Clothing platemaking parameter optimization method and system based on big data

Abstract

The invention is suitable for the technical field of garment plate making, and provides a garment plate making parameter optimization method and system based on big data. The method comprises the following steps: acquiring platemaking parameter data and selling record data of target clothing; constructing a plurality of height and weight sub-data sets; calculating a plurality of distribution probability densities representing height and weight; comparing and screening, and determining target height and weight corresponding to the multiple garment sizes; and carrying out optimization processing on the platemaking parameter data. According to the method, selling record data can be processed, a plurality of height and weight sub-data sets are constructed, a plurality of representative heights and weights are selected, corresponding distribution probability density is calculated, target heights and weights corresponding to a plurality of garment sizes are determined through comparison and screening, and platemaking parameter data are optimized. The pattern making parameters of the target height and weight corresponding to different garment sizes can be optimized, the actual requirements of most customers selecting the corresponding garment sizes can be met, and the wearing experience of the customers is effectively improved.

Description

Technical Field

[0001] This invention belongs to the field of garment pattern making technology, and in particular relates to a method and system for optimizing garment pattern making parameters based on big data. Background Technology

[0002] Pattern making is a technical process that uses professional drafting principles and dimensional data to create paper or electronic patterns based on garment design drawings or sample garment requirements, according to human anatomy and fabric characteristics, to guide garment cutting and sewing. By transforming design concepts into practically operable two-dimensional drawings, the garment can accurately reflect the expected shape, structure, and size in the finished garment stage.

[0003] Precise pattern making can ensure the fit, comfort, and aesthetics of clothing, and achieve size standardization and consistency in mass production.

[0004] In existing technologies, garment pattern making parameters need to be designed based on the standard height and weight corresponding to different garment sizes. However, due to the increasing demand for personalized clothing, the multiple standard heights and weights selected in garment pattern making may not meet the actual needs of customers, affecting their wearing experience. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for optimizing garment pattern making parameters based on big data, aiming to solve the technical problems existing in the prior art mentioned in the background.

[0006] The embodiments of the present invention are implemented as follows: A method for optimizing garment pattern-making parameters based on big data, the method specifically includes the following steps: Identify the target garment and obtain its pattern-making parameters and sales records. The sales record data is effectively filtered, categorized by size, and extracted to determine multiple clothing sizes and obtain multiple height and weight subsets. Statistical analysis was performed on multiple height and weight subsets, multiple representative heights and weights were selected, and the probability density distribution of multiple representative heights and weights was calculated. Based on the multiple distribution probability densities, multiple representative heights and weights are compared and screened to determine the target height and weight corresponding to the multiple clothing sizes. Based on multiple target heights and weights, the plate-making parameter data is optimized to generate optimized parameter data.

[0007] As a further limitation of the technical solution of this embodiment of the invention, the step of determining the target garment and obtaining the pattern-making parameter data and sales record data of the target garment specifically includes the following steps: Receive plate-making optimization requests; The pattern optimization request is used to identify the target garment. Obtain the pattern-making parameter data of the target garment; Obtain the sales record data of the target garment.

[0008] As a further limitation of the technical solution of this invention embodiment, the step of effectively filtering, classifying and extracting data from the sales record data to determine multiple clothing sizes and obtain multiple height and weight subsets specifically includes the following steps: The sales record data is subjected to basic effective filtering to obtain basic effective data; The sales record data is filtered to obtain valid positive reviews. Determine multiple clothing sizes; Based on the multiple clothing sizes, the valid positive review data is categorized according to the corresponding size to obtain multiple categories of valid data. From the multiple valid data sets described above, height and weight data are extracted and organized to obtain multiple height and weight subset datasets.

[0009] As a further limitation of the technical solution of this invention, the step of statistically analyzing multiple height and weight subsets, selecting multiple representative heights and weights, and calculating the probability density distribution of the multiple representative heights and weights specifically includes the following steps: Statistical analysis was performed on multiple height and weight subsets to determine the height and weight of multiple buyers and the corresponding number of purchases. Based on the number of purchases, the height and weight of the buyers are sorted and arranged to obtain multiple sorted data. Organize the data according to the multiple arrangements described above, and select multiple representative height and weight; Based on the multiple height and weight subsets, construct multiple height and weight sample vectors; Based on multiple representative heights and weights, a distribution analysis is performed on multiple height and weight sample vectors to calculate the probability density distribution of multiple representative heights and weights.

[0010] As a further limitation of the technical solution of this embodiment of the invention, the calculation formulas for the plurality of probability densities are as follows: ; in, For clothing sizes The Middle Each clothing size has a probability density function representing the distribution of height and weight. Each represents height and weight. For data dimensions, For bandwidth parameters, This is a sample vector of height and weight. This represents the representative sample vector corresponding to height and weight.

[0011] As a further limitation of the technical solution of this invention, the step of comparing and filtering multiple representative heights and weights according to multiple distribution probability densities to determine the target height and weight corresponding to multiple clothing sizes specifically includes the following steps: Based on the multiple distribution probability densities, multiple representative heights and weights are compared to obtain the density comparison results corresponding to the multiple clothing sizes; Based on multiple density comparison results, multiple target probability densities are selected from multiple distribution probability densities; Based on the probability density of multiple targets, determine the target height and weight corresponding to multiple clothing sizes.

[0012] As a further limitation of the technical solution of this invention, the step of optimizing the pattern-making parameter data based on multiple target heights and weights to generate optimized parameter data specifically includes the following steps: Based on the pattern-making parameter data, determine the pattern-making height and weight corresponding to multiple garment sizes; Based on multiple clothing sizes, the target height and weight are compared with the pattern-making height and weight, and the comparison results are recorded. Based on the corresponding comparison results, determine whether plate-making parameter optimization is necessary; When plate-making parameters need to be optimized, multiple heights and weights to be optimized are selected from multiple plate-making heights and weights, and multiple corresponding optimized heights and weights are selected from multiple target heights and weights. Based on multiple optimized height and weight values, the plate-making parameters corresponding to multiple unoptimized heights and weights in the plate-making parameter data are optimized to generate optimized parameter data.

[0013] A big data-based garment pattern-making parameter optimization system, comprising a garment data acquisition module, a data recording and processing module, a statistical analysis and processing module, a comparison and filtering processing module, and a pattern-making parameter optimization module, wherein: The clothing data acquisition module is used to identify the target clothing and acquire the pattern-making parameter data and sales record data of the target clothing. The data processing module is used to effectively filter, classify, and extract data from the sales record data, determine multiple clothing sizes, and obtain multiple height and weight subsets. The statistical analysis and processing module is used to perform statistical analysis on multiple height and weight subsets, select multiple representative heights and weights, and calculate the probability density distribution of multiple representative heights and weights. The comparison and filtering processing module is used to compare and filter multiple representative heights and weights according to multiple distribution probability densities, and determine the target height and weight corresponding to multiple clothing sizes. The plate-making parameter optimization module is used to optimize the plate-making parameter data based on multiple target heights and weights, and generate optimized parameter data.

[0014] As a further limitation of the technical solution of this embodiment of the invention, the recorded data processing module specifically includes: A basic effective filtering unit is used to perform basic effective filtering on the sales record data to obtain basic effective data; The positive review effective filtering unit is used to filter the sales record data for positive reviews and obtain valid positive review data. Clothing size determination unit, used to determine multiple clothing sizes; The data size classification unit is used to classify the positive review data according to the multiple clothing sizes, and obtain multiple categories of valid data. The data extraction and processing unit is used to extract and process height and weight data from multiple categories of valid data to obtain multiple height and weight subsets.

[0015] As a further limitation of the technical solution of this embodiment of the invention, the statistical analysis and processing module specifically includes: The statistical analysis unit is used to perform statistical analysis on multiple height and weight subsets to determine the height and weight of multiple buyers and the corresponding number of purchases. The sorting and organizing unit is used to sort and organize the height and weight of multiple buyers according to the multiple purchase times, and obtain multiple sorting and organizing data. The height and weight selection unit is used to organize data according to the multiple arrangements and select multiple representative heights and weights. The sample vector construction unit is used to construct multiple height and weight sample vectors based on multiple height and weight subsets. The distribution analysis unit is used to perform distribution analysis on multiple height and weight sample vectors based on multiple representative heights and weights, and to calculate the probability density distribution of multiple representative heights and weights.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention acquires pattern-making parameter data and sales record data of target garments; constructs multiple height and weight subsets; calculates the probability density distribution of multiple representative heights and weights; performs comparative filtering to determine the target height and weight corresponding to multiple garment sizes; and optimizes the pattern-making parameter data. It can process sales record data, construct multiple height and weight subsets, select multiple representative heights and weights, calculate the corresponding probability density distributions, determine the target height and weight corresponding to multiple garment sizes through comparative filtering, and optimize the pattern-making parameter data. This allows the pattern-making parameters for the target height and weight corresponding to different garment sizes to be optimized, meeting the actual needs of the vast majority of customers who choose the corresponding garment sizes, effectively improving the customer's wearing experience. Attached Figure Description

[0017] Figure 1 A flowchart of the garment pattern-making parameter optimization method based on big data provided in an embodiment of the present invention is shown; Figure 2 The diagram illustrates the application architecture of the big data-based garment pattern-making parameter optimization system provided in this embodiment of the invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] Understandably, precise pattern making ensures the fit, comfort, and aesthetics of clothing, achieving dimensional standardization and consistency in mass production. Current garment pattern making requires designing parameters based on standard height and weight corresponding to different clothing sizes. However, due to increasingly personalized demands for clothing, the multiple standard heights and weights selected in pattern making may not meet the actual needs of customers, affecting their wearing experience.

[0020] To address the aforementioned issues, this invention discloses a method and system for optimizing garment pattern-making parameters based on big data. The method involves identifying a target garment and acquiring its pattern-making parameter data and sales record data. The sales record data is effectively filtered, categorized by size, and extracted to determine multiple garment sizes and acquire multiple height-weight subsets. Statistical analysis is performed on these subsets to select representative heights and weights and calculate their probability distribution densities. Based on these probability distribution densities, the representative heights and weights are compared and filtered to determine the target height and weight corresponding to the multiple garment sizes. Finally, the pattern-making parameter data is optimized based on these target heights and weights to generate optimized parameter data. This method processes sales record data, constructs multiple height-weight subsets, selects representative heights and weights, calculates their corresponding probability distribution densities, and determines the target height and weight corresponding to multiple garment sizes through comparison and filtering. The optimization of the pattern-making parameter data ensures that the optimized parameters for the target height and weight corresponding to different garment sizes meet the actual needs of the vast majority of customers choosing the corresponding garment sizes, effectively improving the customer's wearing experience.

[0021] Specifically, Figure 1 The flowchart illustrates a method for optimizing garment pattern-making parameters based on big data, as provided in an embodiment of the present invention.

[0022] In a preferred embodiment of the present invention, a method for optimizing garment pattern-making parameters based on big data specifically includes the following steps: Step S101: Determine the target garment and obtain the pattern-making parameter data and sales record data of the target garment.

[0023] In this embodiment of the invention, a pattern optimization request is received. By performing target identification on the pattern optimization request, the target garment that needs to be optimized for pattern making parameters is determined. Then, the pattern making parameter data of the target garment is obtained, and the online sales records of the target garment are exported to obtain the sales record data of the target garment.

[0024] Specifically, in another preferred embodiment provided by the present invention, determining the target garment and obtaining the pattern-making parameter data and sales record data of the target garment specifically includes the following steps: Receive plate-making optimization requests; The pattern optimization request is used to identify the target garment. Obtain the pattern-making parameter data of the target garment; Obtain the sales record data of the target garment.

[0025] Furthermore, the big data-based garment pattern-making parameter optimization method also includes the following steps: Step S102: Effectively filter, classify and extract the sales record data to determine multiple clothing sizes and obtain multiple height and weight subsets.

[0026] In this embodiment of the invention, basic effective data with evaluation feedback, including the customer's actual height and weight, is extracted from the sales record data through basic effective screening. Then, content recognition is performed on the evaluation feedback in the sales record data to extract positive feedback data. Simultaneously, multiple clothing sizes are determined, and the positive feedback data is categorized according to the multiple clothing sizes to obtain categorized effective data corresponding to the multiple clothing sizes. Afterward, the customer's actual height and weight are extracted and organized from the multiple categorized effective data to obtain a height and weight subset corresponding to the multiple clothing sizes.

[0027] Specifically, in another preferred embodiment provided by the present invention, the step of effectively filtering, classifying and extracting data from the sales record data to determine multiple clothing sizes and obtain multiple height and weight subsets specifically includes the following steps: The sales record data is subjected to basic effective filtering to obtain basic effective data; The sales record data is filtered to obtain valid positive reviews. Determine multiple clothing sizes; Based on the multiple clothing sizes, the valid positive review data is categorized according to the corresponding size to obtain multiple categories of valid data. From the multiple valid data sets described above, height and weight data are extracted and organized to obtain multiple height and weight subset datasets.

[0028] Furthermore, the big data-based garment pattern-making parameter optimization method also includes the following steps: Step S103: Perform statistical analysis on multiple height and weight subsets, select multiple representative heights and weights, and calculate the probability density distribution of multiple representative heights and weights.

[0029] In this embodiment of the invention, statistical analysis is performed on multiple height and weight subsets to determine the height and weight of multiple buyers and their corresponding purchase frequency. Then, based on the purchase frequency, the height and weight of the multiple buyers are arranged and organized to obtain arranged data corresponding to multiple clothing sizes. Subsequently, based on the arranged data and a preset number of representatives, multiple representative heights and weights are selected. Simultaneously, based on the multiple height and weight subsets, height and weight sample vectors corresponding to multiple clothing sizes are constructed. Then, based on the multiple representative heights and weights, distribution analysis is performed on the multiple height and weight sample vectors to calculate the probability density distribution of the multiple representative heights and weights. Specifically, the formula for calculating the multiple probability densities is as follows: ; in, For clothing sizes The Middle Each clothing size has a probability density function representing the distribution of height and weight. Each represents height and weight. For data dimensions, For bandwidth parameters, This is a sample vector of height and weight. This represents the representative sample vector corresponding to height and weight.

[0030] It is understandable that each clothing size corresponds to a representative number of height and weight, and the representative number of height and weight corresponding to each clothing size is the representative number of height and weight in the corresponding height and weight subset arranged from most to least frequent purchases.

[0031] It is understood that, in the embodiments of the present invention, data dimension.

[0032] Specifically, in another preferred embodiment provided by the present invention, the step of statistically analyzing multiple height and weight subsets, selecting multiple representative heights and weights, and calculating the probability density distribution of the multiple representative heights and weights specifically includes the following steps: Statistical analysis was performed on multiple height and weight subsets to determine the height and weight of multiple buyers and the corresponding number of purchases. Based on the number of purchases, the height and weight of the buyers are sorted and arranged to obtain multiple sorted data. Organize the data according to the multiple arrangements described above, and select multiple representative height and weight; Based on the multiple height and weight subsets, construct multiple height and weight sample vectors; Based on multiple representative heights and weights, a distribution analysis is performed on multiple height and weight sample vectors to calculate the probability density distribution of multiple representative heights and weights.

[0033] Furthermore, the big data-based garment pattern-making parameter optimization method also includes the following steps: Step S104: According to the multiple distribution probability densities, compare and filter multiple representative heights and weights to determine the target height and weight corresponding to the multiple clothing sizes.

[0034] In this embodiment of the invention, representative heights and weights belonging to the same clothing size are compared and recorded according to multiple probability distribution densities to obtain density comparison results corresponding to multiple clothing sizes. Based on the multiple density comparison results, target probability densities corresponding to multiple clothing sizes are selected from multiple probability distribution densities of multiple clothing sizes. Then, target heights and weights corresponding to multiple clothing sizes are determined according to multiple target probability densities.

[0035] It is understandable that the target probability density is the probability density with the largest value among the multiple probability densities of the corresponding clothing size.

[0036] Specifically, in another preferred embodiment provided by the present invention, the step of comparing and screening multiple representative heights and weights according to multiple distribution probability densities to determine the target height and weight corresponding to multiple clothing sizes specifically includes the following steps: Based on the multiple distribution probability densities, multiple representative heights and weights are compared to obtain the density comparison results corresponding to the multiple clothing sizes; Based on multiple density comparison results, multiple target probability densities are selected from multiple distribution probability densities; Based on the probability density of multiple targets, determine the target height and weight corresponding to multiple clothing sizes.

[0037] Furthermore, the big data-based garment pattern-making parameter optimization method also includes the following steps: Step S105: Based on multiple target heights and weights, optimize the plate-making parameter data to generate optimized parameter data.

[0038] In this embodiment of the invention, based on pattern-making parameter data, the original pattern-making height and weight corresponding to multiple garment sizes are determined. Based on the multiple garment sizes, multiple target heights and weights are compared with the multiple pattern-making heights and weights, and the comparison results are recorded. Then, based on the comparison results, it is determined whether pattern-making parameter optimization is needed. If it is determined that pattern-making parameter optimization is needed, multiple heights and weights to be optimized are selected from the multiple pattern-making heights and weights, and multiple corresponding optimized heights and weights are selected from the multiple target heights and weights. Then, based on the multiple optimized heights and weights, the multiple heights and weights to be optimized in the pattern-making parameter data are replaced. Finally, using the multiple replaced optimized heights and weights, the pattern-making parameters of the corresponding garment sizes are optimized, and the pattern-making parameter data is optimized and updated to generate optimized parameter data.

[0039] Understandably, comparing multiple target heights and weights with multiple pattern-making heights and weights based on multiple clothing sizes involves comparing the target height and weight of the same clothing size with the pattern-making height and weight to determine if they are the same. Specifically, if the target height and weight for each clothing size are the same as the pattern-making height and weight, then pattern-making parameter optimization is not required; if the target height and weight for at least one clothing size are different from the pattern-making height and weight, then pattern-making parameter optimization is required.

[0040] It is understandable that the height and weight to be optimized are the same as the target height and weight in the template; the height and weight to be optimized are the same as the target height and weight in the template.

[0041] Specifically, in another preferred embodiment provided by the present invention, the step of optimizing the pattern-making parameter data based on multiple target heights and weights to generate optimized parameter data specifically includes the following steps: Based on the pattern-making parameter data, determine the pattern-making height and weight corresponding to multiple garment sizes; Based on multiple clothing sizes, the target height and weight are compared with the pattern-making height and weight, and the comparison results are recorded. Based on the corresponding comparison results, determine whether plate-making parameter optimization is necessary; When plate-making parameters need to be optimized, multiple heights and weights to be optimized are selected from multiple plate-making heights and weights, and multiple corresponding optimized heights and weights are selected from multiple target heights and weights. Based on multiple optimized height and weight values, the plate-making parameters corresponding to multiple unoptimized heights and weights in the plate-making parameter data are optimized to generate optimized parameter data.

[0042] Furthermore, Figure 2The diagram illustrates the application architecture of the big data-based garment pattern-making parameter optimization system provided in this embodiment of the invention.

[0043] Specifically, in another preferred embodiment provided by the present invention, a garment pattern-making parameter optimization system based on big data includes: The clothing data acquisition module 101 is used to identify the target clothing and acquire the pattern-making parameter data and sales record data of the target clothing.

[0044] In this embodiment of the invention, the clothing data acquisition module 101 receives a pattern optimization request, identifies the target clothing that needs to be optimized by performing target identification on the pattern optimization request, then acquires the pattern parameter data of the target clothing, and exports the online sales records of the target clothing to obtain the sales record data of the target clothing.

[0045] The data processing module 102 is used to effectively filter, classify and extract the sales record data, determine multiple clothing sizes, and obtain multiple height and weight subsets.

[0046] In this embodiment of the invention, the data processing module 102 performs basic effective filtering on the sales record data, extracts basic effective data containing evaluation feedback, and the evaluation feedback contains the customer's actual height and weight from the sales record data, then performs content recognition on the evaluation feedback in the sales record data, extracts positive feedback data, and simultaneously determines multiple clothing sizes. According to the multiple clothing sizes, the positive feedback data is classified into corresponding sizes to obtain the category effective data corresponding to the multiple clothing sizes. Then, the actual height and weight of the customer is extracted and organized from the multiple category effective data to obtain the height and weight subset data corresponding to the multiple clothing sizes.

[0047] Specifically, in another preferred embodiment provided by the present invention, the data recording processing module 102 specifically includes: A basic effective filtering unit is used to perform basic effective filtering on the sales record data to obtain basic effective data; The positive review effective filtering unit is used to filter the sales record data for positive reviews and obtain valid positive review data. Clothing size determination unit, used to determine multiple clothing sizes; The data size classification unit is used to classify the positive review data according to the multiple clothing sizes, and obtain multiple categories of valid data. The data extraction and processing unit is used to extract and process height and weight data from multiple categories of valid data to obtain multiple height and weight subsets.

[0048] Furthermore, the big data-based garment pattern-making parameter optimization system also includes: The statistical analysis and processing module 103 is used to perform statistical analysis on multiple height and weight subsets, select multiple representative heights and weights, and calculate the probability density distribution of multiple representative heights and weights.

[0049] In this embodiment of the invention, the statistical analysis and processing module 103 performs statistical analysis on multiple height and weight subsets to determine the height and weight of multiple buyers and their corresponding purchase frequency. Then, based on the purchase frequency, the module arranges the height and weight of the multiple buyers to obtain arranged data corresponding to multiple clothing sizes. Next, based on the arranged data and a preset number of representatives, the module selects the top-ranked representative height and weight. Simultaneously, based on the multiple height and weight subsets, it constructs height and weight sample vectors corresponding to multiple clothing sizes. Then, based on the multiple representative height and weight, it performs distribution analysis on the multiple height and weight sample vectors to calculate the probability density distribution of the multiple representative height and weight. Specifically, the formula for calculating the probability density distribution is as follows:

[0050] in, For clothing sizes The Middle Each clothing size has a probability density function representing the distribution of height and weight. Each represents height and weight. For data dimensions, For bandwidth parameters, This is a sample vector of height and weight. This represents the representative sample vector corresponding to height and weight.

[0051] Specifically, in another preferred embodiment provided by the present invention, the statistical analysis and processing module 103 specifically includes: The statistical analysis unit is used to perform statistical analysis on multiple height and weight subsets to determine the height and weight of multiple buyers and the corresponding number of purchases. The sorting and organizing unit is used to sort and organize the height and weight of multiple buyers according to the multiple purchase times, and obtain multiple sorting and organizing data. The height and weight selection unit is used to organize data according to the multiple arrangements and select multiple representative heights and weights. The sample vector construction unit is used to construct multiple height and weight sample vectors based on multiple height and weight subsets. The distribution analysis unit is used to perform distribution analysis on multiple height and weight sample vectors based on multiple representative heights and weights, and to calculate the probability density distribution of multiple representative heights and weights.

[0052] Furthermore, the big data-based garment pattern-making parameter optimization system also includes: The comparison and filtering processing module 104 is used to compare and filter multiple representative heights and weights according to multiple distribution probability densities, and determine the target height and weight corresponding to multiple clothing sizes.

[0053] In this embodiment of the invention, the comparison and screening processing module 104 compares and records the representative heights and weights of multiple clothing sizes according to multiple distribution probability densities, obtains the density comparison results corresponding to multiple clothing sizes, and then selects the target probability density corresponding to multiple clothing sizes from the multiple distribution probability densities of multiple clothing sizes, and then determines the target height and weight corresponding to multiple clothing sizes according to the multiple target probability densities.

[0054] The plate-making parameter optimization module 105 is used to optimize the plate-making parameter data based on multiple target heights and weights to generate optimized parameter data.

[0055] In this embodiment of the invention, the pattern-making parameter optimization module 105 determines the original pattern-making height and weight corresponding to multiple garment sizes based on the pattern-making parameter data. It then compares multiple target heights and weights with the multiple pattern-making heights and weights based on the multiple garment sizes, records the comparison results, and determines whether pattern-making parameter optimization is needed. If optimization is required, it selects multiple heights and weights to be optimized from the multiple pattern-making heights and weights, and selects multiple corresponding optimized heights and weights from the multiple target heights and weights. Based on these optimized heights and weights, it replaces the multiple heights and weights to be optimized in the pattern-making parameter data. Finally, it uses the replaced optimized heights and weights to optimize the pattern-making parameters of the corresponding garment sizes, and updates the pattern-making parameter data to generate optimized parameter data.

[0056] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for optimizing garment pattern-making parameters based on big data, characterized in that, The method specifically includes the following steps: Identify the target garment and obtain its pattern-making parameters and sales records. The sales record data is effectively filtered, categorized by size, and extracted to determine multiple clothing sizes and obtain multiple height and weight subsets. Statistical analysis was performed on multiple height and weight subsets, multiple representative heights and weights were selected, and the probability density distribution of multiple representative heights and weights was calculated. Based on the multiple distribution probability densities, multiple representative heights and weights are compared and screened to determine the target height and weight corresponding to the multiple clothing sizes. Based on multiple target heights and weights, the plate-making parameter data is optimized to generate optimized parameter data.

2. The method for optimizing garment pattern-making parameters based on big data according to claim 1, characterized in that, The process of identifying the target garment and obtaining its pattern-making parameters and sales records specifically includes the following steps: Receive plate-making optimization requests; The pattern optimization request is used to identify the target garment. Obtain the pattern-making parameter data of the target garment; Obtain the sales record data of the target garment.

3. The method for optimizing garment pattern-making parameters based on big data according to claim 1, characterized in that, The process of effectively filtering, classifying, and extracting data from the sales record data to determine multiple clothing sizes and obtain multiple height and weight subsets specifically includes the following steps: The sales record data is subjected to basic effective filtering to obtain basic effective data; The sales record data is filtered to obtain valid positive reviews. Determine multiple clothing sizes; Based on the multiple clothing sizes, the valid positive review data is categorized according to the corresponding size to obtain multiple categories of valid data. From the multiple valid data sets described above, height and weight data are extracted and organized to obtain multiple height and weight subset datasets.

4. The method for optimizing garment pattern-making parameters based on big data according to claim 1, characterized in that, The statistical analysis of multiple height and weight subsets, the selection of multiple representative heights and weights, and the calculation of the probability density distribution of the multiple representative heights and weights specifically include the following steps: Statistical analysis was performed on multiple height and weight subsets to determine the height and weight of multiple buyers and the corresponding number of purchases. Based on the number of purchases, the height and weight of the buyers are sorted and arranged to obtain multiple sorted data. Organize the data according to the multiple arrangements described above, and select multiple representative height and weight; Based on the multiple height and weight subsets, construct multiple height and weight sample vectors; Based on multiple representative heights and weights, a distribution analysis is performed on multiple height and weight sample vectors to calculate the probability density distribution of multiple representative heights and weights.

5. The method for optimizing garment pattern-making parameters based on big data according to claim 4, characterized in that, The formulas for calculating the probability density of the multiple distributions are as follows: ； in, For clothing sizes The Middle Each clothing size has a probability density function representing the distribution of height and weight. Each represents height and weight. For data dimensions, For bandwidth parameters, This is a sample vector of height and weight. This represents the representative sample vector corresponding to height and weight.

6. The method for optimizing garment pattern-making parameters based on big data according to claim 1, characterized in that, The step of comparing and filtering multiple representative heights and weights according to multiple probability densities to determine the target height and weight corresponding to multiple clothing sizes specifically includes the following steps: Based on the multiple distribution probability densities, multiple representative heights and weights are compared to obtain the density comparison results corresponding to the multiple clothing sizes; Based on multiple density comparison results, multiple target probability densities are selected from multiple distribution probability densities; Based on the probability density of multiple targets, determine the target height and weight corresponding to multiple clothing sizes.

7. The method for optimizing garment pattern-making parameters based on big data according to claim 1, characterized in that, The process of optimizing the pattern-making parameter data based on multiple target heights and weights to generate optimized parameter data specifically includes the following steps: Based on the pattern-making parameter data, determine the pattern-making height and weight corresponding to multiple garment sizes; Based on multiple clothing sizes, the target height and weight are compared with the pattern-making height and weight, and the comparison results are recorded. Based on the corresponding comparison results, determine whether plate-making parameter optimization is necessary; When plate-making parameters need to be optimized, multiple heights and weights to be optimized are selected from multiple plate-making heights and weights, and multiple corresponding optimized heights and weights are selected from multiple target heights and weights. Based on multiple optimized height and weight values, the plate-making parameters corresponding to multiple unoptimized heights and weights in the plate-making parameter data are optimized to generate optimized parameter data.

8. A garment pattern-making parameter optimization system based on big data, characterized in that, The system includes a garment data acquisition module, a data recording and processing module, a statistical analysis and processing module, a comparison and screening processing module, and a pattern-making parameter optimization module, wherein: The clothing data acquisition module is used to identify the target clothing and acquire the pattern-making parameter data and sales record data of the target clothing. The data processing module is used to effectively filter, classify, and extract data from the sales record data, determine multiple clothing sizes, and obtain multiple height and weight subsets. The statistical analysis and processing module is used to perform statistical analysis on multiple height and weight subsets, select multiple representative heights and weights, and calculate the probability density distribution of multiple representative heights and weights. The comparison and filtering processing module is used to compare and filter multiple representative heights and weights according to multiple distribution probability densities, and determine the target height and weight corresponding to multiple clothing sizes. The plate-making parameter optimization module is used to optimize the plate-making parameter data based on multiple target heights and weights, and generate optimized parameter data.

9. The garment pattern-making parameter optimization system based on big data according to claim 8, characterized in that, The recorded data processing module specifically includes: A basic effective filtering unit is used to perform basic effective filtering on the sales record data to obtain basic effective data; The positive review effective filtering unit is used to filter the sales record data for positive reviews and obtain valid positive review data. Clothing size determination unit, used to determine multiple clothing sizes; The data size classification unit is used to classify the positive review data according to the multiple clothing sizes, and obtain multiple categories of valid data. The data extraction and processing unit is used to extract and process height and weight data from multiple categories of valid data to obtain multiple height and weight subsets.

10. The garment pattern-making parameter optimization system based on big data according to claim 8, characterized in that, The statistical analysis and processing module specifically includes: The statistical analysis unit is used to perform statistical analysis on multiple height and weight subsets to determine the height and weight of multiple buyers and the corresponding number of purchases. The sorting and organizing unit is used to sort and organize the height and weight of multiple buyers according to the multiple purchase times, and obtain multiple sorting and organizing data. The height and weight selection unit is used to organize data according to the multiple arrangements and select multiple representative heights and weights. The sample vector construction unit is used to construct multiple height and weight sample vectors based on multiple height and weight subsets. The distribution analysis unit is used to perform distribution analysis on multiple height and weight sample vectors based on multiple representative heights and weights, and to calculate the probability density distribution of multiple representative heights and weights.

Images