A method, system and terminal for creating a talent database based on clothing design factors

Abstract

The present invention relates to a method, system and terminal for creating a talent database based on clothing design factors, belonging to the field of clothing design, which includes: obtaining the style selection resource databases of each platform; generating the current fashion trend based on the style selection resource databases; generating consumer demands based on the fashion trend; generating talent types and required clothing design factors based on the consumer demands; and selecting the style selection resource databases based on the talent types and required clothing design factors to create a talent resource database. This application has the effect of improving the design efficiency.

Description

Technical Field

[0001] The present invention relates to the field of clothing design, and in particular, to a method, a system, and a terminal for creating a talent database based on clothing design factors. Background Art

[0002] Clothing design is a process of systematically conceiving and creating the style, color, fabric, structure, and decoration of clothing by combining art and technology with the human body as the carrier.

[0003] In the prior art, clothing designers first obtain inspiration from various platforms to form a preliminary design concept, and then conduct market research to understand consumer needs, fashion trends, and competitor products to provide a basis for design. Then, they transform the inspiration and ideas into specific sketches, outlining the style, details, and decorative elements of the clothing.

[0004] In view of the above related technologies, designers need to spend a lot of time and energy in obtaining inspiration and conducting market research on various platforms. At the same time, they cannot accurately capture market demand among numerous platforms, resulting in low design efficiency. Summary of the Invention

[0005] In order to improve design efficiency, the present invention provides a method, a system, and a terminal for creating a talent database based on clothing design factors.

[0006] In a first aspect, the present invention provides a method for creating a talent database based on clothing design factors, adopting the following technical solutions:

[0007] A method for creating a talent database based on clothing design factors includes:

[0008] Obtaining a style selection resource database of each platform;

[0009] Generating the current fashion trend based on the style selection resource database;

[0010] Generating consumer needs based on the fashion trend;

[0011] Generating talent types and required clothing design factors based on consumer needs;

[0012] Selecting the style selection resource database based on the talent types and required clothing design factors to create and form a talent resource database.

[0013] By adopting the above technical solutions, by selecting appropriate talents on various platforms to create a talent database and forming talent database resources, breaking the platform boundaries, designers and users can greatly reduce the time and energy spent on obtaining inspiration and conducting market research on numerous platforms through the talent database resources, and can more accurately capture market demand, thereby improving design efficiency.

[0014] Optionally, based on the talent type and the required clothing design factors, the selection resource database is selected to create a talent resource library including:

[0015] Based on the talent type and the required clothing design factors, the selection resource database is selected to form the preliminary selection talents;

[0016] Generate the source platform of preliminary selection talents based on the selection resource database of each platform;

[0017] Get invitation permission based on the source platform;

[0018] Generate candidates based on invitation permission;

[0019] Sending preset invitation information to the selected experts in turn to obtain feedback time values;

[0020] Select candidates whose feedback time values ​​are within the preset reasonable feedback range as candidate candidates, and define the feedback time values ​​that are not within the preset reasonable feedback range as the slow feedback time value;

[0021] Retrieve the candidate values ​​based on the candidate experts;

[0022] Determine whether the candidate number meets the preset required number;

[0023] If yes, then based on the candidate talents, the selection resource database is retrieved to form a talent resource library;

[0024] If not, a candidate expert is generated based on the candidate value and the feedback slow time value, and the selection resource database is retrieved based on the candidate expert and the candidate expert to form an expert resource library.

[0025] By adopting the above technical solution, the qualified preliminary talents will be invited and their opinions will be solicited. The talents who are invited within a reasonable time frame will be selected as candidates. If the talents exceed the reasonable time frame, a new round of evaluation will be conducted to avoid missing the right candidates. At the same time, each talent pool will have a certain number of people required, and the selection of talents from the candidate talents to form a talent resource pool must also meet the number requirements to avoid excessive resources causing mixed resources or too few for reference.

[0026] Optional methods for selecting preliminary candidates include:

[0027] Based on the type of talent, retrieve the talent's activity status, professional skills and clothing style from the selection resource database;

[0028] Generate influence of experts based on their activity;

[0029] Generate talent suitability based on talent influence and professional skills;

[0030] Generate a demand style based on fashion design factors for the demand.

[0031] Compare the fashion style and the demand style to obtain the style fit degree.

[0032] Generate a match degree of the influencer's clothing based on the influencer suitability and the style fit degree.

[0033] Select influencers from the style selection resource database based on the match degree of the influencer's clothing to form a preliminary selection of influencers.

[0034] By adopting the above technical solution, through the dual limitations of the clothing match degree and the suitability, the preliminary selection of influencers can be screened more refinedly, and more demand - fitting recommendation results can be provided for users according to the formed preliminary selection of influencers, optimizing the user experience.

[0035] Optionally, it further includes steps after generating the influencer suitability:

[0036] Retrieve the historical cooperation performance and industry reputation of the influencers from the influencer resource library.

[0037] Generate a preliminary evaluation result based on the historical cooperation performance.

[0038] Generate the content creation quality based on the preliminary evaluation result.

[0039] Generate a comprehensive evaluation index based on the content creation quality and the industry reputation.

[0040] When the comprehensive evaluation index meets the preset comprehensive evaluation criteria, continue to output the influencer suitability.

[0041] When the comprehensive evaluation index does not meet the preset criteria, retrieve the improvement plan and resource investment situation of the influencers from the influencer resource library.

[0042] Generate an index improvement effect based on the improvement plan and the resource investment situation.

[0043] Adjust and replace the influencer suitability based on the index improvement effect.

[0044] By adopting the above technical solution, by evaluating the influencers in the influencer library, according to their performance, data changes, etc., it is determined whether the influencers in the database need to be improved, ensuring the quality of the influencers in the influencer library, and guaranteeing more high - quality influencer resources for the platform and partners.

[0045] Optionally, the method for generating alternative influencers includes:

[0046] Calculate the difference between the feedback slow time value and the preset reasonable feedback range as the feedback time deviation value.

[0047] Calculate the difference between the alternative number value and the required number value, and use it as the alternative deviation number value;

[0048] Generate a performance evaluation rate based on historical cooperation performance;

[0049] Sort the performance evaluation rate and the feedback time deviation value respectively to obtain the evaluation rate sorting value and the feedback time deviation sorting value;

[0050] Calculate the sum value between the evaluation rate sorting value and the feedback time deviation sorting value, and use it as the sorting comprehensive value;

[0051] Sort in descending order based on the sorting comprehensive value, and select the number of candidate talents corresponding to the alternative deviation number value at the very front of the sorting comprehensive value as the promoted talents;

[0052] Combine the promoted talents and the alternative talents to serve as the backup talents.

[0053] By adopting the above technical solution, screening the backup talents through the performance evaluation rate not only provides more opportunities for the talents, but also brings richer choices for the users. At the same time, it optimizes the resource structure of the talent pool and realizes the efficient matching of talents and user needs.

[0054] Optionally, after selecting the number of candidate talents corresponding to the alternative deviation number value at the very front of the sorting comprehensive value as the promoted talents, it further includes:

[0055] Obtain the input style entered by the designer or user;

[0056] Retrieve the wearing frequency, matching method and comfort level based on the promoted talents;

[0057] Generate a resource style based on the wearing frequency, matching method and comfort level;

[0058] Calculate the similarity score between the resource style of each promoted talent and the input style through a preset style matching algorithm;

[0059] Judge whether the similarity score is less than the preset similarity benchmark score;

[0060] If the similarity score is less than the preset similarity benchmark score, calculate the difference between the similarity score and the similarity benchmark score and use it as the similarity deviation value;

[0061] Select the promoted talents based on the similarity deviation value and the wearing frequency and use them as the new promoted talents;

[0062] If the similarity score is not less than the preset similarity benchmark score, continue to output the promoted talents.

[0063] By adopting the above technical solution, by obtaining the user's input style and optimizing the expert library based on the matching of resource style and input style, it is not only possible to provide users with expert recommendations that are more in line with their style, but also to improve the user experience through personalized customization and decision support.

[0064] Optionally, after the qualified talents are merged with the candidates as the reserve talents, the following are also included:

[0065] Based on the candidate influencers, the initial platform weight ratio and real-time live streaming sales conversion ratio are retrieved;

[0066] Generate a demand weight ratio based on the preset live streaming sales conversion ratio per unit time;

[0067] Determine whether the demand weight ratio is consistent with the preset benchmark weight ratio;

[0068] If they are consistent, no adjustment is made;

[0069] If they are inconsistent, calculate the difference between the demand weight ratio and the platform initial weight ratio and use it as the weight difference;

[0070] Retrieve the candidate values ​​based on the candidate expert;

[0071] Calculate the difference between the candidate number and the required number and use it as the candidate deviation number;

[0072] A supplementary expert is generated based on the weight difference and the candidate deviation value, and the supplementary expert is added to the candidate experts.

[0073] By adopting the above technical solutions, we can monitor and analyze the live broadcast, work release, audience interaction and other data of influencers in real time. We can help users predict the future development potential and sales ability of influencers through the analysis of historical data and trends, and provide more powerful support for users' decision-making by adjusting the weight ratio of influencers on each platform.

[0074] Optionally, also include:

[0075] Generate research directions and research focuses based on input style;

[0076] Generate the research scope based on the research direction and research focus;

[0077] Generate survey data based on survey scope and demand clothing design factors;

[0078] Generate research conclusions based on research data;

[0079] Generate design recommendations based on research conclusions;

[0080] Generate recommendation experts based on the design recommendation direction, and retrieve recommendation experts based on the expert resource library to output recommendations.

[0081] By adopting the above technical solutions, data support and more accurate design suggestions are provided for designers, helping users quickly insight into market trends and consumer demands, ensuring that the recommended influencer resources can accurately hit the current consumption hotspots, and ensuring the maximized effect of content dissemination.

[0082] In a second aspect, the present application provides an influencer library creation system, adopting the following technical solutions:

[0083] An influencer library creation system includes an acquisition module for acquiring the product selection resource databases, invitation permission situations, feedback time values, and input styles of various platforms;

[0084] A memory for storing the program of a method for creating an influencer library based on clothing design factors as described in any one of the first aspects;

[0085] A processor capable of loading and executing the program in the memory.

[0086] In a third aspect, the present application provides a terminal, adopting the following technical solutions:

[0087] A terminal includes a memory and a processor, and a computer program stored on the memory can be loaded and executed by the processor to perform a method for creating an influencer library based on clothing design factors as described in any one of the first aspects above.

[0088] In summary, the present application includes at least one of the following beneficial technical effects:

[0089] By selecting suitable influencers on various platforms to create an influencer library and forming influencer library resources, breaking the platform boundaries, designers and users can greatly reduce the time and energy spent on obtaining inspiration and conducting market research on numerous platforms through the influencer library resources, and can more accurately capture market demands, thereby improving design efficiency;

[0090] By evaluating the influencers in the influencer library and deciding whether the influencers in the database need to be improved based on their performance, data changes, etc., ensuring the quality of the influencers in the influencer library and guaranteeing more high-quality influencer resources for the platform and partners;

[0091] By obtaining the input style of the user and optimizing the influencer library according to the matching of the resource style and the input style, not only can more influencers that match the user's style be recommended to the user, but also the user experience can be enhanced through personalized customization and decision support. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] Figure 1 is a flowchart of a method for creating an influencer library based on clothing design factors according to an embodiment of the present invention;

[0093] Figure 2 It is a flowchart of a method for selecting a style selection resource database based on the influencer type and demand clothing design factors in an embodiment of the present invention to create and form an influencer resource library;

[0094] Figure 3 It is a flowchart of a method for selecting primary influencers in an embodiment of the present invention;

[0095] Figure 4 It is a flowchart of a method in an embodiment of the present invention after generating the influencer suitability;

[0096] Figure 5 It is a flowchart of a method for generating candidate influencers in an embodiment of the present invention;

[0097] Figure 6 It is a flowchart of a method in an embodiment of the present invention after selecting the number of candidate influencers corresponding to the alternative deviation number value at the forefront of the selection sorting comprehensive value as promoted influencers. Detailed implementation manners

[0098] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

[0099] An influencer library creation method based on clothing design factors, through the creation of the influencer library, breaks the platform boundaries, promotes the cooperation and resource integration of influencers between different platforms. It reduces the time for users to collect market demand-related information, more accurately captures market demand, realizes the efficient matching of influencers and user needs, and improves the usage efficiency.

[0100] Refer to Figure 1 , an embodiment of the present invention discloses an influencer library creation method based on clothing design factors, which includes:

[0101] Step S100: Obtain the style selection resource databases of each platform.

[0102] Each platform refers to the source channels for collecting style selection resource data. The style selection resource database refers to the collection of data such as clothing styles, design elements, and fashion trends collected from multiple channels such as the above platforms. The style selection resource database crawls the resource data of relevant influencers on each platform through data and integrates these data into a system for subsequent analysis and use.

[0103] Step S101: Generate the current fashion trend based on the style selection resource database.

[0104] The fashion trend refers to identifying the changing rules of design elements such as clothing styles, colors, materials, and cuts that are currently preferred by consumers by analyzing the data in the style selection resource database. For example, "loose cut", "retro style", "environmentally friendly materials", etc.

[0105] Technicians analyze the pictures, text descriptions, and user behavior data in the style selection resource database, such as usage frequency, dressing styles, clothing styles, etc., and input the analysis results into the system. Then, through the system algorithm, the style selection resource database is matched with the current fashion trends. The specific system algorithm is prior art and will not be elaborated here.

[0106] Step S102: Generate consumer demands based on fashion trends.

[0107] Consumer demands refer to the functionality and aesthetics of clothing that target consumers deduce according to fashion trends. Functionality includes the warmth retention, breathability, waterproofness, etc. of clothing; aesthetics includes the color matching, cutting style, etc. of clothing. Different fashion trends correspond to different consumer demands, and consumer demands are obtained by querying the consumer demand database. The consumer demand database pre-stores a comparison table of consumer demands corresponding to different fashion trends, and the consumer demand database is obtained by operators recording the consumer demands corresponding to different fashion trends.

[0108] Step S103: Generate influencer types and demand clothing design factors based on consumer demands.

[0109] Influencer types refer to the influencer classifications that can effectively promote clothing matched according to consumer demands, such as fashion influencers, sports influencers, etc. Demand clothing design factors refer to the key elements of clothing design extracted according to consumer demands, such as color, material, cutting, pattern, etc. Influencer types and demand clothing design factors are generated by extracting the frequently occurring style features in consumer demands and the influencers wearing these styles. For example, when the consumer demand is a sports demand, the corresponding influencer type is a sports influencer, and the corresponding clothing design factors are breathability, elasticity, etc.; when the consumer demand is fashion, the corresponding influencer type is a fashion influencer, and the corresponding clothing design factors are novel styles, special design elements, etc.

[0110] Step S104: Select the style selection resource database based on the influencer types and demand clothing design factors to create an influencer resource library.

[0111] The influencer resource library refers to the set of final cooperative influencers selected according to influencer types and demand clothing design factors, including the historical cooperation performance and industry reputation of the influencers. The formed influencer resource library can be viewed and referenced by users. Match the relevant content in the database with the influencer types and demand clothing design factors, and retrieve and form the influencer resource library for the successfully matched influencer types. This not only improves the screening efficiency but also can discover influencers that are easily missed by traditional methods, making the influencer recommendations in the influencer resource library more accurate.

[0112] Reference Figure 2 , a method for creating a talent database based on fashion design factors, further includes selecting a product selection resource database according to the talent type and required fashion design factors to create a talent resource database, the following steps:

[0113] Step S200: Select a product selection resource database based on the talent type and required fashion design factors to form preliminary selected talents.

[0114] The preliminary selected talents refer to potential cooperation partners selected from the product selection resource database that meet specific talent types and required fashion design factors. The preliminary selected talents are selected from the product selection resource database through keyword matching, talent types and required fashion design factors. This not only improves efficiency but also makes the summarized data more representative.

[0115] Step S201: Generate the source platforms of the preliminary selected talents based on the product selection resource databases of each platform.

[0116] The source platforms refer to the e-commerce platforms or social media platforms where the preliminary selected talents are located. The source platforms of different preliminary selected talents are different. The source platforms of the preliminary selected talents are obtained by querying a preset source platform database. The source platform database pre-stores a comparison table of different preliminary selected talents and the corresponding source platforms of the preliminary selected talents. By automatically querying and recording the product selection resource databases of different platform information in sequence, the source platform database is formed.

[0117] Step S202: Obtain the invitation permission status based on the source platforms.

[0118] The invitation permission status refers to the permissions and approvals required to invite talents to cooperate on the source platforms, including but not limited to platform rules, talent cooperation policies, etc. These information are obtained through the back-end management of the source platforms to ensure the compliance and effectiveness of the invitation process.

[0119] Step S203: Generate candidate talents based on the invitation permission status.

[0120] The candidate talents refer to the talents who can be further invited to cooperate under the condition of meeting the invitation permission. First, extract those talents from the preliminary selected talents who clearly indicate that they allow the brand party or the cooperation party to conduct cooperation invitations, and automatically filter out those talents who set "do not accept new cooperation invitations" or "only accept invitations from specific brands". Through these factors, finally screen out the candidate talents that meet the invitation permission to form a list of candidate talents that can be further contacted and cooperated. This process ensures that talents who meet the requirements and are willing to cooperate can be found efficiently, thereby improving the success rate and efficiency of cooperation.

[0121] Step S204: Sending preset invitation information in sequence to the selected experts to obtain feedback time values.

[0122] Invitation information refers to the invitation content pre-designed according to the cooperation needs, including project details, cooperation requirements and expected feedback time values, etc. The invitation information is pre-set. The invitation is sent to the selected experts in turn through the automated script, and the feedback time of each expert is recorded. The automated script is an existing technology and will not be described here.

[0123] Step S205: selecting candidates whose feedback time values ​​are within the preset reasonable feedback range as candidate candidates, and defining feedback time values ​​that are not within the preset reasonable feedback range as slow feedback time values.

[0124] Reasonable feedback range refers to the reasonable response time interval set according to industry standards and historical data. Reasonable feedback range is pre-set. Slow feedback time value refers to the feedback time value that exceeds the preset reasonable feedback range. Candidates whose feedback time value is within the reasonable feedback range are selected as candidates, and feedback time values ​​outside this range are defined as slow feedback time value.

[0125] Step S206: Retrieve candidate values ​​based on the candidate experts.

[0126] The candidate value refers to the number of current candidate experts. By counting the candidate experts and using the counted result as the candidate value, we can evaluate whether the cooperation requirements are met.

[0127] Step S207: Determine whether the candidate number value meets the preset required number value.

[0128] The required number of people refers to the minimum number of experts set in the plan. The required number of people is pre-set. By comparing the number of candidates with the required number of people, it is determined whether further screening or additional experts are needed.

[0129] Step S208: If yes, then the selection resource database is retrieved based on the candidate experts to form an expert resource library.

[0130] When the number of candidates meets the required number of talents, there is no need to further screen or supplement the number of talents. The detailed information of the candidate talents can be retrieved from the selection resource database to form a complete talent resource library.

[0131] Step S209: If not, a candidate expert is generated based on the candidate value and the feedback slow time value, and the selection resource database is retrieved based on the candidate expert and the candidate expert to form an expert resource library.

[0132] A candidate talent refers to a supplementary talent selected based on the feedback slow time value when the number of alternative talents is insufficient. When the number of alternative talents does not meet the required number of talents, it is necessary to further screen or supplement the number of talents, dynamically adjust the number of candidate talents according to the difference between the required number of talents and the number of alternative talents, and verify the rationality of the feedback time (such as whether the feedback time value of the candidate talent is within the acceptable range) to ensure that the number of talents after supplementation meets the requirements. Finally, the alternative talents and candidate talents that pass the integration verification are retrieved from the product selection resource database to form a talent resource database.

[0133] Refer to Figure 3 , a method for creating a talent database based on clothing design factors, further includes the following steps for selecting primary candidate talents:

[0134] Step S300: Retrieve the activity status, professional skills, and clothing styles of talents from the product selection resource database based on the talent type.

[0135] The activity status refers to the activity frequency and interaction data of the talent on their platform, such as the frequency of posting content, the audience interaction rate, etc. Professional skills refer to the fields in which the talent is proficient, such as fashion matching, fabric evaluation, etc. Clothing style refers to the types and styles of clothing that the talent usually showcases, such as casual style, Chinese new national style, etc. By retrieving the activity status, professional skills, and clothing styles of the talents corresponding to the talent type from the product selection resource database, it is convenient for subsequent use.

[0136] Step S301: Generate talent influence based on the activity status.

[0137] Talent influence refers to the magnitude of the influence of the talent on their audience, usually measured by the activity status. The activity status includes the frequency of posting content, the number of audiences, the interaction rate (such as likes, comments, shares), etc. Calculate the talent influence through the following formula: Influence = a × number of audiences + b × interaction rate + c × posting frequency, where a, b, and c are weight coefficients, and the specific coefficients are preset by the operator based on historical data and industry standards.

[0138] Step S302: Generate talent suitability based on the talent influence and professional skills.

[0139] Talent suitability refers to the degree value indicating whether a talent is suitable for promoting a specific type of clothing. The talent suitability is obtained by querying the preset talent suitability database. The talent suitability database pre-stores a comparison table of different talent influences and professional skills and their corresponding suitability. The talent suitability database is formed after the operator analyzes and records the influences and professional skills of different talent information in sequence.

[0140] Step S303: Generate the required style based on the required clothing design factors.

[0141] The demand style refers to the clothing style characteristics extracted according to consumer demands, such as "retro style", "sports casual style", "eco-fashion style", etc. The demand style is obtained by querying a preset demand style database. The demand style database pre-stores a comparison table of different demand clothing design factors (such as color, material, cutting, pattern) and the corresponding demand styles, and the demand style database is formed after the operator analyzes and records different demand clothing design factors in turn.

[0142] Step S304: Compare the clothing style and the demand style to obtain the style fit degree.

[0143] The style fit degree refers to the matching degree between the clothing style of the talent and the demand style. The style fit degree is determined by calculating the similarity between the two. First, the clothing style and the demand style are respectively disassembled into basic characteristics such as color, style, material, and pattern, and then the similarity of these characteristics is compared one by one (such as both using black series, both containing floral patterns, etc.), and then weights are assigned according to the importance of different characteristics (such as color weight 30%, style weight 40%), and finally the matching scores of each characteristic are weighted and added to obtain the total score. The higher the total score, the higher the style fit degree.

[0144] Step S305: Generate the talent clothing matching degree based on the talent suitability and the style fit degree.

[0145] The talent clothing matching degree refers to whether the talent is suitable for promoting clothing of a specific style, and is determined by comprehensively considering the talent suitability and the style fit degree. The matching degree calculation formula is as follows: matching degree = m × suitability + n × style fit degree, where m and n are weight coefficients, and the specific coefficients are preset by the operator according to historical data and industry standards.

[0146] Step S306: Select talents from the style selection resource database based on the talent clothing matching degree to form the preliminary selected talents.

[0147] The preliminary selected talents refer to the preliminary selected cooperation objects screened from the style selection resource database. By setting the matching degree, talents with a matching degree higher than the set value are selected as the preliminary selected talents. For example, if the set matching degree is 0.8, talents with a matching degree higher than 0.8 will be selected as the preliminary selected talents. In this embodiment, eligible talents are extracted from the database through an automated screening algorithm to ensure that the quality and quantity of the preliminary selected talents meet the promotion requirements. The automated screening algorithm is a technology that automatically analyzes, matches, and sorts target objects through computer programs, which is a prior art and will not be elaborated here.

[0148] Refer to Figure 4 , a method for creating a talent database based on clothing design factors, further includes the following steps after generating the talent suitability:

[0149] Step S400: retrieve the expert's historical cooperation performance and industry reputation from the expert resource library.

[0150] Historical cooperation performance refers to the comprehensive capabilities, performance quality and reliability records of specific parties (such as suppliers, partners, influencers, etc.) in the past cooperation process. Specifically, it includes the number of projects cooperated, the duration of cooperation, the effect of cooperation, etc. Industry reputation refers to the public evaluation and reputation of a company, individual or brand in a specific industry. Historical cooperation performance and industry reputation technicians achieve this by checking the social media evaluation of influencers, discussions on industry forums, and directly asking brands they have cooperated with.

[0151] Step S401: Generate preliminary evaluation results based on historical cooperation performance.

[0152] The preliminary evaluation results refer to the evaluation conclusions drawn by analyzing the historical cooperation performance of the experts. They include the success rate of cooperation, the completion of the project, whether it is delivered on time, whether the content quality is stable, etc. The preliminary evaluation results are obtained by querying the preset historical cooperation performance database. The historical cooperation performance database pre-stores a comparison table of different expert information and the corresponding historical cooperation performance. The historical cooperation performance database is obtained by the operator analyzing the historical cooperation data of different expert information to determine whether their performance in historical cooperation is satisfactory.

[0153] Step S402: Generate content creation quality based on the preliminary evaluation results.

[0154] The quality of content creation refers to the overall performance of the content output by the influencer, including professionalism, visual effects, and interactive effects. It can be evaluated through multiple indicators, such as the production level of the video, the creativity of the content, and the ability to express in words. You can refer to the content created by the influencer in past collaborations, check its interactive data on social media (such as the number of likes, comments, and shares), as well as indicators such as the viewing time and completion rate of the content, to comprehensively judge the quality of content creation. The method of comprehensive judgment is as follows: Comprehensive judgment score = α×professional score + β×visual effect score + γ×interactive effect score, where α, β, and γ are weight coefficients, and the specific coefficients are pre-set by the operator based on historical data and industry standards. In this way, the quality of influencer content creation can be systematically evaluated.

[0155] Step S403: Generate comprehensive evaluation indicators based on content creation quality and industry reputation.

[0156] The comprehensive evaluation index refers to the index value obtained by comprehensively evaluating the content quality and word-of-mouth. Combining the information on content creation quality and industry word-of-mouth, a comprehensive evaluation index system is established. This system can include content quality scores and word-of-mouth scores. Through weighted average or other statistical methods, these dimensions are integrated into a comprehensive evaluation index to comprehensively reflect the overall performance of the talent. The calculation formula for the comprehensive evaluation index is as follows: Comprehensive evaluation index = δ × Quality score + ε × Word-of-mouth score, where δ and ε are weight coefficients, and the specific coefficients are preset by the operator based on historical data and industry standards.

[0157] Step S404: When the comprehensive evaluation index meets the preset comprehensive evaluation criteria, continue to output the talent suitability.

[0158] If the comprehensive evaluation index of the talent reaches the preset standard, it indicates that the talent has high applicability. At this time, the suitability of the talent can be continued to be output, that is, its applicability in a specific cooperation project, providing a reference for subsequent cooperation decisions.

[0159] Step S405: When the comprehensive evaluation index does not meet the preset standard, retrieve the improvement plan and resource investment situation of the talent from the talent resource library.

[0160] The improvement plan refers to the specific strategies and measures formulated to improve the performance of the talent, usually including the analysis results of the deficiencies in the comprehensive evaluation index of the talent, the planning of the improvement direction, and the implementation steps. The resource investment situation refers to the resources willing to be invested to implement these improvement plans, including time, energy, and funds. If the comprehensive evaluation index of the talent does not reach the preset standard, it means that there are some areas that need to be improved. At this time, retrieve the improvement plan of the talent from the talent resource library to understand the strategies and measures it has formulated to improve its performance. At the same time, check the talent's situation in terms of resource investment, such as whether it is willing to invest more time, energy, or funds to improve the cooperation effect.

[0161] Step S406: Generate the index improvement effect based on the improvement plan and resource investment situation.

[0162] The index improvement effect refers to determining whether these improvement measures can effectively improve the comprehensive evaluation index of the talent by evaluating whether the improvement plan of the talent is feasible and whether the resource investment is sufficient. The index improvement effect is obtained by querying the preset improvement effect database. The improvement effect database pre-stores a comparison table of different improvement plans and resource investment situations and the corresponding index improvement effects, and the improvement effect database is formed after the operator analyzes and registers different improvement plans and resource investment situations item by item.

[0163] Step S407: Adjust and replace the talent suitability based on the index improvement effect.

[0164] According to the evaluation results of the improvement effect of the indicators, the suitability of the influencers is adjusted accordingly. Adjust the suitability of the influencers according to the improvement effect of the indicators, and determine whether to continue cooperation by judging whether the improvement is effective.

[0165] Refer to Figure 5 , a method for creating an influencer library based on fashion design factors, further includes the following steps for generating candidate influencers:

[0166] Step S500: Calculate the difference between the slow feedback time value and the preset reasonable feedback range as the feedback time deviation value.

[0167] The reasonable feedback range refers to the reasonable difference between the slow feedback time value and the preset reasonable feedback range, which is a reasonable reply time interval set according to industry standards or historical data, such as within 7 working days. It is obtained by direct calculation.

[0168] Step S501: Calculate the difference between the candidate number value and the required number value as the candidate deviation number value.

[0169] The candidate deviation number value refers to the difference between the candidate number value and the required number value, which is used to evaluate whether additional influencers are needed. Through calculation, it can systematically evaluate whether the current number of candidate influencers meets the requirements and provide a basis for subsequent influencer supplementation or adjustment.

[0170] Step S502: Generate a performance evaluation rate based on historical cooperation performance.

[0171] The performance evaluation rate is an evaluation index calculated by analyzing the historical cooperation data of the influencers, which is used to measure the performance of the influencers in past cooperation. The performance evaluation rate is obtained through calculation, and the calculation formula of the performance evaluation rate is: Performance evaluation rate = Number of successful cooperations / Total number of cooperations × 100%. The number of successful cooperations and the total number of cooperations are obtained by retrieving the influencer resource database.

[0172] By analyzing the historical cooperation data of the influencers and calculating their performance evaluation rates, the objectivity and accuracy of the evaluation are ensured.

[0173] Step S503: Sort the performance evaluation rate and the feedback time deviation value respectively to obtain the evaluation rate sorting value and the feedback time deviation sorting value.

[0174] The evaluation rate sorting value refers to the ranking value corresponding to each influencer after sorting the performance evaluation rates from high to low. The feedback time deviation sorting value refers to the ranking value corresponding to each influencer after sorting the feedback time deviation values from small to large. By using a sorting algorithm to sort these two indicators respectively, the sorting value of each influencer is obtained.

[0175] Step S504: Calculate the sum of the evaluation rate ranking value and the feedback time deviation ranking value as the comprehensive ranking value.

[0176] The comprehensive ranking value refers to the sum of the evaluation rate ranking value and the feedback time deviation ranking value, which is used to comprehensively evaluate the performance and feedback efficiency of the influencers. By summing the evaluation rate ranking value and the feedback time deviation, the two ranking values are integrated into a comprehensive value to ensure the comprehensiveness of the evaluation.

[0177] Step S505: Perform a descending order arrangement based on the comprehensive ranking value, and select the corresponding number of candidate influencers with the highest comprehensive ranking values as the promoted influencers.

[0178] The promoted influencers refer to the influencers who can enter the next judging link through the screening of the above steps. After arranging in descending order according to the comprehensive ranking value, the corresponding number of candidate influencers with the highest comprehensive ranking values are selected. Through the descending order arrangement, it is ensured that the selected promoted influencers have high comprehensive performance and feedback efficiency.

[0179] Step S506: Combine the promoted influencers and the candidate influencers to serve as the reserve influencers.

[0180] By combining the promoted influencers and the candidate influencers, it can be supplemented when the number of candidate influencers is insufficient, ensuring the integrity and flexibility of the influencer resource library.

[0181] Refer to Figure 6 , A method for creating an influencer library based on clothing design factors, further includes the following steps after selecting the corresponding number of candidate influencers with the highest comprehensive ranking values as the promoted influencers:

[0182] Step S600: Obtain the input style entered by the designer or user.

[0183] The input style refers to the expected clothing style of the designer or user through the interface, such as "retro style", "sports casual style", "fashion style", etc. The input style can be obtained through text input boxes or style template selections.

[0184] Step S601: Retrieve the wearing frequency, matching method, and comfort level based on the promoted influencers.

[0185] The wearing frequency refers to the frequency of the promoted influencers wearing clothing of a specific style, which is obtained by retrieving the influencer resource database. The matching method refers to how the promoted influencers match their clothing, such as color matching, dressing style, etc. The comfort level refers to the evaluation of the comfort of the clothing by the promoted influencers, usually obtained through questionnaire surveys or comment analyses. By retrieving these information from the influencer resource database, the comprehensiveness and accuracy of the data are ensured.

[0186] Step S602: Generate a resource style based on wearing frequency, matching style, and comfort level.

[0187] The resource style refers to the clothing style characteristics comprehensively evaluated according to the wearing frequency, matching style, and comfort level of promoted influencers. The weight of the resource style is calculated by the following formula: Resource style weight = e × Wearing frequency + f × Matching style score + g × Comfort level score, where e, f, and g are weight coefficients, and the specific coefficients are preset by the operator based on historical data and industry standards. The wearing frequency reflects the degree of preference of the influencer for a certain style; the matching style reflects the wearing habits of the influencer; the comfort level reflects the acceptance degree of the influencer for the clothing material. By performing linear calculations based on the preset weight coefficients (α, β, γ) corresponding to the wearing frequency, matching style, and comfort level, a unified and quantified resource style weight is finally formed. Then, by inputting the resource style weight into the preset resource style database, the resource style is obtained through matching. Different resource styles correspond to different wearing frequencies, matching styles, and comfort levels. The resource style is obtained by querying the resource style database. The resource style database stores a comparison table of different resource styles in advance, and the resource style database is obtained through pre-input by the operator.

[0188] Step S603: Calculate the similarity score between the resource style of each promoted influencer and the input style through a preset style matching algorithm.

[0189] The similarity score is an evaluation index calculated through the style matching algorithm, which is used to measure the matching degree between the resource style of the promoted influencer and the input style. The style matching algorithm is preset. By splitting both the resource style and the input style into small pieces such as color, style, and material, then comparing the similarity degree of each piece, and then marking the importance according to the similarity degree, and finally adding up the importance of all pieces. The higher the total score, the more similar the two styles are.

[0190] Step S604: Determine whether the similarity score is less than the preset similarity benchmark score.

[0191] The similarity benchmark score refers to the lowest similarity value set according to industry standards or historical data. By comparing the similarity score with the similarity benchmark score, it is judged whether further processing is required.

[0192] Step S605: If the similarity score is less than the preset similarity benchmark score, then calculate the difference between the similarity score and the similarity benchmark score as the similarity deviation value.

[0193] The similarity deviation value refers to the difference between the similarity score and the benchmark score. By calculating the similarity deviation value, the degree of insufficient similarity is evaluated, providing a basis for subsequent adjustment.

[0194] Step S606: Select the promotion expert based on the similarity deviation value and the wearing frequency and use him as the new promotion expert.

[0195] The similarity deviation value and wearing frequency are obtained from the talent resource library, and the comprehensive score is calculated by the following formula: Comprehensive score = p × similarity deviation value + q × wearing frequency, where p and q are weight coefficients, and the specific coefficients are pre-set by the operator based on historical data and industry standards. According to the comprehensive score, the ranking is from low to high, and the lowest-scoring talent is selected as the new promotion talent to ensure the improvement of style matching.

[0196] Step S607: If the similarity score is not less than the preset similarity benchmark score, continue to output the promoted experts.

[0197] When the similarity score is not less than the preset similarity benchmark score, it means that the promoted person is qualified, and these promoted people continue to be output by updating the database.

[0198] A method for creating a talent pool based on clothing design factors, further comprising the following steps after merging the promoted talents with the candidate talents as candidate talents:

[0199] Step S700: Retrieve the initial platform weight ratio and real-time live streaming sales conversion ratio based on the candidate influencers.

[0200] The initial platform weight ratio refers to the ratio of the number of candidate influencers on different platforms at the initial stage. The real-time live streaming sales conversion ratio refers to the sales conversion rate actually promoted by the candidate influencers during the live streaming process, which is obtained through the influencer resource library.

[0201] Step S701: Generate a demand weight ratio based on the live streaming sales conversion ratio in a preset unit time.

[0202] The demand weight ratio refers to the target weight set according to the live broadcast conversion ratio within a preset unit time (such as 24 hours). The demand weight ratio is obtained by querying the preset weight ratio database. The weight ratio database pre-stores a comparison table of different live broadcast conversion ratios per unit time and the corresponding demand weight ratios. The weight ratio database is formed by the operator analyzing and recording the different live broadcast conversion ratios per unit time in sequence. The demand weight ratio is determined through data analysis and model prediction to ensure the scientific nature and feasibility of the target.

[0203] Step S702: Determine whether the demand weight ratio is consistent with a preset reference weight ratio.

[0204] The benchmark weight ratio refers to the reference weight ratio set according to industry standards or historical data. By comparing the demand weight ratio with the benchmark weight ratio, we can determine whether the two are consistent.

[0205] Step S703: If they are consistent, no adjustment is made.

[0206] If the demand weight ratio is consistent with the benchmark weight ratio, it indicates that the current weight setting can meet the promotion demand and no further adjustment is required.

[0207] Step S704: If they are inconsistent, calculate the difference value between the demand weight ratio and the platform initial weight ratio, and take it as the weight difference.

[0208] If the demand weight ratio is inconsistent with the benchmark weight ratio, it indicates that there is a deviation between the current demand weight ratio and the initial weight ratio, and the weights need to be reallocated.

[0209] Step S705: Retrieve the candidate number value based on the candidate influencers.

[0210] The candidate number value refers to the number of current candidate influencers. By counting the number of candidate influencers and taking it as the candidate number value, it is convenient for subsequent use.

[0211] Step S706: Calculate the difference between the candidate number value and the demand number value and take it as the candidate deviation number value.

[0212] The candidate deviation number value refers to the number value corresponding to the situation where the number of candidates does not meet the demand. By calculating the difference between the candidate number value and the demand number value and taking it as the candidate deviation number value, it is convenient for subsequent use.

[0213] Step S707: Generate the added influencers based on the weight difference and the candidate deviation number value, and add the added influencers to the candidate influencers.

[0214] The added influencers refer to the influencers that need to be supplemented determined according to the weight difference and the candidate deviation number value. Adding the added influencers to the candidate influencers ensures that the quantity and quality of the candidate influencers meet the promotion demand.

[0215] A method for creating an influencer database based on clothing design factors further includes the following steps:

[0216] Step S800: Generate a research direction and research focus based on the input style.

[0217] The research direction refers to the direction that needs to be determined before design. The research focus refers to the key points that need to be grasped in the research direction. Different input styles correspond to different research directions and research focuses. The research direction and research focus are obtained by querying the research database. The research database pre-stores a comparison table of different input styles and the corresponding research directions and research focuses, and the research database is obtained by the operator's pre-input.

[0218] Step S801: Generate a research scope based on the research direction and research focus.

[0219] The research scope refers to the specific data range required for conducting research determined according to the research direction and focus. Different research directions and research focuses correspond to different research scopes. The research scope is obtained by querying the research category database. In the research category database, a comparison table of different research directions and research focuses and their corresponding research scopes is pre-stored. The research direction and research focus are obtained by querying the research database.

[0220] Step S802: Generate research data based on the research scope and fashion design factors for demand.

[0221] Research data refers to the specific data collected within the research scope, including consumer preferences, market trends, successful cases, etc. In this embodiment, the research data is collected through data crawling, API calls, and questionnaires, etc., to ensure the accuracy and timeliness of the data.

[0222] Step S803: Generate research conclusions based on the research data.

[0223] Research conclusions refer to the conclusions extracted from the research data. Different research data correspond to different research conclusions. The research conclusions are obtained by querying the research conclusion database. In the research conclusion database, a comparison table of different research conclusions is pre-stored. The research data is obtained through operator input.

[0224] Step S804: Generate design recommendation directions based on the research conclusions.

[0225] Design recommendation directions refer to the reference research directions recommended for users during design based on the research conclusions. Different research conclusions correspond to different design recommendation directions. The design recommendation directions are obtained by querying the design recommendation direction database. In the design recommendation direction database, a comparison table of different recommendation directions is pre-stored. The research conclusions are obtained through operator input.

[0226] Step S805: Generate recommended influencers based on the design recommendation directions, and retrieve the recommended influencers from the influencer resource library for output recommendation.

[0227] Recommended influencers refer to the influencers most suitable for promotion screened from the influencer resource library according to the design direction. By matching the tags of the recommendation direction with the preset tags in the influencer resource library. For example, if the design direction is "Chinese new national style", the recommended influencers include national style influencers and promoters of historical elements. In this embodiment, the recommended influencers are extracted from the influencer resource library through an automated screening algorithm and output through an API interface to ensure the accuracy and efficiency of the recommendation.

[0228] Based on the same inventive concept, an embodiment of the present invention provides a fashion design factor-based system, including:

[0229] An acquisition module, configured to acquire the style selection resource databases, invitation permission statuses, feedback time values, and input styles of each platform;

[0230] A memory, configured to store a program, such as a program of the method for creating a talent database based on clothing design factors as described above;

[0231] A processor, capable of loading and executing the program in the memory.

[0232] Based on the same inventive concept, an embodiment of the present invention provides a terminal, including a memory and a processor, where a computer program stored on the memory can be loaded and executed by the processor to implement the method for creating a talent database based on clothing design factors as described above.

[0233] Those skilled in the art can clearly understand that, for the sake of convenience and brevity of description, only the above division of each functional module is used as an example for illustration. In practical applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0234] The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the concept of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art, several improvements and refinements made without departing from the principle of the present invention should also be regarded as within the protection scope of the present invention.

Claims

1. A method for creating a talent database based on clothing design factors, characterized in that: include: Obtain the selection resource database of each platform; Generate current fashion trends based on the selection resource database; Generate consumer demand based on popular trends; Generate talent types and demand clothing design factors based on consumer demand; Based on the talent type and the required clothing design factors, the selection resource database is selected to create a talent resource library; Based on the talent type and the required clothing design factors, the selection resource database is selected to create a talent resource library including: Based on the talent type and the required clothing design factors, the selection resource database is selected to form the preliminary selection talents; Generate the source platform of preliminary selection talents based on the selection resource database of each platform; Get invitation permission based on the source platform; Generate candidates based on invitation permission; Sending preset invitation information to the selected experts in turn to obtain feedback time values; Select candidates whose feedback time values ​​are within the preset reasonable feedback range as candidate candidates, and define the feedback time values ​​that are not within the preset reasonable feedback range as the slow feedback time value; Retrieve the candidate values ​​based on the candidate experts; Determine whether the candidate number meets the preset required number; If yes, then based on the candidate talents, the selection resource database is retrieved to form a talent resource library; If not, a candidate expert is generated based on the candidate value and the feedback slow time value, and the selection resource database is retrieved based on the candidate expert and the candidate expert to form an expert resource library.

2. A method for creating a talent database based on clothing design factors according to claim 1, characterized in that: The selection method for preliminary candidates includes: Based on the type of talent, retrieve the talent's activity status, professional skills and clothing style from the selection resource database; Generate influence of experts based on their activity; Generate talent suitability based on talent influence and professional skills; Generate demand styles based on demand clothing design factors; Compare the clothing style with the required style to get the style fit; Generate the matching degree of the expert's clothing based on the expert's suitability and style fit; Experts are selected from the selection resource database based on the matching degree of their clothing to form preliminary experts.

3. A method for creating a talent database based on clothing design factors according to claim 2, characterized in that: It also includes the steps after generating the talent suitability: Retrieve the influencer’s historical cooperation performance and industry reputation from the influencer resource library; Generate preliminary assessment results based on historical cooperation performance; Generate content creation quality based on preliminary assessment results; Generate comprehensive evaluation indicators based on content creation quality and industry reputation; When the comprehensive evaluation indicators meet the preset comprehensive evaluation standards, continue to output the talent suitability; When the comprehensive evaluation indicators do not meet the preset standards, the improvement plan and resource investment of the expert are retrieved from the expert resource library; Generate indicators based on improvement plans and resource inputs to improve results; Based on the indicator improvement effect, the suitability of the experts can be adjusted and replaced.

4. A method for creating a talent database based on clothing design factors according to claim 3, characterized in that: Methods for generating candidate experts include: Calculate the difference between the feedback slow time value and the preset feedback reasonable range and use it as the feedback time deviation value; Calculate the difference between the candidate number and the required number and use it as the candidate deviation number; Generate performance evaluation rates based on historical collaboration performance; The performance evaluation rate and the feedback time deviation value are sorted respectively to obtain an evaluation rate ranking value and a feedback time deviation ranking value; Calculate the sum of the evaluation rate ranking value and the feedback time deviation ranking value and use it as the ranking comprehensive value; Arrange in descending order based on the ranking comprehensive value, and select the candidates with the number of candidates corresponding to the candidate deviation value at the front of the ranking comprehensive value as the promoted candidates; The qualified talents will be combined with the reserve talents to form the reserve talents.

5. A method for creating a talent database based on clothing design factors according to claim 4, characterized in that: After the candidates with the number of candidates corresponding to the value of the candidate deviation at the front of the selection ranking comprehensive value are selected as the promoted candidates, the following are also included: Get the input style entered by the designer or user; Based on the promotion experts, the frequency of wearing, matching methods and comfort level are retrieved; Generate resource styles based on wearing frequency, matching methods and comfort level; The similarity score between the resource style of each qualified talent and the input style is calculated through the preset style matching algorithm; Determine whether the similarity score is less than a preset similarity benchmark score; If the similarity score is less than the preset similarity benchmark score, the difference between the similarity score and the similarity benchmark score is calculated and used as the similarity deviation value; Based on the similarity deviation value and wearing frequency, the promotion experts are selected and used as new promotion experts; If the similarity score is not less than the preset similarity benchmark score, the promoted experts will continue to be output.

6. A method for creating a talent database based on clothing design factors according to claim 5, characterized in that: After the qualified talents are merged with the candidates as the reserve talents, the following will also be included: Based on the candidate influencers, the initial platform weight ratio and real-time live streaming sales conversion ratio are retrieved; Generate a demand weight ratio based on the preset live streaming sales conversion ratio per unit time; Determine whether the demand weight ratio is consistent with the preset benchmark weight ratio; If they are consistent, no adjustment is made; If they are inconsistent, calculate the difference between the demand weight ratio and the platform initial weight ratio and use it as the weight difference; Retrieve the candidate values ​​based on the candidate expert; Calculate the difference between the candidate number and the required number and use it as the candidate deviation number; A supplementary expert is generated based on the weight difference and the candidate deviation value, and the supplementary expert is added to the candidate experts.

7. The method for creating a talent database based on clothing design factors according to claim 1, characterized in that: Also includes: Generate research directions and research focuses based on input style; Generate the research scope based on the research direction and research focus; Generate survey data based on survey scope and demand clothing design factors; Generate research conclusions based on research data; Generate design recommendations based on research conclusions; Generate recommendation experts based on the design recommendation direction, and retrieve recommendation experts based on the expert resource library to output recommendations.

8. A talent database creation system based on clothing design factors, characterized in that: include: The acquisition module is used to obtain the selection resource database, invitation permission status, feedback time value and input style of each platform; A memory for storing a program of a method for creating a talent library based on clothing design factors according to any one of claims 1 to 7; The program in the memory can be loaded and executed by the processor.

9. A terminal, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executes a method for creating a talent library based on clothing design factors as described in any one of claims 1 to 7.

Images