid implementation judgment result into the optimized training data set;

[0072] S11. When the number of judgment results in the optimized training data set reaches a second threshold, the optimized training data set is divided into a new training set and a verification set, and S3 to S6 are repeatedly executed to obtain an optimized rapid implementation result judgment model.

[0073] The evaluation accuracy of the model after optimization training should be higher than that of the model without optimization training.

[0074] Another aspect of the present invention relates to a system, the structure of which is as follows: Figure 2 As shown, including:

[0075] A data acquisition module, used for acquiring image data of non-functional testing;

[0076] A classification module is used to classify and label the image data in the training set according to the classification labels;

[0077] A training module is used to input the classified training set into the image classification model, perform iterative training, and obtain a fast implementation result judgment model to be determined;

[0078] A verification module, used for inputting the verification set into the rapid implementation result judgment model to be determined, and evaluating the judged rapid implementation result;

[0079] A determination module, when the evaluation accuracy of the rapid implementation result judgment model to be determined reaches a first threshold, stops training and determines the rapid implementation result judgment model;

[0080] A detection module, inputting the image data of the non-functional test into a rapid implementation result judgment model to detect the result of the non-functional rapid implementation;

[0081] An output module, outputting the rapid implementation judgment result of the financial system;

[0082] The optimization training module divides the optimization training data set into a new training set and a verification set when the number of judgment results in the optimization training data set reaches a second threshold, optimizes the training of the rapid implementation result judgment model, and improves the accuracy of the model judgment.

[0083] The output results of the model are used to determine whether the test results have passed. If one of the models fails (such as CPU usage failure), the test result will be quickly implemented as failed. For the failed results, manual analysis will be performed. Otherwise, it is passed and no manual intervention is required.

[0084] The prompt module is used to prompt manual analysis after learning that the rapid implementation judgment result of the financial system is not passed.

[0085] By using this system, the above-mentioned operation processing method can be executed and the corresponding technical effects can be achieved.

[0086] An embodiment of the present invention also provides a computer-readable storage medium capable of implementing all the steps of the method in the above embodiment, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, all the steps of the artificial intelligence-based non-functional testing method for a financial system in the above embodiment are implemented.

[0087] An embodiment of the present invention also provides an electronic device for executing the above method. As an implementation device of the method, the electronic device has at least a processor and a memory, and in particular, the memory stores data required for executing the method and related input data, such as a CPU usage trend graph, a memory usage trend graph, a TPS trend graph, a response time trend graph, etc., and all steps of the implementation method are implemented by the processor calling the data and program execution in the memory, and obtaining the corresponding technical effect.

[0088] Preferably, the electronic device may include a bus architecture, the bus may include any number of interconnected buses and bridges, and the bus will include various circuits linked together by one or more processors and memories. The bus may also link together various other circuits such as peripherals, voltage regulators, and power management circuits, which are all well known in the art and are therefore not further described herein. The bus interface provides an interface between the bus and the receiver and transmitter. The receiver and transmitter may be the same component, i.e., a transceiver, providing a unit for communicating with various other systems over a transmission medium. The processor is responsible for managing the bus and general processing, while the memory may be used to store data used by the processor when performing operations.

[0089] In addition, the electronic device may further include components such as a communication module, an input unit, an audio processor, a display, and a power supply. The processor (or controller, operation control) used may include a microprocessor or other processor device and / or a logic device, which receives input and controls the operation of various components of the electronic device; the memory may be one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory, or other suitable devices, which may store the above-mentioned related data information, and may also store a program for executing related information, and the processor may execute the program stored in the memory to realize information storage or processing, etc.; the input unit is used to provide input to the processor, for example, it may be a key or a touch input device; the power supply is used to provide power to the electronic device; the display is used to display display objects such as images and text, for example, it may be an LCD display. The communication module is a transmitter / receiver that sends and receives signals via an antenna. The communication module (transmitter / receiver) is coupled to the processor to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal. Based on different communication technologies, multiple communication modules may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless local area network module, etc. The communication module (transmitter / receiver) is also coupled to a speaker and a microphone via an audio processor to provide an audio output via the speaker and receive an audio input from the microphone, thereby realizing a common telecommunication function. The audio processor may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor is also coupled to a central processor, so that recording can be performed on the machine through the microphone, and the sound stored on the machine can be played through the speaker.

[0090] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0091] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A system that specifies the functions of a box or boxes.

[0092] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction system, which is implemented in the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0093] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps of the functions specified in a box or multiple boxes. Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the attached claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0094] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A non-functional testing method for financial systems based on artificial intelligence, characterized in that: include: S1. Acquire image data of historical non-functional tests, wherein the image data of the non-functional tests includes: a CPU usage trend graph, a memory usage trend graph, a TPS trend graph, and a response time trend graph; S2, dividing the image data of the historical non-functional test into a training set and a verification set; S3. Classify and label the image data in the training set and the validation set according to the classification labels, wherein the classification labels include primary classification labels and secondary classification labels: The first-level classification labels include: rapid implementation result is normal, rapid implementation result is abnormal; The secondary classification labels include: CPU usage trend label, memory usage trend label, TPS trend label, response time trend label; S4, inputting the classified training set into the image classification model, performing iterative training, and obtaining a fast implementation result judgment model to be determined; S5, inputting the verification set into the rapid implementation result judgment model to be determined, and evaluating the judged rapid implementation result; S6. When the evaluation accuracy reaches the first threshold, stop training and obtain a fast implementation result judgment model; S7, obtaining image data of non-functional testing of the financial system to be tested; S8, inputting the image data of the non-functional test into the rapid implementation result judgment model; S9, outputting the rapid implementation judgment result of the financial system to be detected; S10, saving the outputted rapid implementation judgment result into the optimized training data set; S11. When the number of judgment results in the optimized training data set reaches a second threshold, the optimized training data set is divided into a new training set and a verification set, and S3 to S6 are repeatedly executed to obtain an optimized rapid implementation result judgment model.

2. The method according to claim 1, characterized in that The rapid implementation result judgment model includes: a CPU usage model, a memory usage model, a TPS model and a response time model.

3. The method according to claim 2, characterized in that When the judgment results of the rapid implementation result judgment model are all passed, the rapid implementation judgment result of the financial system is passed; When the judgment result of any one of the rapid implementation result judgment models is failure, the rapid implementation judgment result of the financial system is failure.

4. The method according to claim 1, characterized in that The first threshold is 95%.

5. The method according to claim 1, characterized in that The image classification model is the Inception V3 image classification model.

6. A non-functional testing system for financial systems based on artificial intelligence, characterized in that: include: A data acquisition module is used to acquire image data of non-functional tests, wherein the image data of non-functional tests includes: a CPU usage trend graph, a memory usage trend graph, a TPS trend graph, and a response time trend graph; The classification module is used to classify and label the image data in the training set according to the classification labels, which include primary classification labels and secondary classification labels: The first-level classification labels include: rapid implementation result is normal, rapid implementation result is abnormal; The secondary classification labels include: CPU usage trend label, memory usage trend label, TPS trend label, response time trend label; A training module is used to input the classified training set into the image classification model, perform iterative training, and obtain a fast implementation result judgment model to be determined; A verification module, used for inputting the verification set into the rapid implementation result judgment model to be determined, and evaluating the judged rapid implementation result; A determination module, when the evaluation accuracy of the rapid implementation result judgment model to be determined reaches a first threshold, stops training and determines the rapid implementation result judgment model; A detection module, inputting the image data of the non-functional test into a rapid implementation result judgment model to detect the result of the non-functional rapid implementation; An output module, outputting the rapid implementation judgment result of the financial system; The optimization training module divides the optimization training data set into a new training set and a verification set when the number of judgment results in the optimization training data set reaches a second threshold, and performs optimization training on the rapid implementation result judgment model; The prompt module is used to prompt manual analysis after learning that the rapid implementation judgment result of the financial system is not passed.

7. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, implements the artificial intelligence-based non-functional testing method for a financial system according to any one of claims 1 to 5.

8. An electronic device, characterized in that: Including processor and memory: The memory is used to store the acquired image data of the non-functional test; The processor is used to execute the artificial intelligence-based non-functional testing method for a financial system according to any one of claims 1 to 5 by calling the acquired non-functional testing image data.

9. A computer program product comprising a computer program and / or instructions, characterized in that: When the computer program and / or instruction is executed by a processor, the non-functional testing method for a financial system based on artificial intelligence described in any one of claims 1 to 5 is implemented.

Images