File downloading prediction model training method and device, file downloading prediction method and device, equipment and medium

Abstract

The invention discloses a training method and device for a file downloading prediction model, a prediction method and device, equipment and a medium, and relates to the technical field of financial science and technology. The training method comprises the following steps: acquiring user downloading data to obtain a training sample set; determining an initial value of each training data in the training sample set to obtain an initial vector, determining a first input vector, a first weight vector and a second weight vector, and determining a first output vector; updating the initial vector to obtain a second initial vector, determining a second input vector, a first updated weight vector and a second updated weight vector, and determining a second output vector; the operation of updating the second initial vector continues to be executed until the output vector meets the preset optimization condition, and a file downloading prediction model is output; according to the scheme, a basis can be provided for fast and accurate downloading prediction of financial system files, the problem that financial system data management and resource allocation are difficult is solved, and the service quality is improved.

Description

Technical Field

[0001] The present invention relates to the field of financial technology, and in particular to a training and prediction method, apparatus, device, and medium for a file download prediction model. Background Art

[0002] With the development of FinTech and the acceleration of digital transformation, the amount of data generated by financial institutions has increased dramatically. This data includes but is not limited to transaction records, customer information, market analysis reports, and more. Effectively managing and utilizing this data has become a major challenge.

[0003] How to quickly and accurately predict the download of financial system files to help financial institutions solve data management and resource allocation problems, improve service quality, and enhance security is a key research issue in the industry. Summary of the Invention

[0004] The present invention provides a file download prediction model training, prediction method, device, equipment and medium, which provides a basis for quickly and accurately predicting the download of financial system files, solves the problems of difficult data management and resource allocation in the financial system, improves the service quality of financial institutions, and enhances the security of files.

[0005] According to one aspect of the present invention, a method for training a file download prediction model is provided, the method comprising:

[0006] Obtaining user download data associated with the target financial system within a preset time period, and preprocessing each user download data to obtain a training sample set; wherein the user download data is recorded in the form of a two-level lookup table;

[0007] Determining an initial value of each training data in the training sample set to obtain an initial vector, determining a first input vector, a first weight vector, and a second weight vector, and determining a first output vector based on the first input vector, the first weight vector, and the second weight vector;

[0008] updating the initial vector to obtain a second initial vector, determining a second input vector, a first updated weight vector, and a second updated weight vector, and determining a second output vector based on the second input vector, the first updated weight vector, and the second updated weight vector;

[0009] The operation of updating the second initial vector is continued until the output vector meets the preset optimization condition, and the output file downloads the prediction model.

[0010] According to another aspect of the present invention, a file download prediction method is provided, the method comprising:

[0011] In response to an operation instruction of a target user on a target financial system, obtaining historical user download data matching the target user;

[0012] Inputting the historical user download data into a file download prediction model, predicting the target user's download files, and pre-caching the predicted download files; wherein the file download prediction model is trained by the file download prediction model training method described in any embodiment of the present invention;

[0013] In response to a download instruction for a target file from a target user, if it is determined that the target file is within the predicted download file, the target file is downloaded based on the predicted download file.

[0014] According to another aspect of the present invention, a device for training a file download prediction model is provided, the device comprising:

[0015] A first acquisition module is configured to acquire user download data associated with a target financial system within a preset time period and pre-process each user download data to obtain a training sample set; wherein the user download data is recorded in the form of a two-level lookup table;

[0016] a first determining module, configured to determine an initial value of each training data in the training sample set to obtain an initial vector, determine a first input vector, a first weight vector, and a second weight vector, and determine a first output vector based on the first input vector, the first weight vector, and the second weight vector;

[0017] a second determining module, configured to update the initial vector to obtain a second initial vector, determine a second input vector, a first updated weight vector, and a second updated weight vector, and determine a second output vector based on the second input vector, the first updated weight vector, and the second updated weight vector;

[0018] The model output module is used to continue to perform the operation of updating the second initial vector until the output vector meets the preset optimization condition, and output the file to download the prediction model.

[0019] According to another aspect of the present invention, a file download prediction device is provided, the device comprising:

[0020] an acquisition module, configured to acquire historical user download data matching the target user in response to an operation instruction of the target user on the target financial system;

[0021] a prediction module, configured to input the historical user download data into a file download prediction model, predict the target user's download files, and pre-cache the predicted download files; wherein the file download prediction model is trained using the file download prediction model training method described in any embodiment of the present invention;

[0022] The download module is configured to respond to a download instruction of a target file from a target user and download the target file based on the predicted download file when it is determined that the target file is within the predicted download file.

[0023] According to another aspect of the present invention, an electronic device is provided, comprising:

[0024] at least one processor; and

[0025] a memory communicatively connected to the at least one processor; wherein,

[0026] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the file download prediction model training method or the file download prediction method described in any embodiment of the present invention.

[0027] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the file download prediction model training method or the file download prediction method described in any embodiment of the present invention when executed.

[0028] According to another aspect of the present invention, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the file download prediction model training method or the file download prediction method according to any embodiment of the present invention.

[0029] The technical solution of an embodiment of the present invention obtains user download data associated with a target financial system within a preset time period and pre-processes each user download data to obtain a training sample set; wherein the user download data is recorded in the form of a two-level lookup table, achieving hierarchical storage of the obtained data; determining an initial value of each training data in the training sample set to obtain an initial vector, determining a first input vector, a first weight vector, and a second weight vector, and determining a first output vector based on the first input vector, the first weight vector, and the second weight vector; updating the initial vector to obtain a second initial vector, determining a second input vector, a first updated weight vector, and a second updated weight vector, and determining a second output vector based on the second input vector, the first updated weight vector, and the second updated weight vector; different output vectors can be determined based on different step sizes and initial vectors; continuing to update the second initial vector until the output vector meets preset optimization conditions, and outputting a file download prediction model; providing a basis for quickly and accurately predicting file downloads in the financial system, solving the problems of difficult data management and resource allocation in the financial system, improving the service quality of financial institutions, and enhancing file security.

[0030] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0032] Figure 1 This is a flowchart of a method for training a file download prediction model according to the first embodiment of the present invention;

[0033] Figure 2 This is a flowchart of a method for training a file download prediction model according to a second embodiment of the present invention;

[0034] Figure 3 This is a flowchart of a file download prediction method provided according to the third embodiment of the present invention;

[0035] Figure 4 is a timing diagram of a file download prediction method provided according to the third embodiment of the present invention;

[0036] Figure 5 2. It is a structural diagram of a training device for a file download prediction model provided according to a fourth embodiment of the present invention;

[0037] Figure 6 This is a structural diagram of a file download prediction device provided according to a fifth embodiment of the present invention;

[0038] Figure 7 It is a structural diagram of an electronic device for implementing a training method for a file download prediction model or a file download prediction method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to enable those skilled in the art to better understand the solutions of the present invention, 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 embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0040] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0041] Example 1

[0042] Figure 1 This is a flowchart of a method for training a file download prediction model according to the first embodiment of the present invention. This embodiment is applicable to the case of training a file download prediction model for a financial system. The method can be executed by a training device for a file download prediction model. The training device for the file download prediction model can be implemented in the form of hardware and / or software. The training device for the file download prediction model can be configured in an electronic device such as a computer, a server, or a tablet computer. Figure 1 As shown, the method includes:

[0043] Step 110: Obtain user download data associated with the target financial system within a preset time period, and pre-process each user download data to obtain a training sample set.

[0044] The user download data is recorded in the form of a two-level lookup table.

[0045] The target financial system may be any system related to financial services, such as a payment processing system, a securities trading system, a customer relationship management system, a wealth management system, or a foreign exchange trading system, and is not limited in this embodiment.

[0046] In this embodiment, the preset time period may be the past year, the past month, or the past week, etc., which is not limited in this embodiment.

[0047] Optionally, in this embodiment, data downloaded by users in the target financial system within a preset time period can be obtained. In this embodiment, this data is referred to as user downloaded data, such as transaction details, account statements, statements, investment reports, or loan-related documents. Furthermore, the obtained downloaded data of each user can be preprocessed. For example, the obtained downloaded data of each user can be processed for outliers, missing values can be filled, or data can be standardized, so as to obtain a training sample set.

[0048] It should be noted that, in this embodiment, the acquisition of user downloaded data associated with the target financial system is only done after authorization by the user, and the acquisition method is reasonable and legal.

[0049] Step 120: Determine the initial value of each training data in the training sample set to obtain an initial vector, determine a first input vector, a first weight vector and a second weight vector, and determine a first output vector based on the first input vector, the first weight vector and the second weight vector.

[0050] Optionally, in this embodiment, after determining to obtain the training sample set, the initial value of each training data in the training sample set can be further followed up to determine the initial vector; further, the first input vector, the first weight vector and the second weight vector can be further determined, and the first output vector can be determined based on the first input vector, the first weight vector and the second weight vector.

[0051] Optionally, in this embodiment, determining the initial value of each training data in the training sample set to obtain an initial vector, and determining the first input vector, the first weight vector, and the second weight vector may include: determining each initial value according to the restriction conditions of the step formula, and determining the initial vector based on each initial value; determining each allocation resource priority coefficient obtained by the initial calculation as the first input vector; and determining the first weight vector and the second weight vector based on the initial vector.

[0052] In this embodiment, the step size formula can be μ(n) = αμ(n-1) + γ|e(n)*e(n-1)*e(n-2)|; wherein, the restriction conditions of the step size formula can be α(0<α<1), γ(γ>0); illustratively, the initial values can be α=0.5, γ=0.1; further, these parameters can be used to generate the initial vector μ(n).

[0053] In an example of this embodiment, assuming that the initial error sequences e(0), e(1), and e(2) are 0.2, 0.3, and 0.4, respectively, and α=0.5, γ=0.1, then μ(0) can be set to a small positive number (for example, 0.01); further, μ(1) can be calculated according to the formula: μ(1)=0.5×0.01+0.1×|0.2×0.3×0.4|=0.005+0.0024=0.0074.

[0054] Furthermore, each priority coefficient of allocated resources obtained by initial calculation may be determined as a first input vector; and a first weight vector and a second weight vector may be determined based on the initial vector.

[0055] Optionally, in this embodiment, the resource allocation priority coefficient d1-d obtained by the initial calculation can be used as n Determined as the first input vector x(n), where the resource priority coefficient d1-d n This will be described in the following text, which is not intended to limit this embodiment.

[0056] In this embodiment, the first weight vector and the second weight vector may be determined based on the following formula:

[0057] ω1(n)=ω1(n-1)+μ(n-1)*e(n-1)*x(n-2); ω2(n)=ω2(n-1)+μ(n-1)*e(n-1)*x(n-3);

[0058] Wherein, ω1(n) is the first weight vector; ω2(n) is the second weight vector.

[0059] Furthermore, determining the first output vector based on the first input vector, the first weight vector, and the second weight vector may include: inputting the first input vector, the first weight vector, and the second weight vector into a first formula to obtain the first output vector; wherein the first formula is:

[0060] y(n)=|ω1(n)*x(n-1)+ω2(n)*x(n-2)+x(n)|;

[0061] Wherein, ω1(n) is the first weight vector, ω2(n) is the second weight vector, and x(n) is the first input vector.

[0062] In an example of this embodiment, for example: ω1(1) = 0.6, ω2(1) = 0.4, x(0) = 0.1, x(1) = 0.2, x(2) = 0.3, then: y(2) = |0.6×0.2+0.4×0.1+0.3|=|0.12+0.04+0.3|=0.46.

[0063] Step 130: Update the initial vector to obtain a second initial vector, determine a second input vector, a first updated weight vector, and a second updated weight vector, and determine a second output vector based on the second input vector, the first updated weight vector, and the second updated weight vector.

[0064] Optionally, in this embodiment, the initial vector is updated to obtain a second initial vector, a second input vector, a first updated weight vector and a second updated weight vector are determined, and a second output vector is determined based on the second input vector, the first updated weight vector and the second updated weight vector. This may include: inputting the initial vector into a second formula to obtain the second initial vector; wherein the second formula is: μ(n)=αμ(n-1)+γ|e(n)e(n-1)e(n-2)|, (α=0.5, γ=0.1); wherein e(n) is the difference between the output vector and the expected vector; determining the second input vector based on the first output vector; determining the first updated weight vector and the second updated weight vector based on the second initial vector; inputting the second input vector, the first updated weight vector and the second updated weight vector into the first formula respectively to obtain the second output vector.

[0065] Optionally, in this embodiment, a new step size μ(n), i.e., a second initial vector, can be calculated based on the second formula μ(n) = αμ(n-1) + γ|e(n)e(n-1)e(n-2)|, (α = 0.5, γ = 0.1); exemplarily, in this embodiment, if it is assumed that e(3) = 0.5, then μ(2) = 0.5×0.0074+0.1×|0.5×0.4×0.3|=0.0037+0.006=0.0097.

[0066] Furthermore, ω1(n), ω2(n), i.e., the first updated weight vector and the second updated weight vector, can be calculated from e(n), x(n), x(n) at the previous moment, and ω1(n), ω2(n), and μ(n) calculated according to the previous iteration, using the formulas ω1(n) = ω1(n-1) + μ(n-1)*e(n-1)*x(n-2); ω2(n) = ω2(n-1) + μ(n-1)*e(n-1)*x(n-3).

[0067] Furthermore, the x(n) at the previous moment, the current x(n), ω1(n), and ω2(n), where x(n+1)=y(n), can be used to predict y(n) at the next moment according to the first formula y(n)=|ω1(n)*x(n-1)+ω2(n)*x(n-2)+x(n)|; wherein, the x(n) at the previous moment is updated to the secondary lookup table to predict the allocation resource priority coefficient, and y(n) is updated to the secondary lookup table to predict the allocation resource priority coefficient.

[0068] Step 140: Continue to update the second initial vector until the output vector meets the preset optimization condition, and then output the file to download the prediction model.

[0069] Optionally, in this embodiment, the operation of updating the initial vector to obtain a second initial vector, determining a second input vector, a first updated weight vector, and a second updated weight vector, and determining a second output vector based on the second input vector, the first updated weight vector, and the second updated weight vector can be repeated until the output second output vector meets the preset optimization conditions, and the file download prediction model is output.

[0070] In this embodiment, the preset optimization condition may be that the difference between the output vector and the expected vector is less than a set threshold, wherein the set threshold may be a relatively small value such as 0.001 or 0.0001, and the specific value is not limited in this embodiment.

[0071] In an optional implementation of this embodiment, when it is determined that the difference between the output vector and the expected vector is less than a set threshold, it can be determined that the output vector meets a preset optimization condition, and the file download prediction model is obtained.

[0072] Optionally, in this embodiment, a loop iteration may be performed based on the above embodiment until the output vector y(n) obtained by model training becomes closer and closer to the actual value.

[0073] The technical solution of this embodiment obtains user download data associated with a target financial system within a preset time period and pre-processes each user download data to obtain a training sample set; wherein the user download data is recorded in the form of a two-level lookup table, achieving hierarchical storage of the obtained data; determining the initial value of each training data in the training sample set to obtain an initial vector, determining a first input vector, a first weight vector, and a second weight vector, and determining a first output vector based on the first input vector, the first weight vector, and the second weight vector; updating the initial vector to obtain a second initial vector, determining a second input vector, a first updated weight vector, and a second updated weight vector, and determining a second output vector based on the second input vector, the first updated weight vector, and the second updated weight vector; different output vectors can be determined based on different step sizes and initial vectors; continuing to update the second initial vector until the output vector meets preset optimization conditions, and outputting a file download prediction model; providing a basis for quickly and accurately predicting downloads of financial system files, solving the problems of difficult data management and resource allocation in the financial system, improving the service quality of financial institutions, and enhancing file security.

[0074] Example 2

[0075] Figure 2 This is a flowchart of a training method for a file download prediction model provided in accordance with the second embodiment of the present invention. This embodiment is a further refinement of the above technical solution. The technical solution in this embodiment can be combined with various optional solutions in one or more of the above embodiments. Figure 2 As shown, the method includes:

[0076] Step 210: Obtain user download data associated with the target financial system within a preset time period.

[0077] Step 220: Determine a first-level lookup table based on the attribute parameters of each user's downloaded data, and store each user's downloaded data based on the first-level lookup table; determine a second-level lookup table based on the calculation parameters of each user's downloaded data, and store each user's downloaded data based on the second-level lookup table.

[0078] Among them, the attribute parameters include at least one of the following: geographic location, download time, file type, download speed and number of downloads; the calculation parameters include at least one of the following: allocation resource priority coefficient, predicted allocation resource priority coefficient, previous moment predicted allocation resource priority coefficient and number of iterations.

[0079] Optionally, in this embodiment, after obtaining the user download data associated with the target financial system within a preset time period, the attribute parameters such as the geographical location, download time, file type, download speed and number of downloads of the individual user download data can be further determined, and the download data of each user can be stored based on the first-level lookup table; the calculation parameters such as the allocation resource priority coefficient, the predicted allocation resource priority coefficient, the predicted allocation resource priority coefficient at the previous moment and the number of iterations can also be determined during the iterative training process of the file download prediction model, and stored in the second-level lookup table.

[0080] In a specific example of this embodiment, the first-level lookup table may be as shown in Table 1, and the second-level lookup table may be as shown in Table 2;

[0081] Table 1

[0082]

[0083] Table 2

[0084]

[0085] It should be noted that the resource priority coefficient allocated by the above-mentioned lookup table can be initially obtained by weighted averaging the data grouped by geographical location according to the first-level lookup table. For example, the ID1 priority coefficient d1 = position 1*position coefficient + time 1*time weighting coefficient + type 1*file type weighting coefficient + n1*download speed weighting coefficient + ID1 download times*download times weighting coefficient +...; the predicted resource allocation priority coefficient can be determined according to the variable step size based on the iterative calculation involved in the above-mentioned embodiment.

[0086] Step 230: Determine the initial value of each training data in the training sample set to obtain an initial vector, determine a first input vector, a first weight vector and a second weight vector, and determine a first output vector based on the first input vector, the first weight vector and the second weight vector.

[0087] Step 240: Update the initial vector to obtain a second initial vector, determine a second input vector, a first updated weight vector, and a second updated weight vector, and determine a second output vector based on the second input vector, the first updated weight vector, and the second updated weight vector.

[0088] Step 250: Continue to update the second initial vector until the output vector meets the preset optimization condition, and then output the file to download the prediction model.

[0089] The solution of this embodiment, after obtaining user download data associated with the target financial system within a preset time period, can further determine attribute parameters such as the geographical location, download time, file type, download speed, and number of downloads of each user's download data, and store the download data of each user based on the first-level lookup table; it can also determine calculation parameters such as the allocation resource priority coefficient, the predicted allocation resource priority coefficient, the previous moment predicted allocation resource priority coefficient, and the number of iterations during the iterative training process of the file download prediction model, and store them in the second-level lookup table. The user download data can be stored in different data tables based on the different properties of the data, which provides assistance for subsequent rapid and accurate training of the file download prediction model.

[0090] Example 3

[0091] Figure 3 This is a flowchart of a file download prediction method provided according to the first embodiment of the present invention. This embodiment is applicable to the case of predicting the user's data to be downloaded. The method can be executed by a file download prediction device. The file download prediction device can be implemented in the form of hardware and / or software. The file download prediction device can be configured in electronic devices such as computers, servers or tablet computers. Figure 3 As shown, the method includes:

[0092] Step 310: In response to the target user's operation instruction on the target financial system, obtain historical user download data matching the target user.

[0093] The target user may be any user who uses the target financial system, such as a legal person user or an individual user, which is not limited in this embodiment.

[0094] In an optional implementation of this embodiment, after receiving the target user's operation instructions on the target financial system, historical user download data matching the target user can be further obtained; illustratively, the target user's download data in the target financial system in the previous quarter can be obtained, such as transaction details, account statements, statements or investment reports.

[0095] Step 320: Input the historical user download data into a file download prediction model, predict the download files of the target user, and pre-cache the predicted download files.

[0096] The file download prediction model is obtained by training using the file download prediction model training method described in any of the above embodiments.

[0097] Optionally, in this embodiment, after obtaining historical user download data that matches the target user, the obtained historical user download data can be further input into the file download prediction model obtained by the above training for prediction, thereby obtaining a predicted download file. Furthermore, the predicted download file can be cached in advance so that when the target user wants to download the file subsequently, it can be quickly fed back to the target user.

[0098] Step 330: In response to a download instruction for a target file from a target user, if it is determined that the target file is within the predicted download file, download the target file based on the predicted download file.

[0099] The target file may be a transaction detail file, an account statement, a statement, an investment report or other file in the target financial system, which is not limited in this embodiment.

[0100] Optionally, in this embodiment, after receiving the download instruction for the target file from the target user, it can be further determined whether the target file is in the predicted download file. If it is determined that the target file is in the predicted download file, the target file can be obtained directly based on the cached predicted download file; illustratively, in this embodiment, the target file can be fed back to the target user through CDN (Content Delivery Network).

[0101] In another optional implementation of this embodiment, when it is determined that the target file is not in the predicted download file, the target file can be further obtained in the target financial system, and the file download prediction model can be adjusted based on the target file, that is, the target file can be added to the training sample set of the file download prediction model to optimize the file download prediction model, thereby helping to improve the accuracy of the file download prediction model.

[0102] The solution of this embodiment obtains historical user download data matching the target user by responding to the target user's operating instructions on the target financial system; inputs the historical user download data into a file download prediction model, predicts the target user's download file, and pre-caches the predicted download file; provides a basis for helping users quickly obtain the target file, thereby improving the user experience; and can also respond to the target user's download instruction for the target file, and when it is determined that the target file is within the predicted download file, download the target file based on the predicted download file, thereby quickly obtaining the target file.

[0103] In order to better understand the file download prediction method involved in this embodiment, Figure 4 This is a timing diagram of a file download prediction method provided according to Example 3 of the present invention, wherein CDN is an intelligent virtual network built on the basis of the existing network, relying on cache clusters deployed in various places, and through the load balancing, content distribution, scheduling and other functional modules of the financial system, enabling users to obtain the required content nearby, thereby reducing network congestion and improving user access response speed and cache hit rate.

[0104] The data collection module collects user download data, including download time, file type, download speed, geographic location, number of downloads and other information.

[0105] The prediction model building module builds a prediction model based on the collected historical data to predict users' future download needs.

[0106] Resource allocation optimization module: Dynamically adjusts the resource allocation of CDN nodes based on the prediction results, giving priority to ensuring the download speed of regions and file types with high predicted download demand.

[0107] Feedback and adjustment module: collects actual download data, compares it with the prediction results, and continuously adjusts the prediction model through iterative optimization of the LMS algorithm to improve prediction accuracy.

[0108] The solution of the embodiment of the present invention can reduce waiting time by predicting user download needs and preloading files in advance; it can dynamically adjust CDN node resources to avoid resource waste and improve overall network efficiency; it can reduce network latency, improve the stability and speed of file downloads, and enhance user satisfaction.

[0109] Example 4

[0110] Figure 5 Schematic diagram of a training device for a file download prediction model according to the fourth embodiment of the present invention. Figure 5 As shown, the device includes: a first acquisition module 510 , a first determination module 520 , a second determination module 530 and a model output module 540 .

[0111] The first acquisition module 510 is configured to acquire user download data associated with the target financial system within a preset time period and pre-process each user download data to obtain a training sample set; wherein the user download data is recorded in the form of a two-level lookup table;

[0112] A first determining module 520 is configured to determine an initial value of each training data in the training sample set to obtain an initial vector, determine a first input vector, a first weight vector, and a second weight vector, and determine a first output vector based on the first input vector, the first weight vector, and the second weight vector;

[0113] a second determining module 530, configured to update the initial vector to obtain a second initial vector, determine a second input vector, a first updated weight vector, and a second updated weight vector, and determine a second output vector based on the second input vector, the first updated weight vector, and the second updated weight vector;

[0114] The model output module 540 is configured to continue updating the second initial vector until the output vector satisfies a preset optimization condition, and output the prediction model as a file download.

[0115] In the solution of this embodiment, a first acquisition module acquires user download data associated with a target financial system within a preset time period, and pre-processes each user download data to obtain a training sample set; wherein the user download data is recorded in the form of a two-level lookup table; a first determination module determines an initial value of each training data in the training sample set to obtain an initial vector, determines a first input vector, a first weight vector, and a second weight vector, and determines a first output vector based on the first input vector, the first weight vector, and the second weight vector; a second determination module updates the initial vector to obtain a second initial vector, determines a second input vector, a first updated weight vector, and a second updated weight vector, and determines a second output vector based on the second input vector, the first updated weight vector, and the second updated weight vector; a model output module continues to execute the operation of updating the second initial vector until the output vector meets a preset optimization condition, thereby outputting a file download prediction model, which can provide a basis for quickly and accurately predicting the download of financial system files, solve the problems of difficult data management and resource allocation in the financial system, improve the service quality of financial institutions, and enhance the security of files.

[0116] In an optional implementation of this embodiment, the first acquisition module 510 is specifically configured to:

[0117] Determining a first-level lookup table based on attribute parameters of each user downloaded data, and storing each user downloaded data based on the first-level lookup table;

[0118] determining a second-level lookup table based on calculation parameters of each user downloaded data, and storing each user downloaded data based on the second-level lookup table;

[0119] The attribute parameters include at least one of the following: geographic location, download time, file type, download speed, and number of downloads;

[0120] The calculation parameters include at least one of the following: an allocation resource priority coefficient, a predicted allocation resource priority coefficient, a previous moment predicted allocation resource priority coefficient, and the number of iterations.

[0121] In an optional implementation of this embodiment, the first determining module 520 is specifically configured to:

[0122] Determine each initial value according to the constraint condition of the step size formula, and determine the initial vector based on each initial value;

[0123] Determine the priority coefficients of each allocated resource obtained by initial calculation as a first input vector;

[0124] A first weight vector and a second weight vector are determined based on the initial vector.

[0125] In an optional implementation of this embodiment, the first determining module 520 is further specifically configured to:

[0126] The first input vector, the first weight vector, and the second weight vector are respectively input into a first formula to obtain the first output vector; wherein the first formula is:

[0127] y(n)=|ω1(n)*x(n-1)+ω2(n)*x(n-2)+x(n)|;

[0128] Wherein, ω1(n) is the first weight vector, ω2(n) is the second weight vector, and x(n) is the first input vector.

[0129] In an optional implementation of this embodiment, the second determining module 530 is specifically configured to:

[0130] The initial vector is input into the second formula to obtain the second initial vector; wherein the second formula is:

[0131] μ(n)=αμ(n-1)+γ|e(n)e(n-1)e(n-2)|, (α=0.5, γ=0.1);

[0132] Where, e(n) is the difference between the output vector and the expected vector;

[0133] determining the second input vector based on the first output vector;

[0134] Determine the first updated weight vector and the second updated weight vector based on the second initial vector;

[0135] The second input vector, the first updated weight vector, and the second updated weight vector are respectively input into the first formula to obtain the second output vector.

[0136] In an optional implementation of this embodiment, the model output module 540 is specifically configured to:

[0137] When it is determined that the difference between the output vector and the expected vector is less than a set threshold, it is determined that the output vector meets a preset optimization condition, and the file download prediction model is obtained.

[0138] The training device for the file download prediction model provided in the embodiment of the present invention can execute the training method for the file download prediction model provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0139] Example 5

[0140] Figure 6 FIG. 1 is a schematic diagram of the structure of a file download prediction device provided according to the fifth embodiment of the present invention. Figure 6 As shown, the device includes: an acquisition module 610 , a prediction module 620 and a download module 630 .

[0141] The acquisition module 610 is configured to acquire historical user download data matching the target user in response to the target user's operation instruction on the target financial system;

[0142] Prediction module 620, configured to input the historical user download data into a file download prediction model, predict the target user's download files, and pre-cache the predicted download files; wherein the file download prediction model is trained using the file download prediction model training method described in any of the embodiments of the present invention;

[0143] The download module 630 is configured to respond to a download instruction for a target file from a target user and, if it is determined that the target file is within the predicted download file, download the target file based on the predicted download file.

[0144] The solution of this embodiment is to obtain historical user download data matching the target user in response to the target user's operation instructions on the target financial system through the acquisition module 610; input the historical user download data into the file download prediction model through the prediction module, predict the download file of the target user, and pre-cache the predicted download file; provide a basis for helping users quickly obtain the target file, thereby improving the user experience; respond to the target user's download instruction for the target file through the download module, and when it is determined that the target file is within the predicted download file, download the target file based on the predicted download file, so that the target file can be quickly obtained.

[0145] The file download prediction device provided by the embodiment of the present invention can execute the file download prediction method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0146] In the technical solutions of the embodiments of the present invention, the collection, storage, use, processing, transmission, provision and disclosure of user downloaded data are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0147] Example 6

[0148] Figure 7 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0149] like Figure 7As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0150] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0151] The processor 11 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors that run machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the training method of the file download prediction model or the file download prediction method.

[0152] In some embodiments, the training method of the file download prediction model, or the file download prediction method, may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, the training method of the file download prediction model described above, or one or more steps of the file download prediction method, may be executed. Alternatively, in other embodiments, the processor 11 may be configured to execute the training method of the file download prediction model, or the file download prediction method, by any other appropriate means (e.g., by means of firmware).

[0153] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0154] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0155] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0156] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0157] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0158] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0159] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0160] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

[0161] An embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the database detection method provided in any embodiment of the present application.

[0162] The computer program product may be implemented by writing computer program code for performing the operations of the present invention in one or more programming languages, or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0163] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned. They should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0164] Note that the above are only preferred embodiments of the present invention and the technical principles employed. Those skilled in the art will appreciate that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments, and substitutions are possible for those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments and may include many other equivalent embodiments without departing from the scope of the present invention. The scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for training a file download prediction model, characterized in that: include: Obtaining user download data associated with the target financial system within a preset time period, and preprocessing each user download data to obtain a training sample set; wherein the user download data is recorded in the form of a two-level lookup table; Determining an initial value of each training data in the training sample set to obtain an initial vector, determining a first input vector, a first weight vector, and a second weight vector, and determining a first output vector based on the first input vector, the first weight vector, and the second weight vector; updating the initial vector to obtain a second initial vector, determining a second input vector, a first updated weight vector, and a second updated weight vector, and determining a second output vector based on the second input vector, the first updated weight vector, and the second updated weight vector; The operation of updating the second initial vector is continued until the output vector meets the preset optimization condition, and the output file downloads the prediction model.

2. The training method of the file download prediction model according to claim 1, characterized in that: The pre-processing of the user downloaded data to obtain a training sample set includes: Determining a first-level lookup table based on attribute parameters of each user downloaded data, and storing each user downloaded data based on the first-level lookup table; determining a second-level lookup table based on calculation parameters of each user downloaded data, and storing each user downloaded data based on the second-level lookup table; The attribute parameters include at least one of the following: geographic location, download time, file type, download speed, and number of downloads; The calculation parameters include at least one of the following: an allocation resource priority coefficient, a predicted allocation resource priority coefficient, a previous moment predicted allocation resource priority coefficient, and the number of iterations.

3. The training method of the file download prediction model according to claim 2, characterized in that: The determining of the initial value of each training data in the training sample set to obtain an initial vector, and determining the first input vector, the first weight vector, and the second weight vector includes: Determine each initial value according to the constraint condition of the step size formula, and determine the initial vector based on each initial value; Determine the priority coefficients of each allocated resource obtained by initial calculation as a first input vector; A first weight vector and a second weight vector are determined based on the initial vector.

4. The training method of the file download prediction model according to claim 3, characterized in that: The determining a first output vector based on the first input vector, the first weight vector, and the second weight vector includes: The first input vector, the first weight vector, and the second weight vector are respectively input into a first formula to obtain the first output vector; wherein the first formula is: y(n)=|ω1(n) * x(n-1)+ω2(n) * x(n-2)+x(n)|; Wherein, ω1(n) is the first weight vector, ω2(n) is the second weight vector, and x(n) is the first input vector.

5. The training method of the file download prediction model according to claim 4, characterized in that: The updating of the initial vector to obtain a second initial vector, determining a second input vector, a first updated weight vector, and a second updated weight vector, and determining a second output vector based on the second input vector, the first updated weight vector, and the second updated weight vector, includes: The initial vector is input into the second formula to obtain the second initial vector; wherein the second formula is: μ(n)=αμ(n-1)+γ|e(n)e(n-1)e(n-2)|, (α=0.5, γ=0.1); Where, e(n) is the difference between the output vector and the expected vector; determining the second input vector based on the first output vector; Determine the first updated weight vector and the second updated weight vector based on the second initial vector; The second input vector, the first updated weight vector, and the second updated weight vector are respectively input into the first formula to obtain the second output vector.

6. The method for training a file download prediction model according to claim 1, wherein: The step of continuing to update the second initial vector until the output vector satisfies a preset optimization condition and outputting a file to download the prediction model comprises: When it is determined that the difference between the output vector and the expected vector is less than a set threshold, it is determined that the output vector meets a preset optimization condition, and the file download prediction model is obtained.

7. A file download prediction method, characterized in that: include: In response to an operation instruction of a target user on a target financial system, obtaining historical user download data matching the target user; Inputting the historical user download data into a file download prediction model, predicting the target user's download files, and pre-caching the predicted download files; wherein the file download prediction model is trained by the file download prediction model training method according to any one of claims 1 to 6; In response to a download instruction for a target file from a target user, if it is determined that the target file is within the predicted download file, the target file is downloaded based on the predicted download file.

8. A training device for a file download prediction model, characterized in that: include: A first acquisition module is configured to acquire user download data associated with a target financial system within a preset time period and pre-process each user download data to obtain a training sample set; wherein the user download data is recorded in the form of a two-level lookup table; a first determining module, configured to determine an initial value of each training data in the training sample set to obtain an initial vector, determine a first input vector, a first weight vector, and a second weight vector, and determine a first output vector based on the first input vector, the first weight vector, and the second weight vector; a second determining module, configured to update the initial vector to obtain a second initial vector, determine a second input vector, a first updated weight vector, and a second updated weight vector, and determine a second output vector based on the second input vector, the first updated weight vector, and the second updated weight vector; The model output module is used to continue to perform the operation of updating the second initial vector until the output vector meets the preset optimization condition, and output the file to download the prediction model.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the file download prediction model training method described in any one of claims 1 to 6, or the file download prediction method described in claim 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the file download prediction model training method described in any one of claims 1 to 6, or the file download prediction method described in claim 7 when executed.

Images