Productive financial service data synthesis method and system based on large model reasoning

Abstract

The present invention relates to the field of data processing technology, and specifically discloses a method and system for synthesizing productive financial service data based on large model reasoning. The method comprises: data set collection, transformation model extraction, product quality synthesizable set matching, transformation model synthesis benefit value analysis and data synthesis service optimization. First, different state data sets of each product in the production process are obtained, and the transformation model is extracted to provide model support for subsequent synthesis operations; the quality compliance index of each product is evaluated, and the quality compliance synthesizable set of each product is matched. Synthesis operations are performed based on the transformation model, which can process and synthesize data more quickly, improve the synthesis speed of the overall quality compliance synthesizable set, analyze the synthesis benefit value of the transformation model, thereby realizing the synthesis process of productive product quality data, thereby providing fast and accurate data synthesis services to meet customer needs.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method and system for synthesizing productive financial service data based on large model reasoning. Background Art

[0002] Currently, in order to promote productivity growth and economic development, large-scale data synthesis operations are required when testing product quality, among which data set synthesis services are one of the important parts. Production organizations need to make full use of data set information to evaluate product quality, formulate quality inspection policies, and optimize quality risk management. By comparing the quality status of products, they can synthesize the degree of adaptability of users and their corresponding products to provide better, more efficient and secure financial services. This may require complex algorithms and a large amount of computing resources. Traditional data synthesis technology cannot efficiently synthesize quality data; at the same time, data synthesis technology relies on specific generation models, so that the quality and diversity of synthesized data will be limited.

[0003] For example, the invention patent with announcement number CN108376069B discloses a data synthesis method and device thereof, which are operated according to the following steps: solidifying the audio data into a first firmware and recompiling the first firmware; debugging the electroacoustic parameters and converting them to generate a data file containing the electroacoustic parameters; synthesizing the data file and the first firmware and generating a second firmware; the data synthesis method and device thereof in the above application synthesize the first firmware generated by the compilation tool and the data file generated by the conversion tool through a data synthesis tool, and generate a second firmware; the data synthesis method and device thereof in the above application can save human resources, reduce the trouble of recompiling firmware, improve work efficiency, and improve accuracy.

[0004] For example, the invention patent with announcement number CN111625523B, the above application embodiment discloses a method, device and equipment for data synthesis, the method includes: obtaining original data; dividing the original data into key value data and residual data; generating new key values ​​according to the key value data; processing the residual data to obtain discrete data and continuous data; performing chi-square distribution processing on the discrete data to obtain new discrete data; performing factor processing on the continuous data to obtain character data and merged digital data; generating new continuous character data according to the character data; generating new continuous digital data according to the digital data Copula process and formula data processing, and processing the digital data obtained by the Copula process into time data; synthesizing the new key values, new discrete data, new continuous character data, new continuous digital data and time data; the above application can batch generate synthetic data with the same set characteristics as the original data to meet a large number of training set problems required for machine learning and other research needs.

[0005] However, in the process of implementing the above-mentioned application embodiments, the present application discovered that the above-mentioned technology has at least the following technical problems: currently in the process of synthesizing data, the data dimension that needs to be synthesized is relatively single, which cannot improve the work efficiency and accuracy of data synthesis, and is not conducive to improving the reliability of the synthesis service; and after the data is synthesized, the synthesis operation is directly terminated, but the accuracy of the synthesized data remains to be investigated, so that the synthesis result will have a certain deviation, which is not conducive to improving the accuracy of the synthesis service. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides a method and system for synthesizing productive financial service data based on large model reasoning, which can effectively solve the problems involved in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: The first aspect of the present invention provides a method for synthesizing productive financial service data based on large model reasoning, including: data set collection: collecting the operating data of the production equipment of each product in the production process, the supply chain information of the production line, and the weight and volume of each synthesized product, which are recorded as the first data set, the second data set, and the third data set of each product respectively; transformation model extraction: constructing the first data set, the second data set, and the third data set of each product into the first data set model, the second data set model, and the third data set model respectively, and obtaining the historical data of each product as the input of each data set model, so as to obtain each data set model. The output results of the data set model are collected, and the data set model with the highest data synthesis adaptability is screened out and recorded as the transformation model; product quality synthesizable set matching: obtain the data set information of each product, evaluate the quality compliance index of each product, and match the quality compliance synthesizable set of each product; transformation model synthesis benefit value analysis: based on the transformation model, the quality compliance synthesizable set of each product is synthesized, and the synthesis benefit value of the transformation model is analyzed; data synthesis service optimization: according to the synthesis benefit value of the transformation model, it is compared with the set synthesis benefit threshold. If the synthesis benefit value of the transformation model is lower than the synthesis benefit threshold, the data synthesis optimization plan is matched, thereby realizing the data synthesis process of productive product quality.

[0008] As a further method, the data synthesis process for realizing the quality of productive products is specifically analyzed as follows: according to the synthetic benefit value of the transformation model, it is compared with the set synthetic benefit threshold; if the synthetic benefit value of the transformation model is lower than the synthetic benefit threshold, the synthetic benefit value of the transformation model is differenced with the synthetic benefit threshold to obtain the synthetic benefit deviation value of the transformation model, and the difference is matched with the data synthesis optimization plan corresponding to each synthetic benefit deviation value interval to obtain the data synthesis optimization plan of the transformation model, thereby realizing the data synthesis process for the quality of productive products.

[0009] As a further method, the synthetic benefit value of the transformation model has a specific analysis process: by synthesizing the synthesizable set of quality standards of each product, the synthetic information of the transformation model is obtained, specifically including the synthetic completion degree and synthetic matching degree of the transformation model, and the synthetic benefit value of the transformation model is obtained by comprehensive analysis.

[0010] As a further method, the quality standard synthesizable set of each product is synthesized, and the specific synthesis process is: the quality standard synthesizable set of each product is input into the transformation model, and the quality standard synthesizable set of each product is synthesized by the decoding output function of the transformation model. The output is the synthetic data of the transformation model, and the Bayesian optimization method is used to screen the synthetic data of the transformation model according to the synthesis specification indicators, thereby realizing the synthesis operation of the quality standard synthesizable set of each product.

[0011] As a further method, the quality compliance set of each product is matched, and the specific matching process is: according to the quality compliance index of each product, the difference processing is performed with the preset quality compliance reference value to obtain the quality compliance deviation value of each product, and the quality compliance set corresponding to each quality compliance deviation value interval defined in the product quality management library is matched to obtain the quality compliance set of each product.

[0012] As a further method, the transformation model has a specific analysis process as follows: the first data set, the second data set and the third data set of each product are respectively constructed as a first data set model, a second data set model and a third data set model, and by obtaining the historical data of each product as the input of each data set model, the output result of each data set model can be obtained, and the data set model with the highest data synthesis adaptability is screened out and recorded as the transformation model.

[0013] The second aspect of the present invention provides a system for a method for synthesizing productive financial service data based on large model reasoning, including: a data set collection module, which is used to collect the operating data of the production equipment of each product during the production process, the supply chain information of the production line, and the weight and volume of each synthesized product, which are recorded as the first data set, the second data set, and the third data set of each product respectively; a transformation model extraction module, which is used to construct the first data set, the second data set, and the third data set of each product into a first data set model, a second data set model, and a third data set model respectively, and by obtaining the historical data of each product as the input of each data set model, the output result of each data set model can be obtained, and the model can be screened. The data set model with the highest data synthesis adaptability is recorded as the transformation model; the product quality synthesizable set matching module is used to obtain the data set information of each product, evaluate the quality compliance index of each product, and match the quality compliance synthesizable set of each product; the transformation model synthesis benefit value analysis module is used to synthesize the quality compliance synthesizable set of each product based on the transformation model, and analyze the synthesis benefit value of the transformation model; the data synthesis service optimization module is used to compare the synthesis benefit value of the transformation model with the set synthesis benefit threshold. If the synthesis benefit value of the transformation model is lower than the synthesis benefit threshold, the data synthesis optimization plan is matched, thereby realizing the data synthesis process of productive product quality.

[0014] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0015] (1) The present invention provides a method and system for synthesizing productive financial service data based on large model reasoning. First, different state data sets of each product in the production process are obtained, and the transformation model is extracted to provide model support for subsequent synthesis operations. The quality compliance index of each product is evaluated, and a quality-compliant synthesizable set of each product is matched. Synthesis operations are performed based on the transformation model, which can process and synthesize data more quickly, reduce synthesis waiting time, and increase the synthesis speed of the overall quality-compliant synthesizable set. The synthesis benefit value of the transformation model is analyzed, thereby realizing the data synthesis process of productive product quality. By optimizing the synthesis benefit of the transformation model, fast and accurate data synthesis services are provided to meet market demand.

[0016] (2) The present invention obtains data set information of each product and evaluates the quality compliance index of each product, so that production organizations can more accurately predict product quality risks, formulate appropriate risk management strategies for these risks, and match the quality compliance index of each product to form a synthesizable set. This can provide an objective and quantifiable basis for determining product quality compliance, helping production organizations make more accurate quality decisions when approving products.

[0017] (3) Based on the transformation model, the present invention synthesizes the quality-compliant synthesizable set of each product and analyzes the synthesis benefit value of the transformation model, which helps production organizations improve the reliability of data synthesis services for each product and enhances the management intensity of product quality risks and the synthesis efficiency of the quality-compliant synthesizable set. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0019] Figure 1 The figure is a schematic flow chart of the method steps of the present invention.

[0020] Figure 2 It is a schematic diagram of system module connection of the present invention.

[0021] Figure 3 This is the moving speed curve involved in the present invention.

[0022] Attached Figure 3 Shown: 1. Speed ​​threshold straight line. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0024] Reference Figure 1 As shown, the first aspect of the present invention provides a method for synthesizing productive financial service data based on large model reasoning, including: data set collection: collecting the operating data of the production equipment of each product in the production process, the supply chain information of the production line, and the weight and volume of each synthesized product, which are recorded as the first data set, the second data set, and the third data set of each product respectively.

[0025] Specifically, the first data set, the second data set, and the third data set of each product are analyzed in the following steps:

[0026] Based on the operating data of the production equipment in the production process, the average operating temperature and the average vibration frequency of the production equipment during the production cycle are extracted, where the equipment's average operating temperature and the average vibration frequency can be obtained through the temperature sensor and fluctuation sensor carried by the equipment itself, and are recorded as the first data set of each product; based on the supply chain information of the production line, the displacement and moving speed of each product are extracted, where the displacement and moving speed of each product can be obtained through each product positioning system in the supply chain, and are recorded as the second data set of each product; at the same time, based on the weight and volume of each synthesized product, where the weight and volume of each synthesized product can be extracted from each product production report, and are recorded as the third data set of each product.

[0027] Transformation model extraction: The first data set, the second data set and the third data set of each product are constructed as the first data set model, the second data set model and the third data set model respectively. By obtaining the historical data of each product as the input of each data set model, the output result of each data set model can be obtained, and the data set model with the highest data synthesis adaptability is screened out and recorded as the transformation model.

[0028] Furthermore, the specific analysis process of the transformation model is as follows:

[0029] The first data set, the second data set and the third data set of each product are divided into corresponding training sets, validation sets and test sets respectively, so as to construct the first data set model, the second data set model and the third data set model. The construction process is as follows: first, according to the characteristics of the data sets of each product, the appropriate model type is selected, the model is trained with the training set, the model parameters are adjusted with the validation set, and the accuracy and generalization ability of the model are evaluated with the test set, so as to determine the model of each data set. By using the historical data of each product as the input of each data set model, the output result of each data set model can be obtained, which is recorded as the output indicator set of each data set model, and the data set model with the highest data synthesis fitness is screened out and recorded as the transformation model, wherein the determination of the data synthesis fitness is determined by experts in the financial field based on historical financial data and their own experience. The specific screening process is as follows: first, the output indicator set intervals are set to determine which interval the output indicator set of each data set model belongs to, the data synthesis fitness corresponding to the interval is assigned to the model corresponding to the output indicator set of the data set model, the models that have been assigned the data synthesis fitness are arranged in order from small to large, and the model corresponding to the data synthesis fitness ranked first is recorded as the transformation model.

[0030] Product quality synthesizable set matching: obtain the data set information of each product, evaluate the quality compliance index of each product, and match the quality compliance synthesizable set of each product.

[0031] Specifically, the influence coefficient of abnormal operation of production equipment of each product is analyzed in the following specific process:

[0032] Based on the data set information of each product, the production equipment operation data of each product and the production line supply chain information of each product are extracted.

[0033] Based on the production equipment operation data of each product, the production equipment operation data includes the average operating temperature and the average vibration frequency of the production equipment during the production cycle.

[0034] The average operating temperature of the production equipment during the production cycle is processed by the difference with the operating reference temperature to obtain the temperature deviation value of the production equipment; the average vibration frequency of the production equipment during the production cycle is processed by the difference with the vibration reference frequency to obtain the vibration deviation value of the production equipment.

[0035] The operating reference temperature and vibration reference frequency are extracted from the product quality management library.

[0036] The data stored in the above-mentioned product quality management library are obtained by fitting after multiple tests on the product quality information. For example, the operating reference temperature is obtained by repeatedly obtaining the operating conditions of the production equipment at different temperatures. The test personnel determine the optimal operating temperature that meets the production equipment, and average all temperatures that meet the conditions to obtain the operating reference temperature.

[0037] The data defined in the above-mentioned product quality management library may change due to factors such as the production equipment status, quality status, and production environment of the product, and this embodiment does not make any special restrictions on this.

[0038] Multiply the temperature deviation value of the production equipment of each product by the correction factor corresponding to the temperature deviation value of the production equipment to obtain the temperature impact index of the production equipment of each product; multiply the vibration deviation value of the production equipment of each product by the correction factor corresponding to the vibration deviation value of the production equipment to obtain the vibration impact index of the production equipment of each product.

[0039] The correction factors corresponding to the temperature deviation values ​​of the production equipment and the correction factors corresponding to the vibration deviation values ​​of the production equipment mentioned above are used in this embodiment to reduce the negative impact of the temperature of the production equipment and the vibration of the production equipment of each product due to system errors in the process of calculating the influence coefficient of abnormal operation of the production equipment of each product; in this embodiment, the correction factors corresponding to the temperature deviation values ​​of the production equipment take a value range of 0.56 to 0.66; the correction factors corresponding to the vibration deviation values ​​of the production equipment take a value range of 0.59 to 0.65.

[0040] The temperature impact index of each product's production equipment and the vibration impact index of each product's production equipment are added together to obtain the abnormal operation impact coefficient β of each product's production equipment i .

[0041] Furthermore, the production line supply and demand impact coefficient of each product is analyzed in detail as follows:

[0042] Based on the production line supply chain information of each product, wherein the production line supply chain information includes the displacement of each product during the supply cycle.

[0043] The displacement of each product during the supply cycle is ratioed to the duration of the supply cycle to obtain the movement speed of each product during the supply cycle. The supply cycle is determined by the supply personnel through a comprehensive analysis of production time, equipment utilization, and worker efficiency.

[0044] According to the moving speed of each product in the supply cycle, a moving speed curve of the product is constructed, from which the number of products with a moving speed greater than a set speed threshold is extracted and recorded as the number of high-speed moving products.

[0045] The above speed threshold is determined by the supply personnel through a comprehensive analysis of supply time, equipment productivity, and worker efficiency.

[0046] According to the moving speed of each product in the supply cycle, the moving speed curve of the product is constructed, such as Figure 3 As shown, the horizontal axis of the product moving speed curve is the product, the unit is pieces, and the vertical axis is the moving speed, the unit is centimeters per minute.

[0047] According to the moving speed curve of the product, a speed threshold straight line 1 is located therefrom, and the number of all products whose moving speed is higher than the speed threshold straight line 1 is obtained, which is recorded as the number of high-speed moving products.

[0048] From the product quality management library, the influencing factors corresponding to the defined displacement and single high-speed moving products are extracted.

[0049] A comprehensive analysis is performed on the production line supply and demand impact coefficient of each product. In a specific embodiment, various information on the production line supply chain can be collected, including but not limited to supply chain logistics information, productive inventory information, and product sales information. Based on the collected data, various product quality risk assessment models can be established, and the production line supply and demand impact coefficient of each product can be obtained through the model.

[0050] The production line supply and demand impact coefficients of the above-mentioned products are obtained by comprehensively analyzing the displacement of each product during the supply cycle and the number of products moving at high speed in this embodiment, and are used to determine the value of the production line supply and demand impact of each product. In this embodiment, a more accurate calculation method is used to obtain the coefficients, and the specific expression is:

[0051]

[0052] where μ i It is expressed as the production line supply and demand impact coefficient of the i-th product. In this embodiment, if the product moves significantly during the supply cycle, the displacement of the product will increase accordingly. At the same time, the moving speed of the product per unit time will also increase significantly. By simplifying this expression, the quality risk status of different products can be more accurately grasped.

[0053] i represents the serial number of each product, i=1,2,3,...,m, and m represents the total number of products.

[0054] FZ i It is expressed as the displacement of the ith product during the supply cycle, which refers to the distance the product moves from the supply chain to the production line during the supply cycle.

[0055] FZ′ represents the limiting displacement and refers to the maximum value of the displacement.

[0056] GS is expressed as the number of products moving at high speed, which refers to the number of products whose moving speed is higher than the speed threshold.

[0057] In this embodiment, the number of high-speed moving products is set to 13, and the impact factor corresponding to a single high-speed moving product is set to 0.53. Then, in this expression The result of the partial calculation is 2.53.

[0058] a1 represents the correction factor corresponding to the predefined displacement, and a2 represents the impact factor corresponding to a single high-speed moving product, wherein the value of the correction factor corresponding to the displacement and the impact factor corresponding to the single high-speed moving product are determined by inviting experts in related fields. The experts finally determine the values ​​of the correction factor and the impact factor based on the displacement of the product, the relationship between the high-speed moving product and the product quality risk, and their own experience; in this embodiment, the correction factor corresponding to the displacement ranges from 0.38 to 0.52; the impact factor corresponding to the single high-speed moving product ranges from 0.47 to 0.59, and e is a natural constant.

[0059] Specifically, the quality compliance index of each product, in a specific embodiment, can be obtained by collecting detailed information of customers, conducting quality compliance risk analysis, building a quality compliance evaluation model based on the results of the quality compliance risk analysis, inputting the quality data of the product into the evaluation model, and the model will output a quality compliance index, thereby obtaining the quality compliance index of each product.

[0060] The quality compliance index of each product mentioned above is obtained by comprehensively analyzing the influence coefficient of abnormal operation of production equipment of each product and the influence coefficient of supply and demand of production line of each product in this embodiment, so as to determine the value of the quality compliance index of each product. In this embodiment, a more accurate calculation method is adopted to obtain the quality compliance index. The specific analysis process is as follows:

[0061]

[0062] where ω i It is expressed as the quality compliance index of the ith product. In this embodiment, if the impact coefficient of abnormal operation of the product's production equipment is high, it means that the equipment of the product has an abnormal condition during operation, which will increase the risk of the product's quality compliance, and will also increase the impact coefficient of the product's production line supply and demand. This expression performs a correlation analysis on the abnormal operation of the equipment and the supply and demand of the production line, which can better understand the quality compliance of the product and improve the accuracy of the prediction of product quality compliance.

[0063] i represents the serial number of each product, i=1,2,3,...,m, and m represents the total number of products.

[0064] β i It is expressed as the influence coefficient of abnormal operation of production equipment of the ith product, and represents the numerical value of the influence degree of abnormal operation of production equipment of each product obtained by comprehensive analysis of the operating temperature and vibration frequency of the production equipment of each product.

[0065] μ i It is expressed as the production line supply and demand impact coefficient of the ith product, which is the value of the production line supply and demand impact of each product obtained by comprehensive analysis of the displacement of each product during the supply cycle and the number of products moving at high speed.

[0066] In this embodiment, the production line supply and demand impact coefficient of the i-th product is set to 33%, and the value range of the weight factor corresponding to the production line supply and demand impact coefficient is set to 0.7. Then, μ in this expression is i *The calculation result of part b2 is 0.231.

[0067] b1 represents the weight factor corresponding to the predefined coefficient of influence of abnormal operation of production equipment, and b2 represents the weight factor corresponding to the predefined coefficient of influence of supply and demand of production line, wherein the weight factor corresponding to the coefficient of influence of abnormal operation of production equipment and the weight factor corresponding to the coefficient of influence of supply and demand of production line are obtained by analyzing the data of past product production cases, including production equipment information and supply chain information, understanding the actual influence of different production equipment information and supply chain information on the quality compliance of each product, and calculating the influence of different production equipment information and supply chain information through regression analysis. The above regression analysis is a statistical method. In this embodiment, the linear regression analysis method is used, which is mainly used to study the relationship between the independent variables of different production equipment information and supply chain information and the dependent variable quality compliance index, so as to obtain the weight factor corresponding to the coefficient of influence of abnormal operation of production equipment and the weight factor corresponding to the coefficient of influence of supply and demand of production line; in this embodiment, the value range of the weight factor corresponding to the coefficient of influence of abnormal operation of production equipment is 0.58 to 0.62, and the value range of the weight factor corresponding to the coefficient of influence of supply and demand of production line is 0.66 to 0.75.

[0068] Furthermore, the quality of each product that meets the standard can be matched and synthesized into a set. The specific matching process is:

[0069] According to the quality compliance index of each product, the difference processing is performed with the preset quality compliance reference value, wherein the quality compliance reference value is the quality compliance reference value given by the production organization after comprehensive evaluation of the product's quality compliance application based on its own product quality risk management policy and internal evaluation model, and the quality compliance deviation value of each product is obtained, and matched with the quality compliance synthesizable set corresponding to each quality compliance deviation value interval defined in the product quality management library, wherein the matching process is: setting each quality compliance deviation value interval, determining the interval to which the quality compliance deviation value of each product belongs, and allocating the quality compliance synthesizable set corresponding to the interval to the customer corresponding to the quality compliance deviation value, thereby obtaining the quality compliance synthesizable set of each product.

[0070] Analysis of the synthetic benefit value of the transformation model: Based on the transformation model, the quality standards of each product can be synthesized into a synthetic set, and the synthetic benefit value of the transformation model is analyzed.

[0071] The quality-compliant synthesizable set of each product is input into the transformation model, which has advantages in processing sequence data and complex relationships. That is, the transformation model relies on the self-attention mechanism to process sequences of variable length in the quality-compliant synthesizable set, relies on the multi-head self-attention mechanism to divide the quality-compliant synthesizable set into various representation subspaces, and performs self-attention calculations in parallel on multiple different representation subspaces.

[0072] The quality-compliant synthesizable set of each product is synthesized through the decoding output function of the transformation model, wherein the decoding output function is implemented based on the decoder in the transformation model. The main function of the decoder is to receive the synthesizable set vector processed by the encoder in the transformation model as input, that is, the output is the quality-compliant synthesized data of each product, and the Bayesian optimization method is used to screen the quality-compliant synthesized data of each product according to the synthesis specification indicators, wherein the Bayesian optimization is to guide the synthesis process by establishing a probability model of the synthesis set, so as to find the data synthesis configuration that makes the synthesis set obtain the optimal value, so that the quality-compliant synthesized data of each product meets the synthesis specification indicators, thereby realizing the synthesis operation of the quality-compliant synthesizable set of each product.

[0073] Furthermore, in a specific embodiment, the composite benefit value of the transformation model can be obtained by assigning different weights to each factor based on factors such as the composite score, composite status and composite error of the transformation model, multiplying the score of each factor by its weight, and then summing them up to obtain the composite benefit value of the transformation model.

[0074] In this embodiment, the synthetic benefit value of the above transformation model is obtained by synthesizing the quality standard synthesizable set of each product to obtain the synthetic information of the transformation model, specifically including the synthetic completion degree and synthetic matching degree of the transformation model, and comprehensively analyzing to obtain the synthetic benefit value of the transformation model, and importing it into the following analysis formula:

[0075]

[0076] Wherein D represents the synthetic benefit value of the transformation model. In this embodiment, when the synthetic completion degree of the transformation model and the synthetic matching degree of the transformation model increase, the synthetic benefit value of the transformation model will also increase accordingly, which will help the production organization to more accurately evaluate the impact of the synthetic benefit of the transformation model on the product quality compliance, so as to formulate more reasonable quality compliance synthesis policies and decisions. At the same time, a more compact algorithm operation is used for this expression, which can more clearly and accurately understand the synthetic benefit of the transformation model, and ultimately make the synthetic benefit of product quality compliance in an efficient state.

[0077] W represents the synthesis completion of the transformation model, which is obtained by comprehensive analysis of the data item completion rate of the transformation model and the data synthesis time efficiency.

[0078] In this embodiment, the synthesis completion of the transformation model is set to 93%, and the weight factor corresponding to the synthesis completion is set to 0.62. Then, the calculation result of the W*(1-g1) part in this expression is 0.35.

[0079] P represents the synthetic matching degree of the transformation model, which means the synthetic matching degree of the transformation model obtained by comprehensive analysis of the data synthesis consistency and data integrity of the transformation model.

[0080] g1 represents the weight factor corresponding to the predefined synthesis completion degree, and g2 represents the weight factor corresponding to the predefined synthesis matching degree, wherein the weight factor corresponding to the synthesis completion degree and the weight factor corresponding to the synthesis matching degree are both obtained by simulating the impact of different synthesis completion degrees and synthesis matching degrees on the final result of the synthesis benefit of the transformation model. According to the results of sensitivity analysis, the impact of different synthesis completion degrees and synthesis matching degrees on the synthesis completion degree is determined, and the weight factors are allocated accordingly. In this embodiment, the weight factor corresponding to the synthesis completion degree ranges from 0.58 to 0.65, and the weight factor corresponding to the synthesis matching degree ranges from 0.49 to 0.55, and e is a natural constant.

[0081] The values ​​of the synthesis completion of the above transformation model are shown in Table 1:

[0082] Table 1 Synthesis completion of transformation models

[0083]

[0084] In this embodiment, according to the table of synthesis completion of the above-mentioned transformation model, when the data item completion rate and the data synthesis time efficiency increase, the synthesis completion of the transformation model will also increase accordingly, that is, there is a positive proportional relationship between the data item completion rate and the data synthesis time efficiency and the synthesis completion of the transformation model. Therefore, if you want to improve the synthesis completion of the transformation model, you need to optimize the synthesis of the data items and the synthesis time, so as to improve the data item completion rate and the data synthesis time efficiency, and ensure that the synthesis completion of the transformation model is in a better state.

[0085] The values ​​of the synthetic matching degree of the above transformation model are shown in Table 2:

[0086] Table 2 Synthetic matching degree of transformation model

[0087]

[0088] In this embodiment, according to the table of the synthetic matching degree of the above-mentioned transformation model, when the data synthesis consistency and data integrity increase, the synthetic matching degree of the transformation model will also increase accordingly, that is, there is a positive proportional relationship between the data synthesis consistency and data integrity and the synthetic matching degree of the transformation model. Therefore, if you want to improve the synthetic matching degree of the transformation model, you need to improve the data item completion rate and the data synthesis time efficiency to ensure that the synthetic matching degree of the transformation model reaches an ideal state.

[0089] Data synthesis service optimization: The synthetic benefit value of the transformation model is compared with the set synthetic benefit threshold. If the synthetic benefit value of the transformation model is lower than the synthetic benefit threshold, a data synthesis optimization plan is matched, thereby realizing the data synthesis process of productive product quality.

[0090] Specifically, the data synthesis process for achieving the quality of productive products includes the following matching processes:

[0091] The synthetic benefit value of the transformation model is compared with a predefined synthetic benefit threshold, where the synthetic benefit threshold can be directly extracted from the product quality management library. If the synthetic benefit value of the transformation model is lower than the synthetic benefit threshold, the synthetic benefit value of the transformation model and the synthetic benefit threshold are differenced to obtain the synthetic benefit deviation value of the transformation model, and the synthetic benefit deviation value is matched with the data synthesis optimization plan corresponding to each synthetic benefit deviation value interval. The specific matching process is: set each synthetic benefit deviation value interval, determine the interval to which the synthetic benefit deviation value of the transformation model belongs, assign the data synthesis optimization plan corresponding to the interval to the transformation model corresponding to the synthetic benefit deviation value, and obtain the data synthesis optimization plan of the transformation model, thereby realizing the data synthesis process of productive product quality.

[0092] The above-mentioned data synthesis optimization plan is specifically as follows: screening the data sources required for product quality data synthesis, introducing transformation models to simplify the data synthesis process, establishing a real-time monitoring system for data synthesis, monitoring and adjusting the synthesis process in real time, and promptly identifying and adjusting problems based on real-time monitoring results to ensure the stability and reliability of data synthesis.

[0093] It needs to be explained that through the data synthesis operation on the quality of the above-mentioned productive products, the products will be put on the market for trading after quality assessment. The degree of transaction and transaction information can serve as an important basis for financial institutions to evaluate the credit of enterprises. Financial institutions can also use the benefit value synthesized by transformation model data to evaluate the risks and opportunities of the entire product production line supply chain, and provide financial services for each link in the product production line supply chain. At the same time, the quality assessment results can also help financial institutions identify potential market risks and promote innovation and development of the financial market.

[0094] Reference Figure 2 As shown, the second aspect of the present invention provides a system for a method for synthesizing productive financial service data based on large model reasoning, including: a data set collection module, a transformation model extraction module, a product quality synthesizable set matching module, a transformation model synthesis benefit value analysis module, and a data synthesis service optimization module.

[0095] The second aspect of the present invention provides a system for a method for synthesizing productive financial service data based on large model reasoning, and also includes: a product quality management library, wherein the product quality management library is used to store quality-compliant synthesizable sets corresponding to each quality-compliant deviation value interval.

[0096] The data set collection module is connected to the transformation model extraction module, the transformation model extraction module and the product quality synthesizable set matching module are both connected to the synthetic benefit value analysis module of the transformation model, the synthetic benefit value analysis module of the transformation model is connected to the data synthesis service optimization module, and the product quality synthesizable set matching module and the synthetic benefit value analysis module of the transformation model are both connected to the product quality management library.

[0097] The data set collection module is used to collect the operating data of the production equipment during the production process of each product, the supply chain information of the production line, and the weight and volume of each synthesized product, which are recorded as the first data set, the second data set and the third data set of each product, respectively, and then enter the transformation model extraction module.

[0098] The transformation model extraction module is used to construct the first data set, the second data set and the third data set of each product into a first data set model, a second data set model and a third data set model respectively. By obtaining the historical data of each product as the input of each data set model, the output result of each data set model can be obtained, and the data set model with the highest data synthesis adaptability is screened out and recorded as the transformation model, and enters the synthetic benefit value analysis module of the transformation model.

[0099] The product quality synthesizable set matching module is used to obtain the data set information of each product, evaluate the quality compliance index of each product, match the quality compliance synthesizable set of each product, and thus enter the synthetic benefit value analysis module of the transformation model.

[0100] The synthetic benefit value analysis module of the transformation model is used to synthesize the quality-compliant synthesizable sets of each product based on the transformation model, analyze the synthetic benefit value of the transformation model, and enter the data synthesis service optimization module.

[0101] The data synthesis service optimization module is used to compare the synthetic benefit value of the transformation model with the set synthetic benefit threshold. If the synthetic benefit value of the transformation model is lower than the synthetic benefit threshold, a data synthesis optimization plan is matched, thereby realizing the data synthesis process of productive product quality.

[0102] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. A method for synthesizing productive financial service data based on large model reasoning, characterized in that: include: Dataset collection: collect the operation data of the production equipment during the production process of each product, the supply chain information of the production line, and the weight and volume of each product, and record them as the first data set, the second data set, and the third data set of each product; Transformation model extraction: The first data set, the second data set and the third data set of each product are respectively constructed into the first data set model, the second data set model and the third data set model. The historical data of each product is obtained as the input of each data set model to obtain the output result of each data set model, and the data set model with the highest data synthesis adaptability is selected and recorded as the transformation model; Product quality synthesizable set matching: obtain the data set information of each product, evaluate the quality compliance index of each product, and match the quality compliance synthesizable set of each product; Analysis of the synthetic benefit value of the transformation model: Based on the transformation model, the synthetic set of quality standards of each product is synthesized to analyze the synthetic benefit value of the transformation model; Data synthesis service optimization: The synthetic benefit value of the transformation model is compared with the set synthetic benefit threshold. If the synthetic benefit value of the transformation model is lower than the synthetic benefit threshold, a data synthesis optimization plan is matched, thereby realizing the data synthesis process of productive product quality; The quality of each product that meets the standard can be matched and synthesized into a set. The specific matching process is: According to the quality compliance index of each product, the difference processing is performed with the preset quality compliance reference value to obtain the quality compliance deviation value of each product, and the quality compliance synthesizable set corresponding to each quality compliance deviation value interval defined in the product quality management library is matched to obtain the quality compliance synthesizable set of each product; The specific analysis process of the quality compliance index of each product is as follows: According to the data set information of each product, the production equipment operation data of each product and the production line supply chain information of each product are extracted, and the influence degree coefficient of abnormal operation of each product's production equipment and the influence degree coefficient of supply and demand of each product's production line are analyzed, and the quality compliance index of each product is comprehensively analyzed; The specific formula for the quality compliance index of each product is as follows: in It is represented as the quality index of the i-th product, i is the number of each product, , m represents the total amount of the product, It is expressed as the influence coefficient of abnormal operation of production equipment of the i-th product, It is expressed as the production line supply and demand impact coefficient of the i-th product, It is expressed as the weight factor corresponding to the predefined coefficient of influence of abnormal operation of production equipment, It is expressed as the weight factor corresponding to the predefined production line supply and demand impact coefficient; The data set model with the highest data synthesis fitness selected is recorded as the transformation model. The specific screening process is: first, set the intervals of each output indicator set, determine to which interval the output indicator set of each data set model belongs, assign the data synthesis fitness corresponding to the interval to the model corresponding to the output indicator set of the data set model, arrange the models that have been assigned data synthesis fitness in order from small to large, and record the model corresponding to the data synthesis fitness that ranks first as the transformation model.

2. The method for synthesizing productive financial service data based on large model reasoning according to claim 1 is characterized by: The data synthesis process for achieving the quality of productive products is specifically analyzed as follows: The synthetic benefit value of the transformation model is compared with the set synthetic benefit threshold. If the synthetic benefit value of the transformation model is lower than the synthetic benefit threshold, the synthetic benefit value of the transformation model and the synthetic benefit threshold are differenced to obtain the synthetic benefit deviation value of the transformation model. The difference is matched with the data synthesis optimization plan corresponding to each synthetic benefit deviation value interval to obtain the data synthesis optimization plan of the transformation model, thereby realizing the data synthesis process of productive product quality.

3. The method for synthesizing productive financial service data based on large model reasoning according to claim 1 is characterized by: The synthetic benefit value of the transformation model is specifically analyzed as follows: By synthesizing the quality-compliant synthesizable set of each product, the synthesis information of the transformation model is obtained, including the synthesis completion and synthesis matching degree of the transformation model. The synthetic benefit value of the transformation model is obtained through comprehensive analysis.

4. The method for synthesizing productive financial service data based on large model reasoning according to claim 3 is characterized by: The quality of each product that meets the standard can be synthesized into a set for synthesis operation, and the specific synthesis process is as follows: The quality-compliant synthesizable set of each product is input into the transformation model, and the synthesis operation is performed on the quality-compliant synthesizable set of each product through the decoding output function of the transformation model, and the output is the synthetic data of the transformation model. The Bayesian optimization method is used to screen the synthetic data of the transformation model according to the synthesis specification indicators, thereby realizing the synthesis operation of the quality-compliant synthesizable set of each product.

5. The method for synthesizing productive financial service data based on large model reasoning according to claim 1 is characterized by: The specific analysis process of the first data set, the second data set and the third data set of each product is as follows: Based on the operating data of the production equipment during the production process, the operating temperature and vibration frequency of the production equipment are extracted and recorded as the first data set of each product; based on the supply chain information of the production line, the displacement and moving speed of each product are extracted and recorded as the second data set of each product; at the same time, based on the weight and volume of each synthesized product, it is recorded as the third data set of each product.

6. A production financial service data synthesis system based on large model reasoning, used to execute the production financial service data synthesis method based on large model reasoning as described in any one of claims 1 to 5, characterized in that: include: The data set collection module is used to collect the operation data of the production equipment of each product in the production process, the supply chain information of the production line, and the weight and volume of each synthesized product, which are recorded as the first data set, the second data set and the third data set of each product, respectively, and then enter the transformation model extraction module; The transformation model extraction module is used to construct the first data set, the second data set and the third data set of each product into the first data set model, the second data set model and the third data set model respectively, obtain the historical data of each product as the input of each data set model, obtain the output result of each data set model, and select the data set model with the highest data synthesis fitness as the transformation model, and enter the synthesis benefit value analysis module of the transformation model; The data set model with the highest data synthesis fitness selected is recorded as a transformation model. The specific screening process is: first, set each output indicator set interval, determine to which interval the output indicator set of each data set model belongs, assign the data synthesis fitness corresponding to the interval to the model corresponding to the output indicator set of the data set model, arrange the models that have been assigned the data synthesis fitness in order from small to large, and record the model corresponding to the data synthesis fitness ranked first as the transformation model; The product quality synthesizable set matching module is used to obtain the data set information of each product, evaluate the quality compliance index of each product, match the quality compliance synthesizable set of each product, and then enter the synthetic benefit value analysis module of the transformation model; The synthetic benefit value analysis module of the transformation model is used to synthesize the quality-compliant synthesizable sets of each product based on the transformation model, analyze the synthetic benefit value of the transformation model, and enter the data synthesis service optimization module; The data synthesis service optimization module is used to compare the synthetic benefit value of the transformation model with the set synthetic benefit threshold. If the synthetic benefit value of the transformation model is lower than the synthetic benefit threshold, a data synthesis optimization plan is matched, thereby realizing the data synthesis process of productive product quality.

Images