Design match information recommendation platform based on designer portrait

Abstract

The invention relates to the technical field of intelligent diagnosis, and discloses a design match information recommendation platform based on a designer portrait, which comprises a user information module, a capability model construction module, a match information recommendation module and an interactive community module, and is characterized in that the user information module is used for collecting various demands of a designer; constructing designer data; the ability model building module is used for building a designer portrait through a fuzzy clustering analysis algorithm and multi-dimensional ability feature analysis based on designer data; the information recommendation module uses a label-based collaborative filtering recommendation algorithm to recommend personalized industry information to designers according to designer portraits and a design match information database; the interactive community module provides a communication platform between designers and promotes knowledge sharing and cooperative innovation; according to the method, a collaborative filtering recommendation algorithm based on labels is applied, and the participants are subjected to tagging setting, so that accurate classification is realized, and personalized requirements of designers are deeply met.

Description

Technical Field

[0001] The present invention relates to the field of intelligent diagnosis technology, and in particular to a design competition information recommendation platform based on designer portraits. Background Art

[0002] At present, in the research of recommendation platforms based on content recommendation, the designer portrait mainly focuses on the designer's experience and knowledge background. However, when extracting the designer's ability characteristics, there are problems such as small number of characteristics and rough division. More importantly, the evaluation values ​​of various ability characteristics often rely on subjective judgment and lack sufficient objectivity, which makes it difficult for the existing designer portrait to accurately reflect the real needs of designers in the process of information search; the design of the industry information collaborative recommendation system should take into account both the ability characteristics of designers and the characteristics of the recommended content. Through the relationship between the designer's ability characteristics and the recommended content, it is estimated which design competitions are more in line with the needs of designers at the current stage, and according to the designer's self-needs, the content suitable for him is recommended to the designer. The important research direction is to maximize the personalization of the recommendation system; in the process of innovative design, the content push method is crucial. The recommendation system should comprehensively consider the designer portrait and push relevant content such as competition information, designers in the same industry, and search content of similar designers in stages. In addition, the recommendation system also needs to pay close attention to the changes of designers at different stages, continuously iterate and collaboratively recommend content, and ensure that the pushed information matches the current needs and capabilities of designers. Through this method, the recommendation system can finally determine the personalized content knowledge provided by collaborative recommendation at each stage, so as to more effectively support the innovative design work of designers. Summary of the invention

[0003] 1. Technical issues to be resolved In view of the shortcomings of the existing technology, the present invention provides a design competition information recommendation platform based on designer portraits, which has the advantages of providing customized information recommendation services for each designer by constructing a comprehensive and in-depth designer portrait in combination with industry trends and market demands. It solves the many shortcomings of the existing technology in constructing designer portraits and recommending industry information, such as few portrait indicators, rough division, and lack of objectivity in designer needs.

[0004] (II) Technical solution In order to achieve the above-mentioned purpose of providing customized information recommendation services for each designer by constructing a comprehensive and in-depth designer portrait in combination with industry trends and market demands, the present invention provides the following technical solutions: a design competition information recommendation platform based on designer portraits, comprising a user information module, a capability model construction module, an information recommendation module and an interactive community module, wherein the user information module is used to collect various data of designers and construct designer portraits; the capability model construction module establishes a designer capability model based on the designer portrait data through a fuzzy clustering analysis algorithm and multi-dimensional capability feature analysis; the information recommendation module utilizes a tag-based collaborative filtering recommendation algorithm to recommend personalized industry information to designers according to the designer capability model and the industry information database; the interactive community module provides a communication platform between designers to promote knowledge sharing and collaborative innovation.

[0005] Preferably, the user information module also includes a data preprocessing unit for cleaning, deduplicating and formatting the collected designer information to ensure data quality and accuracy.

[0006] Preferably, the capability model building module also includes a capability evaluation standard library for defining and analyzing the specific meaning and evaluation standards of various capability characteristics of designers to ensure the comprehensiveness and accuracy of capability evaluation.

[0007] Preferably, the information recommendation module also includes a dynamic adjustment unit for dynamically adjusting the recommendation algorithm and parameters according to the designer's behavior data and feedback on the platform to improve the accuracy and personalization of the recommended content.

[0008] Preferably, the interactive community module also includes a topic classification unit and an interactive feedback unit, which are used to classify and manage the communication content of designers, and collect and analyze the interactive feedback of designers to optimize the community atmosphere and user experience.

[0009] Preferably, it also includes a management module for configuring and managing the functional modules of the platform to ensure stable operation and continuous optimization of the platform; the management module also includes a data analysis unit for in-depth mining and analysis of the data on the platform to provide data support for the management decisions of the platform.

[0010] Preferably, the information recommendation module also has cross-platform data integration capabilities, and can integrate data resources from different design crowdsourcing platforms to provide designers with more comprehensive industry information recommendation services.

[0011] Preferably, the interactive community module also has an international perspective, can provide designers with cross-border design exchange opportunities, and promote the sharing and cooperation of international design resources.

[0012] Preferably, the collaborative filtering recommendation algorithm comprises the following steps: S1: Select five designer portraits as basic data; combine the five designer index scores into a set (also called domain), marked as 1, 2, 3, 4, and 5 respectively, then the domain For the object being classified; S2: Select the maximum-minimum method to establish the fuzzy similarity matrix. The calculation formula of the maximum-minimum method is: ,in is the minimum sum of any two rows, is the sum of the maximum values ​​of any two rows, It means Designer and The amount of similarity among designers on the four performance indicators; S3: With the help of MATLAB calculation, the matrix of fuzzy similarity relationships of the four indicators of the five designers was obtained; S4: Cluster analysis of 5 designers using fuzzy equivalence matrix, intercept level From 1 to 0, press Dynamic clustering of the truncated matrix.

[0013] Preferably, the platform adopts a modular design to facilitate independent development and upgrading of each functional module, thereby improving the flexibility and scalability of the platform.

[0014] (III) Beneficial effects Compared with the prior art, the present invention provides a design competition information recommendation platform based on designer portraits, which has the following beneficial effects: 1. This industry information recommendation platform based on designer portraits has introduced a tag-based collaborative filtering recommendation algorithm. By labeling the participating entities, it has achieved accurate classification to deeply meet the personalized needs of designers. By building an indicator portrait of international design competitions, it can recommend more matching design competitions and other content to designers, thereby further improving the user's personalized experience and enhancing the value of the platform. It not only improves the accuracy of information recommendations, but also provides designers with a broader career development space and more abundant opportunities to participate.

[0015] 2. This industry information recommendation platform based on designer portraits, based on the evolution of designer portraits and fuzzy clustering analysis algorithms, deeply explores the fundamental position of designers and related international design competitions in the industry information recommendation platform, and carefully analyzes the needs of various subjects in the industry information recommendation platform. Through in-depth exploration and classification of designer design fields in different design crowdsourcing platforms, as well as literature surveys and statistical surveys, the designer portrait is portrayed. Based on an in-depth study of other designer competency models and combined with the rich experience of individuals in related design fields, the present invention proposes a new designer competency model. The model provides a detailed analysis of the meaning and evaluation criteria of each competency feature, ensuring a comprehensive and accurate assessment of the designer's capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a block diagram of the platform of the present invention; Figure 2 This is a framework diagram of the industry information recommendation platform based on designer portraits of the present invention; Figure 3 This is a block diagram of the steps of the collaborative filtering recommendation algorithm of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] See also Figure 1 and Figure 2 , a design competition information recommendation platform based on designer portraits, including a user information module, a capability model building module, an information recommendation module and an interactive community module. The user information module is used to collect various data of designers and build designer portraits; the capability model building module is based on the designer portrait data, through fuzzy clustering analysis algorithm and multi-dimensional capability feature analysis, to establish a designer capability model; the information recommendation module uses a tag-based collaborative filtering recommendation algorithm to recommend personalized industry information to designers based on the designer capability model and industry information database; the interactive community module provides a communication platform between designers to promote knowledge sharing and collaborative innovation; 1. User information module Functional description: This module is the data foundation of the platform and is responsible for comprehensively and securely collecting various data of designers, including but not limited to basic information (such as name, age, professional background), professional skills, design style preferences, historical project experience, learning experience, social media activities, etc.

[0019] Data processing: Data encryption and anonymization technologies are used to ensure user privacy and security. At the same time, data mining technology is used to extract valuable information to build accurate designer portraits.

[0020] 2. Capability model building module Algorithm application: Based on the data provided by the user information module, this module uses the fuzzy clustering analysis algorithm to group designers and identify groups of designers with similar ability characteristics. At the same time, through multi-dimensional ability characteristic analysis (such as design ability, innovation ability, market sensitivity, etc.), a personalized ability model is built for each designer.

[0021] Model optimization: During the model building process, iterative optimization will be continuously performed based on newly collected data to ensure the accuracy and timeliness of the capability model.

[0022] 3. Information recommendation module Recommendation algorithm: This module adopts a tag-based collaborative filtering recommendation algorithm, combined with the designer's ability model and industry information database (including the latest design trends, technological innovations, market analysis reports, success cases, etc.), to tailor a personalized industry information recommendation list for each designer.

[0023] User experience: Recommended results are presented in an intuitive and easy-to-understand way, such as smart push notifications and personalized homepage displays, ensuring that designers can quickly obtain information that is beneficial to their career development.

[0024] 4. Interactive community module Communication platform: Provide an open and active online community where designers can share their works, exchange ideas, ask questions, and seek cooperation opportunities. The community supports multiple forms of communication such as pictures, texts, videos, and live broadcasts to meet the communication needs of different designers.

[0025] Knowledge sharing and collaborative innovation: Through interaction within the community, designers can learn from each other, jointly solve design problems, and promote innovation in design concepts and technologies. In addition, the community can also serve as an important channel for designers to find partners and expand their network.

[0026] Interactions between modules Data flow: The user information module provides raw data for the capability model building module. The output of the capability model building module is the basis of the information recommendation module, and the interactive community module in turn promotes further enrichment and updating of data.

[0027] Feedback mechanism: Designers’ activities and feedback in the interactive community, such as click-through rates, comments, and sharing, will be collected and used to optimize the user information module and capability model building module, forming a closed-loop feedback system to continuously improve the platform’s intelligence level and user experience.

[0028] In this embodiment, the user information module further includes a data preprocessing unit for cleaning, deduplicating and formatting the collected designer information to ensure data quality and accuracy; specifically, the role of the data preprocessing unit is reflected in the following aspects: 1. Data cleaning Remove noisy data: Identify and remove invalid, erroneous, or abnormal data points from raw data, such as data with incorrect format, null values, extreme values, or data that is obviously illogical.

[0029] Correct errors: For erroneous data that can be corrected, such as spelling errors, inconsistent date formats, etc., the data should be corrected to make it conform to unified specifications and standards.

[0030] 2. Data deduplication Identify duplicate records: Identify and remove duplicate records or pieces of information in large amounts of data to avoid bias or double counting in subsequent analysis.

[0031] Merge similar records: For records that are partially similar but not completely the same, merge them according to certain rules to maintain the uniqueness and integrity of the data.

[0032] 3. Data formatting Unified data format: Convert data from different sources and in different formats into a unified, standard data format to facilitate subsequent data processing and analysis.

[0033] Data standardization and normalization: standardize numerical data to make it comparable; encode categorical data so that it can be recognized and processed by computers.

[0034] 4. Data Integration Cross-source data integration: Integrate designer information from different channels and systems to form a complete and comprehensive designer portrait dataset.

[0035] Construction of associated data: During the integration process, the association between data is constructed, such as the association between designers and projects, and designers and skill tags, to facilitate subsequent data mining and analysis.

[0036] 5. Quality Control Data quality assessment: Through a series of quality assessment indicators and methods, the quality of the pre-processed data is assessed to ensure the accuracy and reliability of the data.

[0037] Problem feedback and correction: Provide timely feedback and correction for problems or deficiencies found during the preprocessing process to ensure continuous improvement and optimization of data preprocessing.

[0038] In this embodiment, the capability model construction module also includes a capability evaluation standard library, which is used to define and analyze the specific meanings and evaluation standards of each capability characteristic of the designer, so as to ensure the comprehensiveness and accuracy of the capability evaluation; the capability evaluation standard library ensures a comprehensive evaluation of the designer's capabilities by defining and analyzing the specific meanings of each capability characteristic of the designer. These capability characteristics may cover multiple dimensions such as professional knowledge, technical ability, innovative thinking, teamwork ability, leadership, etc., so as to achieve a comprehensive consideration of the designer's comprehensive ability; under each capability dimension, the capability evaluation standard library may be further refined into multiple specific capability elements and indicators, making the evaluation process more detailed and in-depth. This fine-grained division helps to more accurately grasp the performance level of the designer in various capabilities; the capability evaluation standard library provides clear and specific evaluation standards for each capability characteristic, so that the evaluation process has a basis and rules to follow. These standards may include quantitative indicators (such as the number, quality, efficiency, etc. of completed projects) and qualitative descriptions (such as the performance of innovative thinking, attitude of teamwork, etc.), thereby improving the accuracy and objectivity of the evaluation; by introducing the capability evaluation standard library, the subjectivity and bias in the evaluation process can be reduced to a certain extent. Evaluators can make judgments based on unified standards, reducing evaluation errors caused by personal subjective factors; the competency evaluation standard library can also serve as a feedback mechanism for continuous optimization and improvement of the competency model. In practical applications, with the continuous improvement and change of designers' capabilities and the evolution of industry development trends, the competency evaluation standard library also needs to be continuously updated and adjusted. By collecting feedback and data during the evaluation process, the competency evaluation standard library can be continuously optimized and improved to make it more in line with actual needs and industry development trends; the establishment of a competency evaluation standard library can also help to achieve standardized management of competency evaluation. Competency evaluation and management through unified standards and processes can improve work efficiency and management level, and also help to improve designers' recognition and satisfaction with competency evaluation work.

[0039] In this embodiment, the information recommendation module also includes a dynamic adjustment unit, which is used to dynamically adjust the recommendation algorithm and parameters according to the designer's behavior data and feedback on the platform to improve the accuracy and personalization of the recommended content; the dynamic adjustment unit can receive and process the designer's real-time feedback on the recommended content (such as clicks, browsing, likes, comments, etc.), and dynamically adjust the recommendation algorithm according to these feedback data, so that the recommendation results are closer to the designer's actual needs and interests; by deeply analyzing the designer's behavior data on the platform (such as browsing history, search history, interactive behavior, etc.), the dynamic adjustment unit can identify the designer's potential interests and needs, thereby further optimizing the recommendation algorithm and improving the accuracy of the recommendation; Personalized model: The dynamic adjustment unit will combine the designer's personal portrait (such as age, occupation, skills, interests, etc.) and behavior data to build a personalized recommendation model. As the designer's activities on the platform continue to increase, this model will continue to iterate and optimize, so that the recommended content is more in line with the designer's personalized needs; the designer's interests and needs are constantly changing with time and environment. The dynamic adjustment unit can capture these changes in real time and adjust the recommendation algorithm and parameters accordingly to ensure that the recommended content is always synchronized with the designer's current status; reduce information overload: through accurate recommendations, the dynamic adjustment unit can help designers quickly find the content they are interested in, reduce the time cost of screening in massive information, and improve the efficiency of information acquisition; personalized recommendation content can meet the personalized needs of designers and improve their satisfaction and loyalty on the platform. At the same time, this positive user experience will also promote more interaction and contribution of designers on the platform.

[0040] In this embodiment, the interactive community module also includes a topic classification unit and an interactive feedback unit, which are used to classify and manage the communication content of designers, collect and analyze the interactive feedback of designers, and optimize the community atmosphere and user experience; the topic classification unit is mainly used to automatically or semi-automatically classify and manage the communication content posted by designers in the community. This function can ensure the orderliness and accessibility of community content, help designers quickly find topics of interest and participate in discussions; through preset classification tags or content-based automatic classification algorithms, designers' communication content (such as posts, comments, questions and answers, etc.) are classified into corresponding topics or categories; the classified content makes it easy for designers to quickly locate relevant information through keyword searches, thereby improving the efficiency of information retrieval; classification management helps to form professional discussion areas and promote in-depth exchanges and discussions among designers in their respective areas of expertise or interest; the interactive feedback unit is responsible for collecting and analyzing designers' interactive behaviors (such as likes, comments, shares, reports, etc.) and feedback in the community so that community management can Managers can adjust management strategies in a timely manner to optimize the community atmosphere and user experience; track interactive data in the community in real time, such as hot topics, high-frequency words, user activity, etc., to provide decision-making support for community managers; identify potential problems or violations in the community (such as spam, malicious attacks, advertising promotion, etc.) through data analysis, and take corresponding measures in a timely manner, such as deleting content, warning users or banning accounts; adjust community rules, recommendation algorithms or interface design based on user feedback and interactive data to create a more positive, healthy and harmonious community atmosphere; understand user satisfaction and needs for community functions by collecting and analyzing user feedback, and provide a basis for subsequent function optimization and product iteration.

[0041] In this embodiment, it also includes a management module for configuring and managing the functional modules of the platform to ensure the stable operation and continuous optimization of the platform; the management module also includes a data analysis unit for in-depth mining and analysis of the data on the platform to provide data support for the management decisions of the platform; the management module contains four modules, namely login interface, system interface, industry information management, and user management. It is mainly for platform managers to facilitate platform management; Login interface: The administrator must first log in to the management interface and view data uniformly. Data viewing requires attention to confidentiality measures.

[0042] System interface: There are three columns on the system homepage, namely platform field analysis, event details, and content module click-through rate. Platform field analysis: covers trend analysis of some recent fields of the platform, clicks on subdivided industries within the field, details of likes, collections, citations, and clicks on key recommended information within the field, etc., so that administrators can find the current trend and push it in time; Event details: The event details display page shows the number of events added by the platform recently, the event conversion rate, the number of participants in the event, the total number of events on the platform, etc. Administrators can use this page to filter, review and upload events. There are 25 events waiting for review. In this module, the event data is visualized, and the design event situation can be intuitively felt; Content module click-through rate section: In this section, some recommendations similar to the designer module can be displayed. Since the system itself adopts an online recommendation mechanism, the possible errors in the annotation of some industry information can easily affect the online training of the model, resulting in deviations in the industry information pushed to the designer, and ultimately reducing the accuracy. Therefore, the platform also provides a module for monitoring the accuracy of industry information recommendations, which judges the annotation accuracy of the backend data annotation personnel through the pre-set "monitoring information". Not only can the accuracy of each group's annotation staff be compared horizontally by grouping, but the accuracy of the annotations of each design competition can also be compared vertically with the competition itself. This kind of monitoring allows users to find problems in a timely manner and correct them, improving the robustness of the system. In the content module, some indicators of recent content views, push volume, recommendation satisfaction, and content details are displayed. This indicator can be used to directly match indicators with designer portraits in subsequent recommendations and package and output the overall content. The management interface includes user lists (registered + logged in), user lists (unregistered and logged in), user lists (logout), advanced filtering, historical records, etc. The user list contains all users from different fields and levels, including senior designers, design event organizers, corporate parties, etc. Specifically, it includes: user name, user basic information statistics, initial industry information preference status, user level, mobile phone number, active days, industry information preference, obtain the general industry information preference under big data, and use the recommendation algorithm to package and output the industry information liked and collected by designers to the homepage of similar designers, record the update time of the corresponding industry information, assist in timely updating of platform information, clarify the customer name (initial code) in some unregistered user lists, understand the active days, related content pushed before, and check the pop-up window on the initial interface to understand their information preferences, conduct multiple rounds of information push to attract them to register, and achieve the purpose of designer permanent residence.Record the number of active days in some deregistered user lists, conduct research on deregistration reasons through some contact information or email addresses, find key departure nodes to attract return, and improve the overall platform service strength; include various user types in the advanced screening module to assist managers in classifying and obtaining designer information, and at the same time serve as a reference for corresponding information during the update of designer capability portraits. User types are currently divided into three categories: corporate users, individual users, and personnel related to events and exhibitions. The three categories include: senior designers, design event organizers, corporate parties, unregistered users, new users added today, users with an activity level of more than three months, etc.; user conditions are classified according to the user's self-filled categories, including gender, age, occupation, work experience, education, related design works, information preferences, etc.

[0043] In this embodiment, the information recommendation module also has the ability to integrate cross-platform data, and can integrate data resources from different design crowdsourcing platforms to provide designers with more comprehensive industry information recommendation services; the information recommendation module can cross different design crowdsourcing platforms and integrate data resources from these platforms. These platforms may cover a wide range of designer groups, rich design projects, diverse design styles and techniques, etc. By integrating these data resources, the information recommendation module can build a more comprehensive and diverse information library to provide designers with a wider range of information choices; based on the data resources integrated across platforms, the information recommendation module can have a deeper understanding of the designer's interest preferences, design styles and project requirements. Through intelligent algorithm analysis, the module can provide designers with personalized information recommendations, such as hot design trends, related design cases, industry information, etc., to help designers quickly obtain valuable information; cross-platform data integration capabilities also mean that the information recommendation module can obtain the latest data from each platform in real time to ensure the timeliness and accuracy of the recommended content. Designers can keep abreast of industry trends, design trends and technological developments, thereby maintaining competitiveness and innovative thinking; integrating data resources from different platforms can also help promote cross-border exchanges between designers. Designers can be exposed to design works and ideas from different fields and backgrounds, thereby broadening their horizons, stimulating inspiration, and improving their design level and innovation capabilities.

[0044] In this embodiment, the interactive community module also has an international perspective, which can provide designers with cross-border design exchange opportunities and promote the sharing and cooperation of international design resources; through multilingual support or automatic translation functions, the interactive community module breaks the language barrier, allowing designers from different countries and regions to freely exchange ideas, share works and experiences. This borderless communication environment greatly broadens the designer's vision and promotes the collision and integration of design thinking; the international perspective enables the interactive community to capture and display global design trends, popular elements and innovative technologies in a timely manner. Designers can learn about design concepts and styles under different cultural backgrounds here, thereby enriching their design language and expression; the community platform can also serve as an incubator for cross-border design projects, providing designers with opportunities for cooperation. Through the project release, team recruitment and other functions on the platform, designers can easily form a cross-border design team to jointly complete challenging and innovative design tasks; the interactive community module can integrate design resources around the world, including design materials, case libraries, design tools, etc., to provide designers with a rich resource library. These resources can not only help designers improve design efficiency and quality, but also inspire their creative inspiration and innovation; the community platform encourages designers to share their design experience and knowledge, including design skills, software experience, industry insights, etc. This cross-border knowledge sharing helps designers learn from each other, make progress together, and promote the sustainable development of the global design industry; through the interactive community module, designers can establish a wide range of international design networks and meet designer friends and partners from different countries and regions. This cross-border network of contacts will help designers gain more opportunities and resources in their future careers; in the process of cross-border design exchanges, designers can not only share professional knowledge and experience, but also enhance understanding and respect for each other's culture and values. This mutual understanding and respect will help to establish a more stable international cooperative relationship and promote the harmonious development of the global design industry.

[0045] In this embodiment, the collaborative filtering recommendation algorithm includes the following steps: S1: Select five designer portraits as basic data; the total score of the scoring index is 100 points, of which the score of each index is: meta-ability 10 points, cognitive ability 25 points, functional ability 40 points, and social ability 25 points. The index scores of the five designers are combined into a set (also called domain), marked as 1, 2, 3, 4, and 5 respectively, then the domain For the objects to be classified; each designer uses the scores of meta-ability, cognitive ability, functional ability, and social ability as four statistical indicators, namely: ,here Represents the jth indicator of the i-th designer.

[0046] S2: Select the maximum-minimum method to establish the fuzzy similarity matrix. The calculation formula of the maximum-minimum method is: ,in is the minimum sum of any two rows, is the sum of the maximum values ​​of any two rows, It is a quantity that indicates the similarity between the i-th designer and the j-th designer in the four performance indicators; S3: With the help of MATLAB calculation, the matrix of fuzzy similarity relationships of the four indicators of the five designers was obtained; S4: Cluster analysis of 5 designers using fuzzy equivalence matrix, intercept level From 1 to 0, press Cut the matrix to perform dynamic clustering; take , then: From the above matrix, we can see that x is divided into 5 categories: , that is: {No. 1}, {No. 2}, {No. 3}, {No. 4}, {No. 5}.

[0047] Lower intercept level For different Doing the same analysis, we get When , we have: From the above matrix, it can be divided into 3 categories: , that is, {No. 1, No. 2, No. 4}, {No. 3}, {No. 5}.

[0048] when When , we have: , as can be seen from the above matrix, can be divided into 2 categories: , that is, {No. 1, No. 2, No. 4, No. 5}, {No. 3}.

[0049] when When , we have: , from the above matrix, x is divided into 1 category: , that is: {No. 1}, {No. 2}, {No. 3}, {No. 4}, {No. 5}.

[0050] In the classification process, it can be seen that if the intercept level When the value is too high, the designer profiles will be divided into different categories. When the value is too low, the designer portraits will be clustered into one category. In order to better conduct cluster analysis and objectively and reasonably reflect the differences between designers, different intercept levels are used. Conduct research on designer portraits.

[0051] when When , the classification results can better reflect the differences between designers, so The classification is scientific and desirable, indicating that the abilities of designers grouped into one category are relatively similar.

[0052] In this embodiment, the platform adopts a modular design to facilitate the independent development and upgrading of each functional module, thereby improving the flexibility and scalability of the platform.

[0053] Working principle: The industry information recommendation platform based on designer portraits is a professional information service platform for designers. Its core logic is to provide designers with personalized information recommendations by building accurate designer portraits, combining the latest trends in the industry and market demand, helping them improve their professional skills, broaden their design horizons, and seize market opportunities; first, the platform will collect multi-dimensional information about designers and build a comprehensive and in-depth designer portrait. This information includes but is not limited to the designer's personal basic information, education experience, work experience, design works, skills, hobbies, etc. Through in-depth mining and analysis of this information, the platform can accurately grasp the key characteristics of the designer's professional background, design style, innovation ability, and market value; based on the designer portrait, the platform will further analyze industry trends and market demand and provide designers with personalized information recommendations. These recommendations may include the latest design cases, popular design elements, cutting-edge design concepts, practical design tools, and business cooperation opportunities. By accurately matching the designer's portrait information and industry trends, the platform can help designers quickly obtain valuable information and improve their competitiveness and innovation in the design field. In addition, the platform can also analyze the designer's behavior data, feedback and interaction on the platform, and the platform can continuously adjust the recommended content to make it more in line with the designer's actual needs and interest preferences. At the same time, the platform will recommend more challenging and potential project opportunities and cooperation resources for designers based on their professional growth and development trajectory, helping them achieve greater breakthroughs and achievements in the design field. In addition to personalized information recommendations, the platform will also provide a communication community among designers to promote knowledge sharing and collaborative innovation among designers. In this community, designers can share their design experience and insights, discuss hot topics and trends in the industry, and find like-minded partners and sources of inspiration. Through communication and cooperation, designers can continuously broaden their horizons and ideas, and improve their professional skills and innovation level. In summary, the industry information recommendation platform based on designer portraits provides designers with an efficient, convenient and professional information acquisition and career development platform by accurately portraying designer portraits, providing personalized recommendation services and promoting communication and cooperation among designers. Specifically, the platform will first conduct a comprehensive analysis of the designer's personal characteristics, professional background, design style, interest preferences, etc. to form an accurate portrait of the designer. Then, based on this portrait information, the platform will intelligently screen and recommend industry information that is highly matched with the designer's needs, including the latest design concepts, popular design elements, practical design tools, potential cooperation opportunities, etc. This personalized information recommendation can not only help designers quickly obtain valuable information and improve work efficiency, but also guide them to explore new design ideas and methods and inspire innovative inspiration; in addition, the industry information recommendation platform based on designer portraits also has intelligent learning and optimization capabilities.The platform will continuously adjust and optimize the recommendation algorithm based on the designer's behavior data and feedback on the platform, so that the recommendation results are more in line with the designer's actual needs and interest preferences. This intelligent recommendation method not only improves the user experience of the platform, but also enables designers to more easily obtain information and resources that meet their needs.

[0054] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0055] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A design competition information recommendation platform based on designer portraits, including a user information module, a capability model building module, an information recommendation module and an interactive community module, characterized in that: The user information module is used to collect various requirements of designers and construct designer data; The capability model building module builds a designer portrait based on designer data through fuzzy clustering analysis algorithm and multi-dimensional capability feature analysis; the information recommendation module uses a tag-based collaborative filtering recommendation algorithm to recommend personalized industry information to designers based on designer portraits and design competition information database; The interactive community module provides a communication platform among designers to promote knowledge sharing and collaborative innovation.

2. The design competition information recommendation platform based on designer portraits according to claim 1, characterized in that: The user information module also includes a data preprocessing unit for cleaning, deduplicating and formatting the collected designer information to ensure data quality and accuracy.

3. The design competition information recommendation platform based on designer portraits according to claim 1, characterized in that: The capability model building module also includes a capability evaluation standard library, which is used to define and analyze the specific meaning and evaluation standards of various capability characteristics of designers to ensure the comprehensiveness and accuracy of capability evaluation.

4. The design competition information recommendation platform based on designer portraits according to claim 1, characterized in that: The information recommendation module also includes a dynamic adjustment unit for dynamically adjusting the recommendation algorithm and parameters according to the designer's behavior data and feedback on the platform to improve the accuracy and personalization of the recommended content.

5. The design competition information recommendation platform based on designer portraits according to claim 1, characterized in that: The interactive community module also includes a topic classification unit and an interactive feedback unit, which are used to classify and manage the communication content of designers, collect and analyze the interactive feedback of designers, and optimize the community atmosphere and user experience.

6. The design competition information recommendation platform based on designer portraits according to claim 1, characterized in that: It also includes a management module for configuring and managing the functional modules of the platform to ensure the stable operation and continuous optimization of the platform; the management module also includes a data analysis unit for in-depth mining and analysis of the data on the platform to provide data support for the management decisions of the platform.

7. The design competition information recommendation platform based on designer portraits according to claim 1, characterized in that: The information recommendation module also has cross-platform data integration capabilities and can integrate data resources from different design crowdsourcing platforms to provide designers with more comprehensive industry information recommendation services.

8. The design competition information recommendation platform based on designer portraits according to claim 1, characterized in that: The interactive community module also has an international perspective, which can provide designers with cross-border design exchange opportunities and promote the sharing and cooperation of international design resources.

9. The design competition information recommendation platform based on designer portraits according to claim 1, characterized in that: The collaborative filtering recommendation algorithm includes the following steps: S1: Select five designer portraits as basic data; combine the five designer index scores into a set (also called domain), marked as 1, 2, 3, 4, and 5 respectively, then the domain For the object being classified; S2: Select the maximum-minimum method to establish the fuzzy similarity matrix. The calculation formula of the maximum-minimum method is: ,in is the minimum sum of any two rows, is the sum of the maximum values ​​of any two rows, It means Designer and The amount of similarity among designers on the four performance indicators; S3: With the help of MATLAB calculation, the matrix of fuzzy similarity relationships of the four indicators of the five designers was obtained; S4: Cluster analysis of 5 designers using fuzzy equivalence matrix, intercept level From 1 to 0, press Dynamic clustering of the truncated matrix.

10. The platform according to any one of claims 1 to 9, characterized in that: The platform adopts a modular design to facilitate the independent development and upgrading of each functional module, thereby improving the flexibility and scalability of the platform.

Images