Mirror image data processing method and system based on evaluation and prediction

Abstract

The invention discloses a mirroring data processing method and system based on evaluation and prediction. The method comprises the steps that multiple pieces of mirroring data and corresponding recording parameters are acquired; calculating data quality corresponding to each piece of mirror image data according to the recording parameter corresponding to each piece of mirror image data; sorting part or all of the mirror image data according to the data quality from high to low to obtain a mirror image recommendation sequence; and pushing the mirror image recommendation sequence to a display interface of at least one related user for display. It can be seen that the mirror image data evaluation reasonability and management efficiency can be remarkably improved, mirror image data with higher quality can be recommended to the user, and the user experience is further optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a mirror data processing method and system based on evaluation prediction. Background Art

[0002] As artificial intelligence (AI) technology continues to gain traction and advance, many technicians and technical institutions are increasingly leveraging server resources on cloud platforms to train algorithm models. The model architecture, data, and resulting algorithm models used in these training processes are typically stored as image data on the server. Some cloud platforms now allow users to share their server image data for research or application. This sharing model is driving the rapid development of algorithm technology, but managing this image data efficiently and effectively remains a key technical challenge. Current image data management technologies often focus solely on basic storage, search, and download capabilities, failing to fully leverage user usage records or feedback to evaluate the quality of image data. This results in limited management effectiveness and a relatively poor user experience. Clearly, existing technologies have shortcomings that urgently need to be addressed. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a mirror data processing method and system based on evaluation prediction, which can significantly improve the rationality and management efficiency of mirror data evaluation, recommend higher quality mirror data to users, and further optimize the user experience.

[0004] In order to solve the above technical problems, the first aspect of the present invention discloses a mirror data processing method based on evaluation prediction, the method comprising: Obtain multiple image data and corresponding recording parameters; Calculating the data quality corresponding to each mirror data according to the recording parameters corresponding to each mirror data; Sort part or all of the mirror data from high to low according to the data quality to obtain a mirror recommendation sequence; The mirror recommendation sequence is pushed to a display interface of at least one relevant user for display.

[0005] As an optional embodiment, in the first aspect of the present invention, the recorded parameters include historical usage records and usage feedback records; the historical usage records include user information, usage scenario details, usage prediction data, parameters of the usage device, and performance records during use.

[0006] As an optional embodiment, in the first aspect of the present invention, the usage feedback record includes an assessment of data adequacy, an evaluation of the algorithm architecture, an assessment of the algorithm training degree, an analysis of the algorithm training plan, and feedback on the data labeling degree.

[0007] As an optional implementation manner, in the first aspect of the present invention, calculating the data quality corresponding to each mirror data according to the recording parameter corresponding to each mirror data includes: For each mirror data, a plurality of reuse record sets are screened from the corresponding historical usage records; each of the reuse record sets includes the same historical usage records recorded by multiple users, the number of which exceeds a first threshold; Based on the set of reused records, a neural network algorithm is used to calculate the excellent parameters of the usage records corresponding to the mirror data; Calculating the evaluation excellence parameter corresponding to the mirror data using a neural network algorithm based on the usage feedback record corresponding to the mirror data; The usage record excellence parameter and the evaluation excellence parameter are added together to obtain the data quality corresponding to the mirror data.

[0008] As an optional embodiment, in the first aspect of the present invention, the step of calculating the usage record excellence parameters corresponding to the mirrored data using a neural network algorithm based on the reuse record set includes: Calculating the ratio of the number of records in each of the reused record sets to the total number of records in all the reused record sets; Inputting each of the reuse record sets into a trained operation stability prediction neural network to obtain an operation stability parameter corresponding to each of the reuse record sets; the operation stability prediction neural network is trained using a training data set comprising a plurality of training usage records and their corresponding operation stability annotations; The operation stability parameters corresponding to all the reuse record sets are weightedly summed to obtain the usage record excellence parameter corresponding to the mirror data; wherein the weighted calculation weight of each operation stability parameter is positively correlated with the proportion of the corresponding reuse record set.

[0009] As an optional embodiment, in the first aspect of the present invention, the step of calculating the evaluation excellence parameter corresponding to the mirror data using a neural network algorithm based on the usage feedback record corresponding to the mirror data includes: For each evaluation record in the usage feedback record corresponding to the mirror data, input the evaluation record into the trained LLM model to extract all evaluation terms contained in the evaluation record; Identifying positive evaluation words among all the evaluation words according to a preset positive evaluation word database; Identifying negative evaluation words among all the evaluation words according to a preset negative evaluation word database; Calculate the ratio of all positive evaluation words in the evaluation record to all evaluation words to obtain the first word parameter; Calculate the inverse of the ratio of all negative evaluation words in the evaluation record to all evaluation words to obtain the second word parameter; Performing a weighted summation on the first word parameter and the second word parameter to obtain a record excellence parameter of the evaluation record; The sum of the record excellence parameters of all the evaluation records in the usage feedback record corresponding to the mirror data is calculated to obtain the evaluation excellence parameter corresponding to the mirror data.

[0010] As an optional embodiment, in the first aspect of the present invention, determining a display data sequence based on the current user's browsing operation and the mirror recommendation sequence to push to a display interface of at least one relevant user for display includes: Eliminate all mirror data whose data quality is lower than a parameter threshold in the mirror recommendation sequence to obtain a sequence after elimination; Determining the current user's usage record emphasis and evaluation emphasis tendency based on the current user's browsing operation; Calculating a first weight proportional to the tendency to attach importance to the usage record; Calculating a second weight proportional to the evaluation emphasis tendency; For each mirror data in the eliminated sequence, calculating a weighted sum of the usage record excellence parameter and the evaluation excellence parameter of the mirror data to obtain a new data quality corresponding to the mirror data; wherein the weighted calculation weight of the usage record excellence parameter is the first weight, and the weighted calculation weight of the evaluation excellence parameter is the second weight; The eliminated sequence is reordered from largest to smallest according to the quality of the new data to obtain a display data sequence, which is pushed to a display interface of at least one relevant user for display.

[0011] As an optional embodiment, in the first aspect of the present invention, determining the current user's usage record emphasis tendency and evaluation emphasis tendency based on the current user's browsing operation includes: Determine the corresponding cursor movement path and cursor click object in the current user's browsing operation; According to the cursor movement path, a plurality of path passing areas are determined on the current browser page; According to the cursor click object, determining a plurality of clicked objects from a plurality of model objects of the current browser page; Counting the number of first areas belonging to the usage record display area and the number of second areas belonging to the evaluation display area among all areas passed by the path; Counting the number of first objects that are usage record-related objects and the number of second objects that are evaluation-related objects among all the clicked objects; Calculate a first ratio of the number of the first areas to the total number of areas passed by all the paths; Calculate a second ratio of the number of the second areas to the total number of areas; Calculate a third ratio of the number of the first objects to the total number of all clicked objects; Calculate a fourth ratio of the second number of objects to the total number of objects; Calculating the product of the first quantity ratio and the third quantity ratio to obtain the current user's usage record emphasis tendency; The product of the second quantity proportion and the fourth quantity proportion is calculated to obtain the evaluation emphasis tendency of the current user.

[0012] A second aspect of an embodiment of the present invention discloses a mirror data processing system based on evaluation prediction, the system comprising: An acquisition module, used to acquire multiple image data and corresponding recording parameters; a calculation module, configured to calculate the data quality corresponding to each mirror data according to the recording parameters corresponding to each mirror data; A sorting module, configured to sort part or all of the image data from high to low according to the data quality to obtain an image recommendation sequence; The push module is used to push the image recommendation sequence to the display interface of at least one relevant user for display.

[0013] As an optional embodiment, in the second aspect of the present invention, the recorded parameters include historical usage records and usage feedback records; the historical usage records include user information, usage scenario details, usage prediction data, parameters of the usage device, and performance records during use.

[0014] As an optional embodiment, in the second aspect of the present invention, the usage feedback record includes an assessment of data adequacy, an evaluation of the algorithm architecture, an assessment of the algorithm training degree, an analysis of the algorithm training plan, and feedback on the data labeling degree.

[0015] As an optional embodiment, in the second aspect of the present invention, the specific manner in which the calculation module calculates the data quality corresponding to each mirror data according to the recording parameters corresponding to each mirror data includes: For each mirror data, a plurality of reuse record sets are screened from the corresponding historical usage records; each of the reuse record sets includes the same historical usage records recorded by multiple users, the number of which exceeds a first threshold; Based on the set of reused records, a neural network algorithm is used to calculate the excellent parameters of the usage records corresponding to the mirror data; Calculating the evaluation excellence parameter corresponding to the mirror data using a neural network algorithm based on the usage feedback record corresponding to the mirror data; The usage record excellence parameter and the evaluation excellence parameter are added together to obtain the data quality corresponding to the mirror data.

[0016] As an optional embodiment, in the second aspect of the present invention, the specific manner in which the calculation module calculates the excellent parameters of the usage records corresponding to the mirrored data using a neural network algorithm based on the reuse record set includes: Calculating the ratio of the number of records in each of the reused record sets to the total number of records in all the reused record sets; Inputting each of the reuse record sets into a trained operation stability prediction neural network to obtain an operation stability parameter corresponding to each of the reuse record sets; the operation stability prediction neural network is trained using a training data set comprising a plurality of training usage records and their corresponding operation stability annotations; The operation stability parameters corresponding to all the reuse record sets are weightedly summed to obtain the usage record excellence parameter corresponding to the mirror data; wherein the weighted calculation weight of each operation stability parameter is positively correlated with the proportion of the corresponding reuse record set.

[0017] As an optional embodiment, in the second aspect of the present invention, the specific manner in which the calculation module calculates the excellent evaluation parameter corresponding to the mirror data using a neural network algorithm based on the usage feedback record corresponding to the mirror data includes: For each evaluation record in the usage feedback record corresponding to the mirror data, input the evaluation record into the trained LLM model to extract all evaluation terms contained in the evaluation record; Identifying positive evaluation words among all the evaluation words according to a preset positive evaluation word database; Identifying negative evaluation words among all the evaluation words according to a preset negative evaluation word database; Calculate the ratio of all positive evaluation words in the evaluation record to all evaluation words to obtain the first word parameter; Calculate the inverse of the ratio of all negative evaluation words in the evaluation record to all evaluation words to obtain the second word parameter; Performing a weighted summation on the first word parameter and the second word parameter to obtain a record excellence parameter of the evaluation record; The sum of the record excellence parameters of all the evaluation records in the usage feedback record corresponding to the mirror data is calculated to obtain the evaluation excellence parameter corresponding to the mirror data.

[0018] As an optional embodiment, in the second aspect of the present invention, the push module determines, based on the current user's browsing operation and the image recommendation sequence, a specific method for pushing the display data sequence to the display interface of at least one relevant user for display, including: Eliminate all mirror data whose data quality is lower than a parameter threshold in the mirror recommendation sequence to obtain a sequence after elimination; Determining the current user's usage record emphasis and evaluation emphasis tendency based on the current user's browsing operation; Calculating a first weight proportional to the tendency to attach importance to the usage record; Calculating a second weight proportional to the evaluation emphasis tendency; For each mirror data in the eliminated sequence, calculating a weighted sum of the usage record excellence parameter and the evaluation excellence parameter of the mirror data to obtain a new data quality corresponding to the mirror data; wherein the weighted calculation weight of the usage record excellence parameter is the first weight, and the weighted calculation weight of the evaluation excellence parameter is the second weight; The eliminated sequence is reordered from largest to smallest according to the quality of the new data to obtain a display data sequence, which is pushed to a display interface of at least one relevant user for display.

[0019] As an optional embodiment, in the second aspect of the present invention, the push module determines the specific manner in which the current user's usage record emphasis tendency and evaluation emphasis tendency are determined based on the current user's browsing operation, including: Determine the corresponding cursor movement path and cursor click object in the current user's browsing operation; According to the cursor movement path, a plurality of path passing areas are determined on the current browser page; According to the cursor click object, determining a plurality of clicked objects from a plurality of model objects of the current browser page; Counting the number of first areas belonging to the usage record display area and the number of second areas belonging to the evaluation display area among all areas passed by the path; Counting the number of first objects that are usage record-related objects and the number of second objects that are evaluation-related objects among all the clicked objects; Calculate a first ratio of the number of the first areas to the total number of areas passed by all the paths; Calculate a second ratio of the number of the second areas to the total number of areas; Calculate a third ratio of the number of the first objects to the total number of all clicked objects; Calculate a fourth ratio of the second number of objects to the total number of objects; Calculating the product of the first quantity ratio and the third quantity ratio to obtain the current user's usage record emphasis tendency; The product of the second quantity proportion and the fourth quantity proportion is calculated to obtain the evaluation emphasis tendency of the current user.

[0020] A third aspect of the present invention discloses another image data processing system based on evaluation prediction, the system comprising: a memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute part or all of the steps in the image data processing method based on evaluation prediction disclosed in the first aspect of the present invention.

[0021] A fourth aspect of the present invention discloses a computer storage medium storing computer instructions. When the computer instructions are called, they are used to execute some or all of the steps in the image data processing method based on evaluation prediction disclosed in the first aspect of the present invention.

[0022] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: The present invention can calculate the data quality corresponding to each mirror data according to the recording parameters, and sort and display it based on this, showing users more reasonably sorted mirror data recommendations, thereby significantly improving the rationality and management efficiency of mirror data evaluation, recommending higher quality mirror data to users, and further optimizing the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] 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.

[0024] Figure 1 This is a flowchart of a mirror data processing method based on evaluation prediction disclosed in an embodiment of the present invention.

[0025] Figure 2 This is a structural diagram of a mirror data processing system based on evaluation prediction disclosed in an embodiment of the present invention.

[0026] Figure 3 It is a structural diagram of another mirror data processing system based on evaluation prediction disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] 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 accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0028] The terms "first," "second," and so on, in the description and claims of the present invention and the accompanying drawings are used to distinguish between different objects, not to describe a specific order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product, or device comprising a series of steps or elements is not limited to the listed steps or elements but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or device.

[0029] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0030] This invention discloses a mirror data processing method and system based on evaluation and prediction. This method calculates the data quality of each mirror data item based on recorded parameters, sorts and displays the data based on the data quality, and presents users with more rationally ranked mirror data recommendations. This significantly improves the rationality and management efficiency of mirror data evaluation, recommends higher-quality mirror data to users, and further optimizes the user experience. Details are provided below.

[0031] Example 1 See also Figure 1 , Figure 1 This is a flow chart of a mirror data processing method based on evaluation prediction disclosed in an embodiment of the present invention. Figure 1 The mirror data processing method based on evaluation prediction described above can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 1 As shown, the mirror data processing method based on evaluation prediction may include the following operations: 101. Obtain multiple mirror data and corresponding recording parameters.

[0032] 102. Calculate data quality corresponding to each mirror data according to recording parameters corresponding to each mirror data. 103. Sort part or all of the image data from high to low according to data quality to obtain an image recommendation sequence. 104. Push the image recommendation sequence to a display interface of at least one relevant user for display.

[0033] It can be seen that the above-mentioned embodiment of the invention can calculate the data quality corresponding to each mirror data according to the recording parameters, and sort and display it based on this, showing the user a more reasonably sorted mirror data recommendation, thereby significantly improving the rationality and management efficiency of the mirror data evaluation, recommending higher quality mirror data to the user, and further optimizing the user experience.

[0034] As an optional embodiment, in the above steps, the recorded parameters include historical usage records and usage feedback records; historical usage records include user information, usage scenario details, usage prediction data, parameters of the device used, and performance records during use.

[0035] It can be seen that through the above optional embodiments, the content of the recording parameters is limited to more comprehensively and accurately characterize the characteristics of the image data being used, assist in improving the rationality of image data evaluation and management efficiency, recommend higher-quality image data to users, and improve user experience.

[0036] As an optional embodiment, in the above steps, the feedback records used include an assessment of data adequacy, an evaluation of the algorithm architecture, an assessment of the algorithm training level, an analysis of the algorithm training plan, and feedback on the data labeling level.

[0037] It can be seen that through the above optional embodiments, the content of the feedback record is limited to more comprehensively and accurately represent the user's evaluation of different aspects of the mirror data, so as to achieve accurate evaluation of the excellence of the mirror data in the future, assist in improving the rationality of the evaluation and management efficiency of the mirror data, recommend higher-quality mirror data to users, and improve the user experience.

[0038] As an optional embodiment, in the above step, calculating the data quality corresponding to each mirror data according to the recording parameters corresponding to each mirror data includes: For each mirror data, multiple sets of reused records are filtered out from its corresponding historical usage records; optionally, each set of reused records includes the same historical usage records of multiple users whose number exceeds a first threshold; Based on the set of repeated use records, a neural network algorithm is used to calculate the excellent parameters of the use records corresponding to the mirror data; According to the usage feedback records corresponding to the mirror data, a neural network algorithm is used to calculate the evaluation excellence parameters corresponding to the mirror data; The data quality corresponding to the mirror data is obtained by adding the usage record excellent parameters and the evaluation excellent parameters.

[0039] It can be seen that through the above optional embodiments, based on the screening of repeated use records and the calculation of the neural network algorithm, the usage record excellence parameters and evaluation excellence parameters corresponding to the mirror data can be calculated, so as to accurately evaluate the data quality corresponding to the mirror data, facilitate subsequent sorting and recommendation display, and assist in improving the evaluation rationality and management efficiency of the mirror data, recommending higher-quality mirror data to users, and improving the user experience.

[0040] As an optional embodiment, in the above steps, based on the set of reused records, a neural network algorithm is used to calculate the excellent parameters of the usage records corresponding to the mirrored data, including: Calculate the ratio of the number of records in each reused record set to the total number of records in all reused record sets; Inputting each reuse record set into a trained operation stability prediction neural network to obtain an operation stability parameter corresponding to each reuse record set; optionally, the operation stability prediction neural network is trained using a training data set comprising a plurality of training usage records and their corresponding operation stability annotations; The operation stability parameters corresponding to all reused record sets are weightedly summed to obtain the usage record excellence parameter corresponding to the mirror data; wherein the weighted calculation weight of each operation stability parameter is positively correlated with the proportion of the corresponding reused record set.

[0041] It can be seen that through the above optional embodiments, based on the weight calculation of the record quantity ratio and the trained stable operation prediction neural network, the excellent parameters of the usage records corresponding to the mirror data can be fully and accurately calculated, so as to accurately evaluate the data quality corresponding to the mirror data based on this in the future, facilitate subsequent sorting and recommendation display, and assist in improving the rationality of the evaluation and management efficiency of the mirror data, recommending better quality mirror data to users, and improving the user experience.

[0042] As an optional embodiment, in the above steps, using a neural network algorithm to calculate the evaluation excellence parameter corresponding to the mirror data based on the usage feedback record corresponding to the mirror data includes: For each evaluation record in the usage feedback record corresponding to the mirror data, input the evaluation record into the trained LLM model to extract all evaluation terms contained in the evaluation record; According to a preset positive evaluation word database, identifying positive evaluation words among all evaluation words; According to a preset negative evaluation word database, identifying negative evaluation words among all evaluation words; Calculate the ratio of all positive evaluation words in the evaluation record to all evaluation words to obtain the first word parameter; Calculate the inverse of the ratio of all negative evaluation words in the evaluation record to all evaluation words to obtain the second word parameter; Perform weighted summation on the first word parameter and the second word parameter to obtain the record excellence parameter of the evaluation record; The sum of the record excellence parameters of all evaluation records in the usage feedback record corresponding to the mirror data is calculated to obtain the evaluation excellence parameter corresponding to the mirror data.

[0043] It can be seen that through the above optional embodiments, all corresponding evaluation words in the evaluation records can be determined based on the trained LLM model, and positive and negative evaluations can be identified according to the preset word database, so as to calculate the word proportion, so as to fully and accurately calculate the evaluation excellence parameters corresponding to the mirror data, and then accurately evaluate the data quality corresponding to the mirror data based on this, so as to facilitate subsequent sorting and recommendation display, and assist in improving the evaluation rationality and management efficiency of the mirror data, recommending better quality mirror data to users, and improving user experience.

[0044] As an optional embodiment, in the above step, determining a display data sequence based on the current user's browsing operation and the mirror recommendation sequence to push to the display interface of at least one relevant user for display includes: Eliminate all mirror data whose data quality is lower than the parameter threshold in the mirror recommendation sequence to obtain the eliminated sequence; According to the current user's browsing operation, determine the current user's usage record emphasis tendency and evaluation emphasis tendency; Calculate the first weight proportional to the tendency to attach importance to the use record; Calculate the second weight proportional to the evaluation emphasis tendency; For each mirror data in the sequence after elimination, calculate the weighted sum of the usage record excellent parameters and the evaluation excellent parameters of the mirror data to obtain the new data quality corresponding to the mirror data; optionally, the weight of the usage record excellent parameters is the first weight, and the weight of the evaluation excellent parameters is the second weight; The eliminated sequences are reordered from largest to smallest according to the quality of the new data to obtain a display data sequence, which is pushed to a display interface of at least one relevant user for display.

[0045] It can be seen that through the above optional embodiments, low-excellence image data can be eliminated, and then the usage record emphasis tendency and evaluation emphasis tendency can be determined based on the user's browsing operation, so as to re-sort the image recommendation sequence in a more targeted and more suitable way for the current user, thereby improving the evaluation rationality and management efficiency of the image data, recommending to the user image data that is more suitable for his or her preferences and of higher quality, and improving the user experience.

[0046] As an optional embodiment, in the above steps, determining the current user's usage record emphasis tendency and evaluation emphasis tendency based on the current user's browsing operation includes: Determine the corresponding cursor movement path and cursor click object in the current user's browsing operation; According to the cursor movement path, multiple path passing areas are determined on the current browser page; According to the cursor click object, multiple clicked objects are determined from multiple model objects on the current browser page; Counting the number of first areas belonging to the usage record display area and the number of second areas belonging to the evaluation display area among all areas passed by the path; Counting the number of first objects that are usage record-related objects and the number of second objects that are evaluation-related objects among all clicked objects; Calculate the first quantity ratio of the number of the first area to the total number of areas passed by all paths; Calculate the second quantity ratio of the second region quantity to the total region quantity; Calculate a third quantity ratio of the first object quantity to the total number of all clicked objects; Calculate a fourth quantity ratio of the second object quantity to the total object quantity; Calculate the product of the first quantity ratio and the third quantity ratio to obtain the current user's usage record emphasis tendency; The product of the second quantity proportion and the fourth quantity proportion is calculated to obtain the current user's evaluation emphasis tendency.

[0047] It can be seen that through the above optional embodiments, it is possible to calculate usage record-related and evaluation-related attention parameters by analyzing the cursor movement path and cursor click objects of the user's browsing operation, so as to accurately calculate the user's usage record emphasis tendency and evaluation emphasis tendency, thereby facilitating the subsequent reordering of the image recommendation sequence to be more targeted and more suitable for the current user, thereby improving the rationality of image data evaluation and management efficiency, recommending to users image data that is more suitable for their preferences and of higher quality, and improving the user experience.

[0048] Example 2 See also Figure 2 , Figure 2 This is a schematic diagram of the structure of a mirror data processing system based on evaluation prediction disclosed in an embodiment of the present invention. Figure 2 The mirror data processing system based on evaluation prediction described above can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 2 As shown, the mirror data processing system based on evaluation prediction may include: The acquisition module 201 is used to acquire a plurality of mirror data and corresponding recording parameters.

[0049] The calculation module 202 is configured to calculate the data quality corresponding to each mirror data according to the recording parameters corresponding to each mirror data. The sorting module 203 is configured to sort part or all of the image data from high to low according to data quality to obtain an image recommendation sequence. The push module 204 is configured to push the image recommendation sequence to a display interface of at least one relevant user for display.

[0050] It can be seen that the above-mentioned embodiment of the invention can calculate the data quality corresponding to each mirror data according to the recording parameters, and sort and display it based on this, showing the user a more reasonably sorted mirror data recommendation, thereby significantly improving the rationality and management efficiency of the mirror data evaluation, recommending higher quality mirror data to the user, and further optimizing the user experience.

[0051] As an optional embodiment, the recorded parameters include historical usage records and usage feedback records; the historical usage records include user information, usage scenario details, usage prediction data, parameters of the device used, and performance records during use.

[0052] It can be seen that through the above optional embodiments, the content of the recording parameters is limited to more comprehensively and accurately characterize the characteristics of the image data being used, assist in improving the rationality of image data evaluation and management efficiency, recommend higher-quality image data to users, and improve user experience.

[0053] As an optional embodiment, the feedback record includes an assessment of data adequacy, an evaluation of the algorithm architecture, an assessment of the algorithm training level, an analysis of the algorithm training plan, and feedback on the data annotation level.

[0054] It can be seen that through the above optional embodiments, the content of the feedback record is limited to more comprehensively and accurately represent the user's evaluation of different aspects of the mirror data, so as to achieve accurate evaluation of the excellence of the mirror data in the future, assist in improving the rationality of the evaluation and management efficiency of the mirror data, recommend higher-quality mirror data to users, and improve the user experience.

[0055] As an optional embodiment, the specific manner in which the calculation module calculates the data quality corresponding to each mirror data according to the recording parameters corresponding to each mirror data includes: For each mirror data, multiple sets of reused records are filtered out from its corresponding historical usage records; optionally, each set of reused records includes the same historical usage records of multiple users whose number exceeds a first threshold; Based on the set of repeated use records, a neural network algorithm is used to calculate the excellent parameters of the use records corresponding to the mirror data; According to the usage feedback records corresponding to the mirror data, a neural network algorithm is used to calculate the evaluation excellence parameters corresponding to the mirror data; The data quality corresponding to the mirror data is obtained by adding the usage record excellent parameters and the evaluation excellent parameters.

[0056] It can be seen that through the above optional embodiments, based on the screening of repeated use records and the calculation of the neural network algorithm, the usage record excellence parameters and evaluation excellence parameters corresponding to the mirror data can be calculated, so as to accurately evaluate the data quality corresponding to the mirror data, facilitate subsequent sorting and recommendation display, and assist in improving the evaluation rationality and management efficiency of the mirror data, recommending higher-quality mirror data to users, and improving the user experience.

[0057] As an optional embodiment, the calculation module uses a neural network algorithm based on the reuse record set to calculate the specific method of the excellent parameters of the usage record corresponding to the mirror data, including: Calculate the ratio of the number of records in each reused record set to the total number of records in all reused record sets; Inputting each reuse record set into a trained operation stability prediction neural network to obtain an operation stability parameter corresponding to each reuse record set; optionally, the operation stability prediction neural network is trained using a training data set comprising a plurality of training usage records and their corresponding operation stability annotations; The operation stability parameters corresponding to all reused record sets are weightedly summed to obtain the usage record excellence parameter corresponding to the mirror data; wherein the weighted calculation weight of each operation stability parameter is positively correlated with the proportion of the corresponding reused record set.

[0058] It can be seen that through the above optional embodiments, based on the weight calculation of the record quantity ratio and the trained stable operation prediction neural network, the excellent parameters of the usage records corresponding to the mirror data can be fully and accurately calculated, so as to accurately evaluate the data quality corresponding to the mirror data based on this in the future, facilitate subsequent sorting and recommendation display, and assist in improving the rationality of the evaluation and management efficiency of the mirror data, recommending better quality mirror data to users, and improving the user experience.

[0059] As an optional embodiment, the specific method of calculating the excellent evaluation parameters corresponding to the mirror data by the calculation module using a neural network algorithm based on the usage feedback record corresponding to the mirror data includes: For each evaluation record in the usage feedback record corresponding to the mirror data, input the evaluation record into the trained LLM model to extract all evaluation terms contained in the evaluation record; According to a preset positive evaluation word database, identifying positive evaluation words among all evaluation words; According to a preset negative evaluation word database, identifying negative evaluation words among all evaluation words; Calculate the ratio of all positive evaluation words in the evaluation record to all evaluation words to obtain the first word parameter; Calculate the inverse of the ratio of all negative evaluation words in the evaluation record to all evaluation words to obtain the second word parameter; Perform weighted summation on the first word parameter and the second word parameter to obtain the record excellence parameter of the evaluation record; The sum of the record excellence parameters of all evaluation records in the usage feedback record corresponding to the mirror data is calculated to obtain the evaluation excellence parameter corresponding to the mirror data.

[0060] It can be seen that through the above optional embodiments, all corresponding evaluation words in the evaluation records can be determined based on the trained LLM model, and positive and negative evaluations can be identified according to the preset word database, so as to calculate the word proportion, so as to fully and accurately calculate the evaluation excellence parameters corresponding to the mirror data, and then accurately evaluate the data quality corresponding to the mirror data based on this, so as to facilitate subsequent sorting and recommendation display, and assist in improving the evaluation rationality and management efficiency of the mirror data, recommending better quality mirror data to users, and improving user experience.

[0061] As an optional embodiment, the push module determines a specific method of pushing a display data sequence to a display interface of at least one relevant user for display based on the current user's browsing operation and the mirror recommendation sequence, including: Eliminate all mirror data whose data quality is lower than the parameter threshold in the mirror recommendation sequence to obtain the eliminated sequence; According to the current user's browsing operation, determine the current user's usage record emphasis tendency and evaluation emphasis tendency; Calculate the first weight proportional to the tendency to attach importance to the use record; Calculate the second weight proportional to the evaluation emphasis tendency; For each mirror data in the sequence after elimination, calculate the weighted sum of the usage record excellent parameters and the evaluation excellent parameters of the mirror data to obtain the new data quality corresponding to the mirror data; optionally, the weight of the usage record excellent parameters is the first weight, and the weight of the evaluation excellent parameters is the second weight; The eliminated sequences are reordered from largest to smallest according to the quality of the new data to obtain a display data sequence, which is pushed to a display interface of at least one relevant user for display.

[0062] It can be seen that through the above optional embodiments, low-excellence image data can be eliminated, and then the usage record emphasis tendency and evaluation emphasis tendency can be determined based on the user's browsing operation, so as to re-sort the image recommendation sequence in a more targeted and more suitable way for the current user, thereby improving the evaluation rationality and management efficiency of the image data, recommending to the user image data that is more suitable for his or her preferences and of higher quality, and improving the user experience.

[0063] As an optional embodiment, the push module determines the specific manner in which the current user attaches importance to usage records and evaluations based on the current user's browsing operation, including: Determine the corresponding cursor movement path and cursor click object in the current user's browsing operation; According to the cursor movement path, multiple path passing areas are determined on the current browser page; According to the cursor click object, multiple clicked objects are determined from multiple model objects on the current browser page; Counting the number of first areas belonging to the usage record display area and the number of second areas belonging to the evaluation display area among all areas passed by the path; Counting the number of first objects that are usage record-related objects and the number of second objects that are evaluation-related objects among all clicked objects; Calculate the first quantity ratio of the number of the first area to the total number of areas passed by all paths; Calculate the second quantity ratio of the second region quantity to the total region quantity; Calculate a third quantity ratio of the first object quantity to the total number of all clicked objects; Calculate a fourth quantity ratio of the second object quantity to the total object quantity; Calculate the product of the first quantity ratio and the third quantity ratio to obtain the current user's usage record emphasis tendency; The product of the second quantity proportion and the fourth quantity proportion is calculated to obtain the current user's evaluation emphasis tendency.

[0064] It can be seen that through the above optional embodiments, it is possible to calculate usage record-related and evaluation-related attention parameters by analyzing the cursor movement path and cursor click objects of the user's browsing operation, so as to accurately calculate the user's usage record emphasis tendency and evaluation emphasis tendency, thereby facilitating the subsequent reordering of the image recommendation sequence to be more targeted and more suitable for the current user, thereby improving the rationality of image data evaluation and management efficiency, recommending to users image data that is more suitable for their preferences and of higher quality, and improving the user experience.

[0065] Example 3 See also Figure 3 , Figure 3 This is another mirror data processing system based on evaluation prediction disclosed in an embodiment of the present invention. Figure 3 The mirror data processing system based on evaluation prediction is applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 3 As shown, the mirror data processing system based on evaluation prediction may include: A memory 301 storing executable program code; a processor 302 coupled to the memory 301; The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the image data processing method based on evaluation prediction described in the first embodiment.

[0066] Example 4 An embodiment of the present invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of the image data processing method based on evaluation prediction described in the first embodiment.

[0067] Example 5 An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute the steps of the image data processing method based on evaluation prediction described in the first embodiment.

[0068] The foregoing description of specific embodiments of the present disclosure is intended to illustrate a method for performing a multi-tasking process. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0069] The systems, devices, modules, or units described in the above embodiments may be implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0070] For the convenience of description, the above devices are described as being divided into various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0071] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the embodiments of this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0072] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0073] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0074] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0075] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0076] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0077] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0078] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0079] This specification may be described in the general context of computer-executable instructions, such as program modules, executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media, including storage devices.

[0080] The various embodiments in this specification are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the description of the method embodiments.

[0081] Finally, it should be noted that the mirror data processing method and system based on evaluation prediction disclosed in the embodiments of the present invention are only preferred embodiments of the present invention, and are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A mirror data processing method based on evaluation prediction, characterized in that: The method comprises: Obtain multiple image data and corresponding recording parameters; Calculating the data quality corresponding to each mirror data according to the recording parameters corresponding to each mirror data; Sort part or all of the mirror data from high to low according to the data quality to obtain a mirror recommendation sequence; The mirror recommendation sequence is pushed to a display interface of at least one relevant user for display.

2. The image data processing method based on evaluation prediction according to claim 1, characterized in that: The record parameters include historical usage records and usage feedback records; the historical usage records include user information, usage scenario details, usage prediction data, parameters of the device used, and performance records during use.

3. The image data processing method based on evaluation prediction according to claim 2, characterized in that: The usage feedback record includes an assessment of data adequacy, an evaluation of the algorithm architecture, an assessment of the degree of algorithm training, an analysis of the algorithm training plan, and feedback on the degree of data labeling.

4. The image data processing method based on evaluation prediction according to claim 3, characterized in that: The calculating the data quality corresponding to each mirror data according to the recording parameter corresponding to each mirror data includes: For each mirror data, a plurality of reuse record sets are screened from the corresponding historical usage records; each of the reuse record sets includes the same historical usage records recorded by multiple users, the number of which exceeds a first threshold; Based on the set of reused records, a neural network algorithm is used to calculate the excellent parameters of the usage records corresponding to the mirror data; Calculating the evaluation excellence parameter corresponding to the mirror data using a neural network algorithm based on the usage feedback record corresponding to the mirror data; The usage record excellence parameter and the evaluation excellence parameter are added together to obtain the data quality corresponding to the mirror data.

5. The mirror data processing method based on evaluation prediction according to claim 4, characterized in that: The method of calculating the excellent parameters of the usage records corresponding to the mirror data based on the set of reused records using a neural network algorithm includes: Calculating the ratio of the number of records in each of the reused record sets to the total number of records in all the reused record sets; Inputting each of the reuse record sets into a trained operation stability prediction neural network to obtain an operation stability parameter corresponding to each of the reuse record sets; the operation stability prediction neural network is trained using a training data set comprising a plurality of training usage records and their corresponding operation stability annotations; The operation stability parameters corresponding to all the reuse record sets are weightedly summed to obtain the usage record excellence parameter corresponding to the mirror data; wherein the weighted calculation weight of each operation stability parameter is positively correlated with the proportion of the corresponding reuse record set.

6. The image data processing method based on evaluation prediction according to claim 4, characterized in that: The method of calculating the evaluation excellence parameter corresponding to the mirror data using a neural network algorithm based on the usage feedback record corresponding to the mirror data includes: For each evaluation record in the usage feedback record corresponding to the mirror data, input the evaluation record into the trained LLM model to extract all evaluation terms contained in the evaluation record; Identifying positive evaluation words among all the evaluation words according to a preset positive evaluation word database; Identifying negative evaluation words among all the evaluation words according to a preset negative evaluation word database; Calculate the ratio of all positive evaluation words in the evaluation record to all evaluation words to obtain the first word parameter; Calculate the inverse of the ratio of all negative evaluation words in the evaluation record to all evaluation words to obtain the second word parameter; Performing a weighted summation on the first word parameter and the second word parameter to obtain a record excellence parameter of the evaluation record; The sum of the record excellence parameters of all the evaluation records in the usage feedback record corresponding to the mirror data is calculated to obtain the evaluation excellence parameter corresponding to the mirror data.

7. The image data processing method based on evaluation prediction according to claim 4, characterized in that: The determining, based on the browsing operation of the current user and the mirror recommendation sequence, of a display data sequence to push to a display interface of at least one relevant user for display, includes: Eliminate all mirror data whose data quality is lower than a parameter threshold in the mirror recommendation sequence to obtain a sequence after elimination; Determining the current user's usage record emphasis and evaluation emphasis tendency based on the current user's browsing operation; Calculating a first weight proportional to the tendency to attach importance to the usage record; Calculating a second weight proportional to the evaluation emphasis tendency; For each mirror data in the eliminated sequence, calculating a weighted sum of the usage record excellence parameter and the evaluation excellence parameter of the mirror data to obtain a new data quality corresponding to the mirror data; wherein the weighted calculation weight of the usage record excellence parameter is the first weight, and the weighted calculation weight of the evaluation excellence parameter is the second weight; The eliminated sequence is reordered from largest to smallest according to the quality of the new data to obtain a display data sequence, which is pushed to a display interface of at least one relevant user for display.

8. The mirror data processing method based on evaluation prediction according to claim 7, characterized in that: The determining, based on the browsing operation of the current user, the usage record emphasis tendency and the evaluation emphasis tendency of the current user includes: Determine the corresponding cursor movement path and cursor click object in the current user's browsing operation; According to the cursor movement path, a plurality of path passing areas are determined on the current browser page; According to the cursor click object, determining a plurality of clicked objects from a plurality of model objects of the current browser page; Counting the number of first areas belonging to the usage record display area and the number of second areas belonging to the evaluation display area among all areas passed by the path; Counting the number of first objects that are usage record-related objects and the number of second objects that are evaluation-related objects among all the clicked objects; Calculate a first ratio of the number of the first areas to the total number of areas passed by all the paths; Calculate a second ratio of the number of the second areas to the total number of areas; Calculate a third ratio of the number of the first objects to the total number of all clicked objects; Calculate a fourth ratio of the second number of objects to the total number of objects; Calculating the product of the first quantity ratio and the third quantity ratio to obtain the current user's usage record emphasis tendency; The product of the second quantity proportion and the fourth quantity proportion is calculated to obtain the evaluation emphasis tendency of the current user.

9. A mirror data processing system based on evaluation prediction, characterized in that: The system comprises: An acquisition module, used to acquire multiple image data and corresponding recording parameters; a calculation module, configured to calculate the data quality corresponding to each mirror data according to the recording parameters corresponding to each mirror data; A sorting module, configured to sort part or all of the image data from high to low according to the data quality to obtain an image recommendation sequence; The push module is used to push the image recommendation sequence to the display interface of at least one relevant user for display.

10. A mirror data processing system based on evaluation prediction, characterized in that: The system comprises: a memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the image data processing method based on evaluation prediction according to any one of claims 1 to 8.

Images