A method and system for generating quality evaluation information based on big data

Abstract

The present invention discloses a method and system for generating quality evaluation information based on big data, including collecting service data and test data of a preset product, and preprocessing the service data and the test data; performing validity screening on the service data to obtain valid data, and obtaining quality data through the valid data; the quality data includes a first quality and a second quality; the first quality represents the product quality score based on the service data; the second quality represents the service satisfaction of users; using the test data to correct the quality data to obtain corrected data, and constructing a quality evaluation information generation model according to the corrected data; optimizing the quality evaluation information generation model according to the evaluation error, inputting the data to be evaluated into the quality evaluation information generation model, and outputting a generation result. This method improves the accuracy of generating quality evaluation information, and at the same time has good interpretability, and can be directly used in a quality evaluation information generation system.

Description

Technical Field

[0001] The present invention relates to the field of quality evaluation, and particularly to a method and system for generating quality evaluation information based on big data. Background Art

[0002] In the current era of rapid development of information technology, the evaluation of product quality no longer solely relies on traditional detection and user feedback methods, but increasingly tends to utilize big data technology for comprehensive, objective, and efficient analysis. With the wide application of technologies such as cloud computing, the Internet of Things, and artificial intelligence, enterprises can collect a vast amount of product service data and test data, which provide unprecedented rich materials for product quality evaluation. However, how to effectively utilize this big data to generate accurate and reliable quality evaluation information has become an urgent problem to be solved.

[0003] Traditional quality evaluation methods often focus on the physical performance testing of products or the subjective perception surveys of users. Although these methods can to a certain extent reflect the quality status of products, they often have limitations. For example, physical performance testing may not comprehensively cover all scenarios during the actual use of products, and the subjective perceptions of users may be affected by various factors such as personal preferences and usage habits, resulting in a lack of objectivity and accuracy in the evaluation results.

[0004] Therefore, there is an urgent need to invent a new method for generating quality evaluation information to improve the accuracy and reliability of quality evaluation. Summary of the Invention

[0005] The objective of the present invention is to provide a method for generating quality evaluation information based on big data.

[0006] To achieve the above objective, the present invention is implemented according to the following technical solution:

[0007] The present invention includes the following steps:

[0008] Collect service data and test data of a preset product, and preprocess the service data and the test data;

[0009] Perform validity screening on the service data to obtain valid data, and obtain quality data through the valid data; the quality data includes a first quality and a second quality; the first quality represents the product quality score based on the service data; the second quality represents the service satisfaction of users;

[0010] Use the test data to correct the quality data to obtain corrected data, and construct a quality evaluation information generation model according to the corrected data;

[0011] Optimize the quality evaluation information generation model according to the evaluation error, input the data to be evaluated into the quality evaluation information generation model, and output the generation result.

[0012] Further, the method for performing validity screening on the service data to obtain valid data includes:

[0013] Compare the product problems feedback by users with the service data, and calculate the relevance between the product problems and the service data:

[0014] ,

[0015] where the s-th product problem is , the j-th service data is , the relevance between the product problem and the service data is , the control factor is , the optimization coefficient is , the reference value of the service data is , the reference value of the product problem is , the number of service data is , the number of product problems is , the probability of the service data appearing when the product problem appears is ;

[0016] Output the service data with a relevance greater than 0.3719 as valid data.

[0017] Further, the method for obtaining the first quality includes:

[0018] Extract the quality-related data from the valid data, and calculate the product performance relevance of the quality-related data;

[0019] Sort the quality-related data according to time to obtain time-series quality data, and calculate the product quality score according to the time-series quality data:

[0020] ,

[0021] where the product quality score of the i-th batch is , the number of customers is Q, the adjustable coefficient is , the evaluation score of the z-th valid data of the v-th customer is , the product performance relevance of the z-th valid data of the v-th customer is , the z-th valid data of the v-th customer is , the standard value of the valid data is , the number of valid data is , the error coefficient is , valid data and the standard value of the variance is , the product repair ratio of the i-th batch is , the time weight is ;

[0022] Output the product quality score as the first quality.

[0023] Further, the method for obtaining the second quality includes:

[0024] Extract the user feedback data from the valid data, sort the user feedback data according to time, and obtain the time-series feedback data;

[0025] Based on the user feedback data in the valid data, obtain the emotion score using an emotion dictionary, and obtain the user evaluation score using the emotion score and the excited feedback frequency;

[0026] Calculate the user service satisfaction according to the time-series feedback data:

[0027] ,

[0028] ,

[0029] where the first parameter is , the second parameter is , the emotion score of the z-th valid data of the v-th customer is , the excited feedback frequency of the z-th valid data of the v-th customer is , the offset coefficient is , the service satisfaction of the i-th batch of products is , the z-th valid data of the v-th customer is , the standard value of the valid data is , the number of customers is Q, and the number of valid data is ;

[0030] Output the service satisfaction as the second quality.

[0031] Further, the method for correcting the quality data using the test data to obtain the corrected data includes:

[0032] Compare the product data and the test data in the service data, and calculate the deviation degree between the product data and the test data in the service data:

[0033] ,

[0034] Among them, the c-th product data is , the c-th test data is , the kernel density estimation function is , the number of test data is N, and the square of the norm is , the error factor is , and the control coefficient is ;

[0035] Classify according to the similarity of test data using a decision tree, and calculate the information content of the test data:

[0036] ,

[0037] Among them, the information content of the x-th classification is , the number of test data in the x-th classification is , the c-th test data in the x-th classification is , the average value of the test data in the x-th classification is , and the probability of the test data appearing is ;

[0038] Correct the quality data through the deviation degree and information content, and the expression is:

[0039] ,

[0040] Among them, the c-th quality data is , and the c-th corrected data is .

[0041] Furthermore, the method for constructing a quality evaluation information generation model based on the corrected data includes:

[0042] Construct the objective function of the quality evaluation information generation model according to the corrected data, and the expression is:

[0043] ,

[0044] Among them, the loss function is , the objective function of the i-th batch is , the quality weight is , the satisfaction weight is , the service satisfaction of the i-th batch of products is , and the product quality score of the i-th batch is ;

[0045] The quality evaluation information generation model includes a random forest algorithm, a feature extraction algorithm, and a deep learning algorithm;

[0046] The random forest algorithm divides the input data into training data and test data according to 7:2;

[0047] The feature extraction algorithm extracts the features of the training data through the encoder part of the neural network to obtain service features;

[0048] The deep learning algorithm takes the objective function as the learning object, uses the self-attention mechanism to capture global dependencies, and generates quality evaluation information according to the global dependencies;

[0049] Input the test data into the quality evaluation information generation model to obtain the predicted quality evaluation information, and adjust the regularization parameter of the quality evaluation information generation model according to the variance between the predicted quality evaluation information and the actual quality evaluation information.

[0050] Furthermore, the method for optimizing the quality evaluation information generation model according to the evaluation error includes:

[0051] Introduce a particle population, take the variance of the evaluation error as the fitness error, and perform a chaotic mapping on the particle population. The expression is:

[0052] ,

[0053] where the chaotic random number from 0 to 1 is b, and the (w + 1)-th dimension is , and the w-th dimension is ;

[0054] Calculate the fitness of the particles, take the position of the particle with the maximum fitness as the optimal position, and update the position of the particles. The expression is:

[0055] ,

[0056] where the random number from 1 to 2 is , the updated position of the a-th particle in the w-th dimension at the (t + 1)-th iteration is , the initial position of the a-th particle in the w-th dimension is , the random number from 0 to 1 is , the Cauchy operator is , and the optimal position at the t-th iteration is ;

[0057] Introduce a perturbation factor, update the particle position to obtain the perturbed position. The expression is:

[0058] ,

[0059] ,

[0060] where the perturbed position of the a-th particle in the w-th dimension at the (t + 1)-th iteration is , and the updated position of the a-th particle in the w-th dimension at the t-th iteration is , the random numbers from 0 to 1 are respectively , , , the perturbation factor at the t-th iteration is , the maximum number of iterations is , the current number of iterations is t;

[0061] The corrected position is obtained by updating the position with the standard deviation, and the expression is:

[0062] ,

[0063] where the standard deviation of the normally distributed random number is , the perturbed position of the a-th particle in the w-th dimension at the t-th iteration is , the corrected position of the a-th particle in the w-th dimension at the (t + 1)-th iteration is , the lower bound of the w-th dimension is , the upper bound of the w-th dimension is ;

[0064] Iterate continuously until the maximum number of iterations is reached, then stop the iteration, otherwise update the perturbation factor.

[0065] In a second aspect, a quality evaluation information generation system based on big data includes:

[0066] Data acquisition module: used to collect service data and test data of a preset product, and preprocess the service data and the test data;

[0067] Quality screening module: used to perform validity screening on the service data to obtain valid data, and obtain quality data through the valid data; the quality data includes first quality and second quality; the first quality represents the product quality score based on the service data; the second quality represents the service satisfaction of users;

[0068] Correction and construction module: used to correct the quality data with the test data to obtain corrected data, and construct a quality evaluation information generation model according to the corrected data;

[0069] Optimization and generation module: used to optimize the quality evaluation information generation model according to the evaluation error, input the data to be evaluated into the quality evaluation information generation model, and output the generation result.

[0070] In a third aspect, an embodiment of the present application further provides an electronic device, including:

[0071] A processor; and a memory arranged to store computer-executable instructions, and the executable instructions, when executed, cause the processor to execute the method steps described in the first aspect.

[0072] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium storing one or more programs, which when executed by an electronic device including a plurality of application programs, cause the electronic device to execute the method steps described in the first aspect.

[0073] The beneficial effects of the present invention are as follows:

[0074] The present invention is a method and system for generating quality evaluation information based on big data. Compared with the prior art, the present invention has the following technical effects:

[0075] Through steps of preprocessing, validity screening, obtaining quality data, correcting data, model construction, and model optimization, the present invention can improve the accuracy of generating quality evaluation information, thereby improving the precision of generating quality evaluation information. Optimizing the generation of quality evaluation information can greatly save resources and improve work efficiency. It can realize the automatic generation of quality evaluation information for product services, perform data correction and multi-data fusion on the generation of product service quality evaluation information in real time, which is of great significance for the generation of product service quality evaluation information, and can adapt to different standards of quality evaluation information generation and different quality evaluation information generation requirements, having a certain universality. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 It is a flowchart of the steps of a method for generating quality evaluation information based on big data according to the present invention;

[0077] Figure 2 It is a schematic structural diagram of an electronic device in an embodiment of this specification. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0078] The present invention will be further described below through specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention, but do not limit the present invention.

[0079] A method and system for generating quality evaluation information based on big data according to the present invention include the following steps:

[0080] As Figure 1 shown, in this embodiment, the following steps are included:

[0081] Collect service data and test data of a preset product, and preprocess the service data and the test data;

[0082] In actual evaluation, service data includes transaction data, user data, product data, market data, user research data, third-party data, service request data, service result data, technical data, and feedback data;

[0083] Take the XXX product as the research object, and collect service data and test data in the autumn of 2024;

[0084] Perform validity screening on the service data to obtain valid data, and obtain quality data through the valid data; the quality data includes the first quality and the second quality; the first quality represents the product quality score based on the service data; the second quality represents the service satisfaction of users;

[0085] In actual evaluation, the valid data includes user data, product data, market data, user research data, third-party data, service request data, service result data, and feedback data;

[0086] The first quality is 0.872, and the second quality is 0.638;

[0087] Use the test data to correct the quality data to obtain corrected data, and construct a quality evaluation information generation model according to the corrected data;

[0088] In actual evaluation, the corrected data of the first quality and the second quality are 0.859 and 0.651 respectively;

[0089] Optimize the quality evaluation information generation model according to the evaluation error, input the data to be evaluated into the quality evaluation information generation model, and output the generation result.

[0090] In this embodiment, the method for performing validity screening on the service data to obtain valid data includes:

[0091] Compare the product problems feedback by users with the service data, and calculate the correlation between the product problems and the service data:

[0092] ,

[0093] where the s-th product problem is , the j-th service data is , the product problem and the service data The correlation is , the control factor is , the optimization coefficient is , the reference value of the service data is , the reference value of the product problem is , the number of service data is , the number of product problems is , the probability of the service data appearing when the product problem appears is ;

[0094] Output service data with a relevance greater than 0.3719 as valid data.

[0095] In this embodiment, the method for obtaining the first quality includes:

[0096] Extract quality-related data from the valid data and calculate the product performance correlation degree of the quality-related data;

[0097] Sort the quality-related data according to time to obtain time-series quality data, and calculate the product quality score according to the time-series quality data:

[0098] ,

[0099] where the product quality score of the i-th batch is , the number of customers is Q, the adjustable coefficient is , the evaluation score of the z-th valid data of the v-th customer is , the product performance correlation degree of the z-th valid data of the v-th customer is , the z-th valid data of the v-th customer is , the standard value of the valid data is , the number of valid data is , the error coefficient is , the valid data and the standard value The variance of is , the product repair ratio of the i-th batch is , the time weight is ;

[0100] Output the product quality score as the first quality.

[0101] In this embodiment, the method for obtaining the second quality includes:

[0102] Extract user feedback data from the valid data, sort the user feedback data according to time to obtain time-series feedback data;

[0103] Based on the user feedback data in the valid data, obtain an emotion score using an emotion dictionary, and obtain the evaluation score of the user using the emotion score and the excitement feedback frequency;

[0104] Calculate the service satisfaction of the user according to the time-series feedback data:

[0105] ,

[0106] ,

[0107] where the first parameter is , the second parameter is , the sentiment score of the z-th valid data of the v-th customer is , the excitement feedback frequency of the z-th valid data of the v-th customer is , the offset coefficient is , the service satisfaction of the i-th batch of products is , the z-th valid data of the v-th customer is , the standard value of the valid data is , the number of customers is Q, and the number of valid data is ;

[0108] Output the service satisfaction as the second quality.

[0109] In this embodiment, the method for correcting the quality data by using the test data to obtain the corrected data includes:

[0110] Compare the product data and the test data in the service data, and calculate the deviation degree between the product data and the test data in the service data:

[0111] ,

[0112] where the c-th product data is , the c-th test data is , the kernel density estimation function is , the number of test data is N, and the square of the norm is , the error factor is , the control coefficient is ;

[0113] Classify according to the similarity of the test data using a decision tree, and calculate the information content of the test data:

[0114] ,

[0115] where the information content of the x-th classification is , the number of test data in the x-th classification is , the c-th test data in the x-th classification is , the average value of the test data in the x-th classification is , the test data appears with a probability of ;

[0116] Correct the quality data through the deviation degree and the information content, and the expression is:

[0117] ,

[0118] where the c-th quality data is , the c-th correction data is .

[0119] In this embodiment, the method for constructing a quality evaluation information generation model according to the correction data includes:

[0120] Construct an objective function of the quality evaluation information generation model according to the correction data, and the expression is:

[0121] ,

[0122] where the loss function is , the objective function of the i-th batch is , the quality weight is , the satisfaction weight is , the service satisfaction of the i-th batch of products is , the product quality score of the i-th batch is ;

[0123] The quality evaluation information generation model includes a random forest algorithm, a feature extraction algorithm, and a deep learning algorithm;

[0124] The random forest algorithm divides the input data into training data and test data according to 7:2;

[0125] The feature extraction algorithm extracts the features of the training data through the encoder part of the neural network to obtain service features;

[0126] The deep learning algorithm takes the objective function as the learning object, uses the self-attention mechanism to capture global dependencies, and generates quality evaluation information according to the global dependencies;

[0127] Input the test data into the quality evaluation information generation model to obtain the predicted quality evaluation information, and adjust the regularization parameter of the quality evaluation information generation model according to the variance between the predicted quality evaluation information and the actual quality evaluation information.

[0128] In this embodiment, the method for optimizing the quality evaluation information generation model according to the evaluation error includes:

[0129] Introduce a particle population, take the variance of the evaluation error as the fitness error, and perform a chaotic mapping on the particle population. The expression is:

[0130] ,

[0131] where the chaotic random number from 0 to 1 is b, the (w + 1)-th dimension is , the w-th dimension is ;

[0132] Calculate the fitness of the particles, take the position of the particle with the maximum fitness as the optimal position, and update the position of the particle. The expression is:

[0133] ,

[0134] where the random number from 1 to 2 is , the updated position of the ath particle in the wth dimension at the (t + 1)th iteration is , the initial position of the ath particle in the wth dimension is , the random number from 0 to 1 is , the Cauchy operator is , the optimal position at the tth iteration is ;

[0135] Introduce a perturbation factor to update the particle position to obtain the perturbed position. The expression is:

[0136] ,

[0137] ,

[0138] where the perturbed position of the ath particle in the wth dimension at the (t + 1)th iteration is , the updated position of the ath particle in the wth dimension at the tth iteration is , the random numbers from 0 to 1 are respectively 、 、 , the perturbation factor at the tth iteration is , the maximum number of iterations is , and the current number of iterations is t;

[0139] Use the position updated by the standard deviation to obtain the corrected position. The expression is:

[0140] ,

[0141] where the standard deviation of the normal distribution random number is , the perturbed position of the ath particle in the wth dimension at the tth iteration is , the corrected position of the ath particle in the wth dimension at the (t + 1)th iteration is , the lower bound of the wth dimension is , and the upper bound of the wth dimension is ;

[0142] Iterate continuously until the maximum number of iterations is reached, then stop the iteration; otherwise, update the perturbation factor.

[0143] In the second aspect, a quality evaluation information generation system based on big data includes:

[0144] Data acquisition module: used to acquire service data and test data of a preset product, and preprocess the service data and the test data;

[0145] Screening quality module: used to screen the service data for effectiveness to obtain valid data, and obtain quality data through the valid data; the quality data includes a first quality and a second quality; the first quality represents the product quality score based on the service data; the second quality represents the user's service satisfaction;

[0146] Correction and construction module: used to correct the quality data with the test data to obtain corrected data, and construct a quality evaluation information generation model according to the corrected data;

[0147] Optimization and generation module: used to optimize the quality evaluation information generation model according to the evaluation error, input the data to be evaluated into the quality evaluation information generation model, and output the generation result.

[0148] Figure 2 It is a schematic structural diagram of an electronic device according to an embodiment of the present application. Please refer to Figure 2 , at the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. Among them, the memory may include a memory, such as a high-speed random access memory (Random-Access Memory, RAM), and may also include a non-volatile memory (non-volatile memory), such as at least 1 disk memory, etc. Of course, the electronic device may also include other hardware required for other services.

[0149] The processor, network interface, and memory can be interconnected through an internal bus, and the internal bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, Figure 2 only a bidirectional arrow is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0150] Memory, used to store programs. Specifically, the program may include program code, and the program code includes computer operation instructions. The memory may include a memory and a non-volatile memory, and provide instructions and data to the processor.

[0151] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a quality evaluation information generation device based on big data at the logical level. The processor executes the program stored in the memory and is specifically used to execute any one of the aforementioned quality evaluation information generation methods based on big data.

[0152] As described in the present application Figure 1 A quality evaluation information generation method based on big data disclosed in the embodiments shown above can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by the integrated logic circuit in the hardware of the processor or by instructions in software form. The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as being executed by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. This storage medium is located in the memory, and the processor reads the information in the memory and combines its hardware to complete the steps of the above method.

[0153] The electronic device can also execute Figure 1 a quality evaluation information generation method based on big data in Figure 1 and implement the functions of the embodiments shown. The embodiments of the present application will not be elaborated here.

[0154] The embodiments of the present application also propose a computer-readable storage medium. The computer-readable storage medium stores one or more programs, and the one or more programs include instructions. When the instructions are executed by an electronic device including multiple application programs, they execute any one of the aforementioned quality evaluation information generation methods based on big data.

[0155] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0156] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0157] These computer program instructions can 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, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that realizes the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0158] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0159] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0160] The memory may include non-permanent memory in the computer-readable medium, random access memory (RAM), and / or non-volatile memory in the form of, for example, read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0161] A computer-readable medium includes permanent and non-permanent, removable and non-removable media that can implement information storage by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transitory medium that can be used to store information accessible by a computing device. As defined herein, a computer-readable medium does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0162] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0163] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0164] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for generating quality evaluation information based on big data, characterized in that: The following steps are involved: Collecting service data and test data of a preset product, and preprocessing the service data and the test data; Performing validity screening on the service data to obtain valid data, and obtaining quality data through the valid data; the quality data includes a first quality and a second quality; the first quality represents a product quality score based on the service data; The second quality represents the user's service satisfaction; The test data is used to correct the quality data to obtain corrected data, and a quality evaluation information generation model is constructed according to the corrected data; including: Compare the product data and test data in the service data, and calculate the deviation between the product data and test data in the service data: ， The cth product data is , the cth test data is , the kernel density estimation function is , the number of test data is N, and the square of the norm is , the error factor is , the control coefficient is ; According to the similarity of the test data, a decision tree is used to classify and calculate the amount of information of the test data: ， The amount of information for the xth category is , the number of test data for the xth category is , the cth test data of the xth category is , the average value of the test data for the xth category is , test data The probability of occurrence is ; The quality data is corrected by the deviation and information content, and the expression is: ， The cth quality data is , the cth correction data is ; Optimizing the quality evaluation information generation model according to the evaluation error, inputting the data to be evaluated into the quality evaluation information generation model, and outputting the generation result; comprising: Introducing particle population, taking the variance of evaluation error as fitness error, and performing chaotic mapping on particle population, the expression is: ， The chaotic random number from 0 to 1 is b, and the w+1th dimension is , the wth dimension is ; Calculate the fitness of the particle, take the position of the particle with the largest fitness as the optimal position, and update the position of the particle. The expression is: ， The random numbers from 1 to 2 are , the updated position of the wth dimension of the ath particle in the t+1th iteration is , the initial position of the ath particle in the wth dimension is , a random number from 0 to 1 is , the Cauchy operator is , the optimal position of the tth iteration is ; Introduce the disturbance factor and update the particle position to obtain the disturbance position. The expression is: ， ， The perturbed position of the wth dimension of the ath particle in the t+1th iteration is , the updated position of the wth dimension of the ath particle in the tth iteration is , the random numbers from 0 to 1 are , , , the perturbation factor for the tth iteration is , the maximum number of iterations is , the current number of iterations is t; The corrected position is obtained by using the position updated by the standard deviation, and the expression is: ， The normally distributed random numbers The standard deviation of , the perturbed position of the ath particle in the wth dimension at the tth iteration is , the corrected position of the wth dimension of the ath particle at the t+1th iteration is , the lower bound of the w-th dimension is , the upper bound of the w-th dimension is ; Continue to iterate until the maximum number of iterations is reached, then stop iterating, otherwise update the perturbation factor.

2. According to the method for generating quality evaluation information based on big data as described in claim 1, it is characterized in that: The method for screening the service data for validity to obtain valid data comprises: Compare the product problems and service data reported by users to calculate the correlation between product problems and service data: ， The sth product problem is , the jth service data is , product issues and service data The correlation is , the control factor is , the optimization coefficient is , service data The reference value is , product issues The reference value is , the amount of service data is , the number of product issues is , product problems Service data appears in the case of The probability of ; The service data with a correlation greater than 0.3719 is output as valid data.

3. According to the method for generating quality evaluation information based on big data as claimed in claim 1, it is characterized in that: The method for obtaining the first quality includes: Extract quality-related data from valid data and calculate product performance correlation of quality-related data; Sort the quality-related data by time to obtain the time series quality data, and calculate the product quality score based on the time series quality data: ， The product quality score of the i-th batch is , the number of customers is Q, and the adjustable coefficient is , the evaluation score of the vth customer’s zth valid data is , the product performance correlation of the zth valid data of the vth customer is , the zth valid data of the vth customer is The standard value of valid data is , the number of valid data is , the error coefficient is , valid data and standard value The variance of , the maintenance ratio of the i-th batch of products is , the time weight is ; Output the product quality score as the first quality.

4. The method for generating quality evaluation information based on big data according to claim 1, characterized in that: The method for obtaining the second quality comprises: Extract user feedback data from valid data, sort the user feedback data according to time, and obtain time series feedback data; Based on the user feedback data in the valid data, the sentiment dictionary is used to obtain the sentiment score, and the sentiment score and the excited feedback frequency are used to obtain the user's evaluation score; Calculate user service satisfaction based on time series feedback data: ， ， The first parameter is , the second parameter is , the sentiment score of the vth customer’s zth valid data is , the frequency of excited feedback of the zth valid data of the vth customer is , the offset coefficient is , the service satisfaction of the i-th batch of products is , the zth valid data of the vth customer is The standard value of valid data is , the number of customers is Q, and the number of valid data is ; Service satisfaction is output as the second quality.

5. The method for generating quality evaluation information based on big data according to claim 1, characterized in that: The method for constructing a quality evaluation information generation model according to the correction data comprises: The objective function of the quality evaluation information generation model is constructed based on the corrected data. The expression is: ， The loss function is , the objective function of the i-th batch is , the quality weight is , the satisfaction weight is , the service satisfaction of the i-th batch of products is , the product quality score of the i-th batch is ; The quality evaluation information generation model includes random forest algorithm, feature extraction algorithm, and deep learning algorithm; The random forest algorithm divides the input data into training data and test data in a ratio of 7:2; The feature extraction algorithm extracts the features of the training data through the encoder part of the neural network to obtain the service features; The deep learning algorithm takes the target function as the learning object, uses the self-attention mechanism to capture the global dependency, and generates quality evaluation information based on the global dependency; The test data is input into the quality evaluation information generation model to obtain the predicted quality evaluation information, and the regularization parameter of the quality evaluation information generation model is adjusted according to the variance of the predicted quality evaluation information and the actual quality evaluation information.

6. A quality evaluation information generation system based on big data, used to execute the method according to any one of claims 1 to 5, characterized in that: include: Data collection module: used to collect service data and test data of preset products, and pre-process the service data and the test data; A screening quality module: used for screening the service data for validity to obtain valid data, and obtaining quality data through the valid data; the quality data includes a first quality and a second quality; the first quality represents a product quality score based on the service data; The second quality represents the user's service satisfaction; Correction construction module: used to correct the quality data using the test data to obtain corrected data, and to construct a quality evaluation information generation model according to the corrected data; Optimization generation module: used to optimize the quality evaluation information generation model according to the evaluation error, input the data to be evaluated into the quality evaluation information generation model, and output the generation result.

7. An electronic device comprising: processor; as well as A memory arranged to store computer executable instructions, which, when executed, cause the processor to perform the method according to any one of claims 1 to 5.

8. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, enables the electronic device to execute the method according to any one of claims 1 to 5.

Images