Insight analysis method and system for brand costume design tidal current and wind direction

Abstract

The invention relates to the technical field of costume design analysis, and discloses a brand costume design trend wind direction insight analysis method and system, and the system comprises a target customer portrait setting module, a data collection module, a screening and matching module, and a trend wind direction analysis module. Portrayal data of a target customer is set through a target customer portrayal setting module; the data acquisition module is used for acquiring clothing arrival person information data of each clothing arrival person; screening out a target clothing arrival person through a screening and matching module; and finally, performing analysis according to the work data of the target clothing arrival work through a trend and wind direction analysis module to obtain a recommendation index of each work. Screening reference works through the recommendation indexes, and analyzing according to the work data of the reference works to obtain the recommendation indexes of the clothing elements; according to the recommendation indexes of the clothing elements, the favorite degree of the target customer to the clothing elements of the clothing characteristics of the various clothing types can be visually and quickly known, so that clothing products more favored by the target customer are designed, and the sales volume is increased.

Description

Technical Field

[0001] This application relates to the technical field of fashion design analysis, and particularly to a method and system for insight analysis of the fashion trend direction of brand fashion design. Background Art

[0002] With the development of the economy and the improvement of consumers' living standards, consumers' demands for clothing are diverse and will change with time, seasons, and social trends. In order to design clothing that better meets market demands, designers and brands need to capture consumers' preferences and demands.

[0003] Currently, designers and brands generally rely on questionnaires, social media, sales data, and customer feedback. Designers combine the obtained data with their own experience to judge consumers' preferences and demands, and then design clothing based on consumers' preferences and demands.

[0004] However, designers need to rely on their own experience to judge, which is somewhat subjective and not necessarily accurate. If the judgment result has a large deviation, the designed clothing may not meet market demands, possibly leading to a decrease in sales. In addition, if the objects of the questionnaire are not consistent with the brand's user group, the results of the questionnaire do not have reference value, thus affecting the designers' judgment results. If one wants to find interviewees with the same user group as the brand for the questionnaire, it also takes a lot of time. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for insight analysis of the fashion trend direction of brand fashion design to solve the above technical problems: How to accurately judge the fashion trend direction of the brand's user group in clothing design.

[0006] In a first aspect, this application provides a method and system for insight analysis of the fashion trend direction of brand fashion design, adopting the following technical solutions: An insight analysis system for the fashion trend direction of brand fashion design includes: A target customer portrait setting module for setting portrait data of target customers; the portrait data includes several parameters; A data collection module for collecting clothing influencer information data of each clothing influencer; the clothing influencer information data includes portrait data of the audience of the clothing influencer and work data of the clothing influencer; A screening and matching module for analyzing based on the portrait data of the audience of the clothing influencer and the portrait data of the target customer, obtaining a portrait matching index of the portrait data of the audience of the clothing influencer and the portrait data of the target customer, and screening out target clothing influencers according to the portrait matching index; The trend wind direction analysis module is used to analyze according to the work data of the target clothing influencer's works to obtain the recommendation index of each work; screen reference works through the recommendation index, and analyze according to the work data of the reference works to obtain the recommendation index of each clothing element.

[0007] Optionally, the parameters include multiple distribution types; the work data includes image data and release interaction data; the release interaction data includes the number of plays, the number of likes, the number of collections, and the number of comments.

[0008] Optionally, through the formula: Calculate the portrait matching index of the portrait data of the audience of any clothing influencer and the portrait data of the target customer ; Wherein, is any clothing influencer; is a judgment function. When , ; when , ; is the audience volume of this clothing influencer; is the preset audience volume; is the total number of parameters of the portrait data, is the number of any parameter of the portrait data, ; is the number of the portrait data The total number of distribution types of the parameters; is the serial number of any distribution type, ; is the number of the portrait data of the audience of this clothing influencer The serial number of the parameter The percentage of the distribution type in all distribution types of the parameter numbered ; is the number of the portrait data of the target customer The serial number of the parameter The percentage of the distribution type in all distribution types of the parameter numbered ; is the number of the portrait data The weight coefficient of the parameter, .

[0009] Optionally, the screening process of the target clothing influencer is as follows: Compare the portrait matching index of the portrait data of the audience of this clothing influencer and the portrait data of the target customer with the preset portrait matching index , ; When , this clothing influencer is the target clothing influencer; When it is the case, the clothing influencer is a non-target clothing influencer.

[0010] Optionally, the recommendation index of any work of the target clothing influencer is calculated by the formula ; ; wherein, is any work of the target clothing influencer; is the playback volume of this work; is the preset playback volume; is the number of likes of this work; is the number of collections of this work; is the number of keywords in the comments; is the number of comments of this work; is the first recommendation weight coefficient; is the second recommendation weight coefficient; is the third recommendation weight coefficient; .

[0011] Optionally, the screening process of the reference work is as follows; Compare the recommendation index of the work with the preset work recommendation index ; When it is the case, this work is a non-reference work; When it is the case, this work is a reference work.

[0012] Optionally, the process of obtaining the recommendation index of each clothing element includes the following steps: S10: Identify the image data of the reference work one by one, identify the clothing in each image data, and segment and classify the image data according to the clothing type; S20: Identify the clothing features of the image data of each clothing type, and identify the corresponding clothing elements of the clothing features; S30: Analyze through the number of occurrences of the corresponding clothing elements to obtain the recommendation index of the clothing elements.

[0013] Optionally, the recommendation index of the th clothing element of the th clothing type and the th clothing feature is calculated by the formula ; ; wherein, is the number of occurrences of the th clothing type, the th clothing feature, and the th clothing element; For the total number of occurrences of the th clothing feature of the

[0014] An insight analysis method for the trend direction of brand clothing design, the analysis method comprising the following steps: S1: Set the portrait data of the target customers through the target customer portrait setting module; the portrait data includes a number of parameters; S2: Collect the clothing influencer information data of each clothing influencer through the data collection module; the clothing influencer information data includes the portrait data of the clothing influencer's audience and the work data of the clothing influencer; S3: Analyze through the screening and matching module according to the portrait data of the clothing influencer's audience and the portrait data of the target customers, obtain the portrait matching index of the portrait data of the clothing influencer's audience and the portrait data of the target customers, and screen out the target clothing influencers according to the portrait matching index; S4: Analyze through the trend direction analysis module according to the work data of the target clothing influencer's works to obtain the recommendation index of each work; screen out the reference works through the recommendation index, and analyze according to the work data of the reference works to obtain the recommendation index of each clothing element.

[0015] In summary, the present application includes the following beneficial technical effects: (1) The present invention sets the portrait data of the target customers through the target customer portrait setting module; then the data collection module collects the clothing influencer information data of each clothing influencer; then analyzes through the screening and matching module according to the portrait data of the clothing influencer's audience and the portrait data of the target customers, obtains the portrait matching index of the portrait data of the clothing influencer's audience and the portrait data of the target customers, and screens out the target clothing influencers according to the portrait matching index; finally analyzes through the trend direction analysis module according to the work data of the target clothing influencer's works to obtain the recommendation index of each work; screens out the reference works through the recommendation index, and analyzes according to the work data of the reference works to obtain the recommendation index of each clothing element; according to the recommendation index of the clothing element, it is possible to intuitively and quickly understand the degree of preference of the target customers for the clothing elements of the clothing features of each clothing type, so as to design clothing products that are more popular with the target customers and increase sales; (2) The present invention screens out the works that are highly liked by the audience, that is, the target customers, through the recommendation index of the works for subsequent analysis, so as to improve the accuracy of the recommendation index of the clothing elements, and realize that according to the recommendation index of the clothing elements, it is possible to intuitively and quickly understand the degree of preference of the target customers for the clothing elements of the clothing features of each clothing type, so as to design clothing products that are more popular with the target customers and increase sales. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a system module framework diagram of an embodiment of the present invention; Figure 2 It is a method flow diagram of an embodiment of the present invention. Specific implementation manners

[0017] The following details the implementation manners of the present application, and examples of the implementation manners are shown in the accompanying drawings.

[0018] In the description of this specification, the description with reference to the terms "certain implementation manners", "one implementation manner", "some implementation manners", "illustrative implementation manners", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the implementation manner or example are included in at least one implementation manner or example of the present application. In this specification, the schematic descriptions of the above terms do not necessarily refer to the same implementation manner or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more implementation manners or examples.

[0019] The embodiment of the present application discloses an insight analysis system for the trend direction of brand clothing design. Referring to Figure 1 , the insight analysis system includes: A target customer portrait setting module, which is used to set the portrait data of the target customer; the portrait data includes several parameters; A data collection module, which is used to collect the clothing expert information data of each clothing expert; the clothing expert information data includes the portrait data of the audience of the clothing expert and the work data of the clothing expert; A screening and matching module, which is used to analyze based on the portrait data of the audience of the clothing expert and the portrait data of the target customer, obtain the portrait matching index of the portrait data of the audience of each clothing expert and the portrait data of the target customer, and screen out the target clothing experts according to the portrait matching index; A trend direction analysis module, which is used to analyze based on the work data of the works of the target clothing experts to obtain the recommendation index of each work; screen out reference works through the recommendation index, and analyze based on the work data of the reference works to obtain the recommendation index of each clothing element; Through the above technical solution, in this embodiment, the portrait data of the target customer is first set by the target customer portrait setting module; then the data acquisition module acquires the clothing influencer information data of each clothing influencer; then the screening and matching module analyzes according to the portrait data of the clothing influencer's audience and the portrait data of the target customer to obtain the portrait matching index of the portrait data of each clothing influencer's audience and the portrait data of the target customer, and screens out the target clothing influencers according to the portrait matching index; finally, the trend analysis module analyzes according to the work data of the target clothing influencer's works to obtain the recommendation index of each work; screens the reference works through the recommendation index, and analyzes according to the work data of the reference works to obtain the recommendation index of each clothing element; according to the recommendation index of the clothing element, it is possible to intuitively and quickly understand the degree of preference of the target customer for the clothing elements of the clothing characteristics of each clothing type, so as to design clothing products that are more liked by the target customer and increase the sales volume.

[0020] As an implementation manner of the present invention, the parameters include multiple distribution types; the work data includes image data and release interaction data; the release interaction data includes the play volume, the like volume, the collection volume, and the comment volume.

[0021] As an implementation manner of the present invention, through the formula: Calculate the portrait matching index of the portrait data of any clothing influencer's audience and the portrait data of the target customer ; Wherein, is any clothing influencer; is a judgment function. When , ; when , ; is the audience volume of this clothing influencer; is the preset audience volume; is the total number of parameters of the portrait data, is the number of any parameter of the portrait data, ; is the number of the portrait data The total number of distribution types of the parameters; is the serial number of any distribution type, ; is the number of the portrait data of this clothing influencer's audience The serial number of the parameter The percentage of the distribution type in the number of all distribution types of the parameter; is the number of the portrait data of the target customer The serial number of the parameter The percentage of the distribution type in the number Percentage of all distribution types of parameters; Is the number of the portrait data Weight coefficient of the parameter, ; Through the above technical solution, in this embodiment, by Determine whether the audience volume of the fashion influencer exceeds the preset audience volume; if the audience volume of the fashion influencer does not exceed the preset audience volume, it means that the fashion influencer does not have enough audience data for portrait matching index analysis, , so the portrait matching index ; if the audience volume of the fashion influencer exceeds the preset audience volume, it means that the fashion influencer has enough audience data for portrait matching index analysis, ; so the portrait matching index is ; Is the difference between the percentage of the k-th distribution type of the m-th parameter of the portrait data of the audience of the fashion influencer in all distribution types of the m-th parameter and the percentage of the k-th distribution type of the m-th parameter of the portrait data of the target customer in all distribution types of the m-th parameter; Is the cumulative value of the differences between the percentages of each distribution type of the m-th parameter of the portrait data of the audience of the fashion influencer in all distribution types of the m-th parameter and the percentages of each distribution type of the m-th parameter of the portrait data of the target customer in all distribution types of the m-th parameter; The cumulative value of the differences between the percentages of each distribution type of the m-th parameter of the portrait data of the audience of the fashion influencer in all distribution types of the m-th parameter and the percentages of each distribution type of the m-th parameter of the portrait data of the target customer in all distribution types of the m-th parameter The smaller it is, the higher the distribution similarity of each distribution type of the m-th parameter between the portrait data of the audience of the fashion influencer and the portrait data of the target customer; Is the influence degree of the cumulative value of the differences between the percentages of each distribution type of the m-th parameter of the portrait data of the audience of the fashion influencer in all distribution types of the m-th parameter and the percentages of each distribution type of the m-th parameter of the portrait data of the target customer in all distribution types of the m-th parameter on the portrait matching index; Is the cumulative value of the influence degrees of the cumulative values of the differences between the percentages of each distribution type of all parameters in the portrait data of the audience of the fashion influencer and the portrait data of the target customer on the portrait matching index; The smaller it is, the higher the portrait matching index between the portrait data of the audience of the fashion influencer and the portrait data of the target customer; On the contrary, The larger it is, the lower the portrait matching index between the portrait data of the audience of the fashion influencer and the portrait data of the target customer; It should be noted that the percentage of the k-th distribution type of the m-th parameter of the portrait data of the audience of the fashion influencer in all distribution types of the m-th parameter The percentage of the distribution type of the sequence number k of the number m parameter of the portrait data of the target customer in all distribution types of the number m parameter For the prior art, the number of the portrait data The weight coefficient of the parameter Is a preset value, obtained based on experience, and will not be elaborated in the specification

[0022] As an implementation manner of the present invention, the screening process of the target clothing expert is as follows: The portrait matching index of the portrait data of the audience of the clothing expert and the portrait data of the target customer Is compared with the preset portrait matching index ; ; When , this clothing expert is the target clothing expert; When , this clothing expert is a non-target clothing expert; Through the above technical solution, in this embodiment, the portrait matching index of the portrait data of the audience of the clothing expert and the portrait data of the target customer Is compared with the preset portrait matching index ; When , it indicates that the portrait matching index of the portrait data of the audience of the clothing expert and the portrait data of the target customer Is relatively low, so this clothing expert is a non-target clothing expert; when , it indicates that the portrait matching index of the portrait data of the audience of the clothing expert and the portrait data of the target customer Is relatively high, so this clothing expert is the target clothing expert; It should be noted that the preset portrait matching index Is a preset value, obtained based on experience, and will not be elaborated in the specification

[0023] As an implementation manner of the present invention, the recommendation index of any work of the target clothing expert is calculated by the formula ; ; Wherein, Is any work of the target clothing expert; Is the playback volume of this work; Is the preset playback volume; Is the number of likes of this work; Is the number of collections of this work; Is the number of keywords in the comments; Is the number of comments of this work; Is the first recommendation weight coefficient; Is the second recommendation weight coefficient; is the third recommended weight coefficient; ; Through the above technical solution, in this embodiment, by judging whether the playback volume of the work exceeds the preset playback volume, if the playback volume of the work does not exceed the preset playback volume, it means that the work has no recommendation value; ; Recommended index ; If the playback volume of the work exceeds the preset playback volume, it means that the work has recommendation value; , Recommended index ; is the ratio of the number of likes to the playback volume of the work; the ratio of the number of likes to the playback volume of the work The larger it is, the higher the degree of preference of the audience for the content of the work, so the higher the reference value and the higher the recommended index; is the ratio of the number of collections to the playback volume of the work; the ratio of the number of collections to the playback volume of the work The larger it is, the higher the degree of preference of the audience for the content of the work, so the higher the reference value and the higher the recommended index; is the ratio of the number of keywords in the comments to the number of comments; the ratio of the number of keywords in the comments to the number of comments The larger it is, the higher the degree of preference of the audience for the content of the work, so the higher the reference value and the higher the recommended index; On the contrary, the ratio of the number of likes to the playback volume of the work The smaller it is, the lower the degree of preference of the audience for the content of the work, so the lower the reference value and the lower the recommended index; the ratio of the number of collections to the playback volume of the work The smaller it is, the lower the degree of preference of the audience for the content of the work, so the lower the reference value and the lower the recommended index; the ratio of the number of keywords in the comments to the number of comments The smaller it is, the lower the degree of preference of the audience for the content of the work, so the lower the reference value and the lower the recommended index; By screening out the works with a high degree of preference from the audience (target customers) through the recommended index of the works for subsequent analysis, the accuracy of the recommended index of clothing elements can be improved, and it can be realized that according to the recommended index of clothing elements, it is possible to directly and quickly understand the degree of preference of the target customers for the clothing elements of the clothing characteristics of each clothing type, so as to design clothing products that are more popular with the target customers and increase sales volume; It should be noted that the preset playback volume , the first recommended weight coefficient , the second recommended weight coefficient and the third recommended weight coefficient are preset values, obtained based on experience, and will not be elaborated in the specification; It should be noted that the keywords are preset in the system, and obtaining the number of keywords in the work review content according to the work review is the prior art, which will not be elaborated in detail in the specification.

[0024] As an implementation manner of the present invention, the screening process of the reference works is as follows: Compare the recommended index of the work with the preset recommended index of the work ; When , this work is a non-reference work; When , this work is a reference work; Through the above technical solution, in this embodiment, the recommended index of the work is compared with the preset recommended index of the work ; when , it indicates that the recommended index of the work is lower than the preset recommended index of the work , so this work is a non-reference work; when , it indicates that the recommended index of the work is higher than the preset recommended index of the work , so this work is a reference work; It should be noted that the preset recommended index of the work is a preset value, obtained according to experience, and will not be elaborated in detail in the specification.

[0025] As an implementation manner of the present invention, the process of obtaining the recommended index of each clothing element includes the following steps: S10: Identify the image data of the reference works one by one, identify the clothing in each image data, and segment and classify the image data according to the clothing type; S20: Perform clothing feature recognition on the image data of each clothing type, and identify the corresponding clothing elements of the clothing features; S30: Analyze through the occurrence times of the corresponding clothing elements to obtain the recommended index of the clothing elements; Through the above technical solution, in this embodiment, first, the image data of the reference works are identified one by one, the clothing in each image data is identified, and the image data are segmented and classified according to the clothing type; then, the clothing feature recognition is performed on the image data of each clothing type to identify the corresponding clothing elements of the clothing features; finally, the recommended index of the clothing elements is obtained through the analysis of the occurrence times of the corresponding clothing elements; according to the recommended index of the clothing elements, it is possible to intuitively and quickly understand the degree of preference of the target customers for the clothing elements of the clothing features of each clothing type, so as to design clothing products that are more popular with the target customers and increase sales volume; It should be noted that in steps S10 and S20, recognition is performed through the trained convolutional neural network model. The training process is prior art and will not be elaborated in the specification.

[0026] As an embodiment of the present invention, through the formula calculate the recommendation index of the th clothing element of the th clothing feature of the th clothing type ; wherein, is the occurrence times of the th clothing element of the th clothing feature of the th clothing type; is the total occurrence times of the th clothing feature of the th clothing type; Through the above technical solution, in this embodiment, the recommendation index of the th clothing element of the th clothing type and the th clothing feature is calculated through the formula ; ; is the ratio of the occurrence times of the th clothing element of the th clothing feature of the th clothing type to the total occurrence times of the th clothing feature of the th clothing type; the ratio of the occurrence times of the th clothing element of the th clothing feature of the th clothing type to the total occurrence times of the th clothing feature of the th clothing type The larger it is, the higher the preference degree for the th clothing element of the th clothing feature of the th clothing type. Therefore, the recommendation index of the th clothing element of the th clothing feature of the th clothing type is higher; on the contrary, the ratio of the occurrence times of the th clothing element of the th clothing feature of the th clothing type to the total occurrence times of the th clothing feature of the th clothing type The smaller it is, it indicates that the th clothing type's th clothing feature's th clothing element is less liked. Therefore, the th clothing type's th clothing feature's th clothing element's recommendation index is lower; According to the recommendation index of clothing elements, it is possible to intuitively and quickly understand the degree of preference of target customers for the clothing elements of the clothing features of each clothing type, so as to design clothing products that are more liked by target customers and increase sales volume; It should be noted that, for ease of understanding, the following gives examples of clothing types, clothing features, and clothing elements. Clothing types refer to tops, pants, skirts, etc.; Upper clothing features refer to collars, sleeves, etc.; The clothing elements of collars refer to the shapes of various collars.

[0027] The embodiment of the present application also discloses an insight analysis method for the trend direction of brand clothing design, as Figure 2 shown. The analysis method includes the following steps: S1: Set the portrait data of the target customer through the target customer portrait setting module; The portrait data includes several parameters; S2: Collect the clothing expert information data of each clothing expert through the data collection module; The clothing expert information data includes the portrait data of the audience of the clothing expert and the work data of the clothing expert; S3: Analyze according to the portrait data of the audience of the clothing expert and the portrait data of the target customer through the screening and matching module, obtain the portrait matching index of the portrait data of the audience of each clothing expert and the portrait data of the target customer, and screen out the target clothing experts according to the portrait matching index; S4: Analyze according to the work data of the target clothing expert's works through the trend direction analysis module to obtain the recommendation index of each work; Evaluate the recommendation level of the work data of each clothing expert through the recommendation index, and analyze the work data of each recommended level of clothing experts to obtain the recommendation index of each clothing element.

[0028] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. An insight analysis system for the trend direction of brand clothing design, characterized in that, Including: A target customer portrait setting module for setting portrait data of target customers; the portrait data includes a number of parameters; A data collection module for collecting clothing expert information data of each clothing expert; the clothing expert information data includes portrait data of the audience of the clothing expert and work data of the clothing expert; A screening and matching module for analyzing based on the portrait data of the audience of the clothing expert and the portrait data of the target customer, obtaining a portrait matching index between the portrait data of the audience of each clothing expert and the portrait data of the target customer, and screening out target clothing experts according to the portrait matching index; A trend analysis module for analyzing based on the work data of the works of the target clothing experts to obtain a recommendation index for each work; Screening reference works through the recommendation index, and analyzing based on the work data of the reference works to obtain a recommendation index for each clothing element.

2. The insight analysis system for the trend direction of a brand clothing design according to claim 1, wherein The parameters include multiple distribution types; the work data includes image data and release interaction data; the release interaction data includes the number of views, likes, collections, and comments.

3. An insight analysis system for the trend direction of a brand clothing design, as claimed in claim 1, wherein Through the formula: Calculate the portrait matching index of the portrait data of any clothing expert audience and the portrait data of the target customer ; Among them, is any clothing expert; is a judgment function. When , ; when , ; is the audience volume of this clothing expert; is the preset audience volume; is the total number of parameters of the portrait data, is the number of any parameter of the portrait data, ; is the number of the portrait data The total number of distribution types of the parameters; is the serial number of any distribution type, ; is the number of the portrait data of the audience of this clothing expert The serial number of the parameter The percentage of the distribution type in all distribution types of the numbered parameters; is the number of the portrait data of the target customer The serial number of the parameter The percentage of the distribution type in all distribution types of the numbered parameters; is the number of the portrait data The weight coefficient of the parameter, .

4. An insight analysis system for the trend direction of a brand clothing design, according to claim 1, characterized in that The screening process of the target clothing experts is as follows: The portrait matching index of the portrait data of the clothing expert audience and the portrait data of the target customers is compared with the preset portrait matching index , ; When this fashion influencer is the target fashion influencer. When the fashion influencer is a non-target fashion influencer.

5. An insight analysis system for the trend direction of a brand clothing design, as claimed in claim 1, wherein Calculate the recommendation index of any work of the target fashion influencer through the formula ;​ Among them, is any work of the target clothing expert; is the playback volume of this work; is the preset playback volume; is the number of likes for this work; is the number of collections for this work; is the number of keywords in the comments; is the number of comments on this work; is the first recommendation weight coefficient; is the second recommendation weight coefficient; is the third recommendation weight coefficient; .

6. The insight analysis system for the trend direction of a brand clothing design according to claim 1, characterized in that The screening process of the reference works is as follows; Compare the recommended index of the work with the preset recommended index of the work for comparison; When the work is a non-reference work; When this work is a reference work.

7. An insight analysis system for the trend direction of a brand clothing design, according to claim 1, characterized in that The process of obtaining the recommendation index for each clothing element includes the following steps: S10: Identifying the clothing in each image data one by one from the image data of the reference works, and segmenting and classifying the image data according to the clothing type; S20: Conducting clothing feature recognition on the image data of each clothing type to identify the corresponding clothing elements of the clothing features; S30: Analyzing through the occurrence times of the corresponding clothing elements to obtain the recommendation index of the clothing elements.

8. An insight analysis system for the trend direction of a brand clothing design, as described in claim 1, characterized in that Calculate through the formula the th clothing feature of the th clothing element's recommendation index ; Among them, is the th occurrence count of the th clothing feature for the th clothing element; is the total occurrence count of the th clothing feature for the th clothing type.

9. A method for analyzing the trend direction of brand clothing design, applicable to the system for analyzing the trend direction of brand clothing design described in any one of claims 1-8, characterized in that, The analysis method includes the following steps: S1: Setting the portrait data of the target customer through the target customer portrait setting module; the portrait data includes a number of parameters; S2: Collecting the clothing expert information data of each clothing expert through the data collection module; the clothing expert information data includes the portrait data of the audience of the clothing expert and the work data of the clothing expert; S3: Analyzing through the screening and matching module based on the portrait data of the audience of the clothing expert and the portrait data of the target customer, obtaining the portrait matching index between the portrait data of the audience of each clothing expert and the portrait data of the target customer, and screening out the target clothing experts according to the portrait matching index; S4: Analyzing through the trend analysis module based on the work data of the works of the target clothing experts to obtain the recommendation index for each work; screening the reference works through the recommendation index, and analyzing based on the work data of the reference works to obtain the recommendation index for each clothing element.

Images