Method and device for online customized recommendation of personalized national culture products

Abstract

The invention discloses a personalized ethnic culture product online customization recommendation method and device, and relates to the field of commodity recommendation. In the method, a culture preference label is generated according to behavior data, and a consumption habit label is generated according to consumption data; constructing a user portrait of the target user according to the registration information, the culture preference tag and the consumption habit tag; calculating the similarity between the users according to the user portraits, determining a first user whose similarity is greater than a first threshold, and obtaining a first product purchased by the first user within a first preset time period; extracting feature information of the national culture products from a product database, determining a matching degree between the target user and the feature information according to the user portrait, and determining a second product of which the matching degree is greater than a second threshold value; and determining weights of the first product and the second product, and performing weighting processing on the first product and the second product according to the weights to determine a recommendation sequence. By implementing the technical scheme provided by the invention, accurate and personalized national culture product recommendation is realized.

Description

Technical Field

[0001] The present application relates to the technical field of commodity recommendation, and in particular to a method and device for online customized recommendation of personalized ethnic cultural products. Background Art

[0002] With the rapid development of Internet technology, e-commerce platforms have gradually become one of the main channels for consumers to purchase goods. In particular, for ethnic products with unique cultural and historical backgrounds, online sales have not only expanded the market coverage, but also improved consumers' shopping experience. Through personalized recommendation mechanisms, e-commerce platforms can better meet the needs of different users, improve user satisfaction and loyalty, and thus promote sales growth.

[0003] Existing personalized recommendation systems mainly rely on a single data source or a simple data analysis model, resulting in a lack of accuracy and diversity in recommendation results. For example, relying solely on a user's browsing history may not fully reflect the user's real needs, while ignoring the user's registration information and specific cultural preferences may result in the recommended products not matching the user's real preferences. In addition, traditional recommendation algorithms are prone to cold start problems when dealing with large user groups, meaning that it is difficult to obtain effective recommendations for new users or new products.

[0004] Therefore, how to combine multi-dimensional user data to achieve efficient and accurate personalized recommendations is a key issue that needs to be solved urgently. Summary of the invention

[0005] The present application provides a method and device for online customized recommendation of personalized ethnic cultural products, which not only improves the personalization and accuracy of recommendations, but also enhances user participation and satisfaction, improves the dynamics and adaptability of the recommendation system, optimizes the recommendation order and user experience, promotes the sales and dissemination of ethnic cultural products, and provides strong support for the operation and decision-making of the platform.

[0006] In a first aspect of the present application, a method for online customized recommendation of personalized ethnic cultural products is provided, which is applied to an ethnic cultural product platform, and the method comprises: Obtain the target user's registration information, behavior data and consumption data, generate a cultural preference tag based on the behavior data, and generate a consumption habit tag based on the consumption data. The registration information includes age, gender, occupation, and region. The behavior data includes browsing history, search keywords, and product information in favorites on the platform. The consumption data includes purchase frequency and purchase price. Constructing a user profile of the target user according to the registration information, the cultural preference tag, and the consumption habit tag; Calculate the similarity between users based on the user portraits, determine a first user whose similarity is greater than a first threshold, and obtain a first product purchased by the first user within a first preset time period; Extracting characteristic information of ethnic cultural products from a product database, the characteristic information including cultural elements, style characteristics, material, craftsmanship and price, determining a matching degree between the target user and the characteristic information according to the user portrait, and determining a second product with a matching degree greater than a second threshold; Determine the weights of the first product and the second product according to the number of times the target user browses ethnic cultural products within a second preset time period, perform weighted processing on the first product and the second product according to the weights to determine a recommendation order, and make recommendations according to the recommendation order; Receive feedback information from the target user on the recommended product for customization.

[0007] Optionally, generating a cultural preference label according to the behavior data includes: Determining a second number of times the target user browses each national cultural product category according to the first number of times the target user browses each national cultural product page; Determining a second time that the target user stays in each ethnic cultural product category according to the first time that the target user stays on each ethnic cultural product page; Generate a first cultural preference sub-label according to a first target ethnic cultural product category, wherein the first target ethnic cultural product category is an ethnic cultural product category whose second number is greater than a third threshold and whose second time is greater than a fourth threshold; generating a second cultural preference sub-label according to a second target ethnic cultural product category, wherein the second target ethnic cultural product category is the ethnic cultural product category to which the target ethnic cultural product belongs, and the target ethnic cultural product is the ethnic cultural product corresponding to the keyword whose search frequency exceeds the fifth threshold; Generate a third cultural preference sub-label based on the cultural elements and style characteristics of the national cultural products in the favorites; A cultural preference tag is generated according to the first cultural preference subtag, the second cultural preference subtag, and the third cultural preference subtag.

[0008] Optionally, generating a consumption habit label according to the consumption data includes: Generate a first consumption habit sub-tag according to the target purchase frequency of the target user within a third preset time period; Generating a second consumption habit sub-label according to the median and maximum values ​​of the purchase prices of the target user within a third preset time period; A consumption habit tag is generated according to the first consumption habit sub-tag and the second consumption habit sub-tag.

[0009] Optionally, calculating the similarity between users according to user portraits includes: Converting the registration information of the user portrait into a first feature vector, converting the cultural preference label of the user portrait into a second feature vector, and converting the consumption habit label of the user portrait into a third feature vector; The first feature vector, the second feature vector and the third feature vector are combined into a comprehensive feature vector, and the similarity between the comprehensive feature vector of the target user and the comprehensive feature vectors of other users is calculated, where the other users are users other than the target user in the national cultural product platform.

[0010] Optionally, determining the matching degree between the target user and the feature information according to the user portrait includes: Converting the feature information into a fourth feature vector, calculating a first matching degree between the second feature vector and the fourth feature vector, and calculating a second matching degree between the third feature vector and the fourth feature vector; The first matching degree and the second matching degree are weighted to determine a matching degree between the target user and the feature information.

[0011] Optionally, determining the weights of the first product and the second product according to the number of times the target user browses the ethnic cultural product within the second preset time period includes: Calculate the difference between the third time the target user browses the national cultural product within the second preset time period and a sixth threshold, and calculate the ratio of the difference to the sixth threshold; When the difference is greater than zero, reducing the basic weight of the first product according to the proportion to obtain a first weight of the first product, and increasing the basic weight of the second product according to the proportion to obtain a second weight of the second product; When the difference is less than zero, the basic weight of the first product is increased according to the proportion to obtain the first weight of the first product, and the basic weight of the second product is reduced according to the proportion to obtain the second weight of the second product.

[0012] Optionally, performing weighted processing on the first product and the second product according to the weight to determine the recommendation order includes: The first target product that overlaps between the first product and the second product is recommended as the most prioritized product; Using the similarity as the first matching value of the first product, using the matching degree as the second matching value of the second product, calculating the total score of the first product according to the first matching value and the first weight, and calculating the total score of the second product according to the second matching value and the second weight; The second target product is recommended in descending order of total score, and the second target product is the product between the first product and the second product except the first target product.

[0013] In a second aspect of the present application, a system for online customization and recommendation of personalized ethnic cultural products is provided, including a label module, a portrait module, a first recommendation module, a second recommendation module, an execution module and a customization module, wherein: A label module is configured to obtain registration information, behavior data and consumption data of target users, generate cultural preference labels based on the behavior data, and generate consumption habit labels based on the consumption data, wherein the registration information includes age, gender, occupation, and region, the behavior data includes browsing tracks, search keywords, and product information in favorites on the platform, and the consumption data includes purchase frequency and purchase price; A portrait module, configured to construct a user portrait of the target user according to the registration information, the cultural preference tag and the consumption habit tag; A first recommendation module, configured to calculate the similarity between users according to the user portraits, determine a first user whose similarity is greater than a first threshold, and obtain a first product purchased by the first user within a first preset time period; A second recommendation module is configured to extract characteristic information of ethnic cultural products from a product database, the characteristic information including cultural elements, style characteristics, material, craftsmanship and price, determine a matching degree between the target user and the characteristic information according to the user portrait, and determine a second product whose matching degree is greater than a second threshold; an execution module, configured to determine weights of the first product and the second product according to the number of times the target user browses ethnic cultural products within a second preset time period, perform weighted processing on the first product and the second product according to the weights to determine a recommendation order, and make recommendations according to the recommendation order; The customization module is configured to receive the target user's feedback information on the recommended product for customization.

[0014] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes any one of the methods described above.

[0015] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, any of the methods described above is executed.

[0016] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By integrating the target user's registration information, behavior data and consumption data, a comprehensive user profile is constructed. This includes not only the user's basic information (such as age, gender, occupation, and region), but also cultural preferences and consumption habits, so that the recommendation system can more accurately understand the user's needs and preferences, and extract the characteristic information of national cultural products from the product database, including cultural elements, style characteristics, materials, craftsmanship and price, so that the recommendation system can more comprehensively evaluate the match between the product and the user profile, thereby improving the accuracy of the recommendation; 2. Recommend national cultural products that best suit users’ interests and needs based on the matching degree between user portraits and product feature information. This personalized recommendation can improve user engagement, increase user stay time and interaction frequency on the platform, and provide online customization functions, allowing users to choose product materials, colors, patterns, etc. according to their preferences, further meeting users’ personalized needs and improving user satisfaction; 3. Dynamically adjust the weights of the first and second products based on the number of times the target user browses ethnic cultural products within a preset time period. This dynamic adjustment mechanism enables the recommendation system to adjust the recommended content in a timely manner according to the user's current behavior and interest changes, improve the timeliness and adaptability of the recommendation, and calculate the similarity between users to identify users with high similarity and obtain the products they have purchased for recommendation. This method not only takes into account the user's historical behavior, but also draws on the purchase experience of similar users, further improving the accuracy and diversity of the recommendation; 4. Weight the first product and the second product according to the weights to determine the recommendation order. This weighted processing method can comprehensively consider the user's historical behavior and current interests, generate a more reasonable recommendation list, improve the user experience, sort the products according to the weighted scores, ensure that the recommended product order is more in line with the user's current needs and interests, and improve the user's purchase conversion rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flowchart of a method for online customization and recommendation of personalized ethnic cultural products disclosed in an embodiment of the present application; Figure 2 It is a module diagram of a system for online customization and recommendation of personalized ethnic cultural products disclosed in an embodiment of the present application; Figure 3 It is a structural schematic diagram of an electronic device disclosed in an embodiment of the present application.

[0018] Explanation of the accompanying drawings: 201, label module; 202, portrait module; 203, first recommendation module; 204, second recommendation module; 205, execution module; 206, customization module; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0019] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0020] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0021] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0022] This embodiment discloses a method for online customization and recommendation of personalized ethnic cultural products, which is applied to an ethnic cultural product platform. Figure 1 is a flow chart of a method for online customization and recommendation of personalized ethnic cultural products disclosed in an embodiment of the present application, such as Figure 1 As shown, the method comprises the following steps: S101, obtaining the registration information, behavior data and consumption data of the target user, generating a cultural preference tag based on the behavior data, and generating a consumption habit tag based on the consumption data, wherein the registration information includes age, gender, occupation, and region, the behavior data includes browsing tracks, search keywords, and product information in favorites on the platform, and the consumption data includes purchase frequency and purchase price; S102, constructing a user profile of the target user according to the registration information, the cultural preference tag, and the consumption habit tag; S103, calculating the similarity between users according to the user portraits, determining a first user whose similarity is greater than a first threshold, and obtaining a first product purchased by the first user within a first preset time period; S104, extracting characteristic information of ethnic cultural products from a product database, the characteristic information including cultural elements, style characteristics, material, craftsmanship and price, determining a matching degree between the target user and the characteristic information according to the user portrait, and determining a second product whose matching degree is greater than a second threshold; S105, determining weights of the first product and the second product according to the number of times the target user browses ethnic cultural products within a second preset time period, performing weighted processing on the first product and the second product according to the weights to determine a recommendation order, and making recommendations according to the recommendation order; S106: Receive feedback information from the target user on the recommended product for customization.

[0023] Registration information includes basic information such as age, gender, occupation, and region. Behavioral data includes browsing tracks, search keywords, and favorite product information on the platform. Consumption data includes purchase frequency and purchase price. Generate cultural preference tags based on behavioral data (browsing tracks, search keywords, favorite product information). For example, if a user frequently browses the Miao silver jewelry page, searches for Miao silver jewelry keywords, and collects multiple Miao silver jewelry products, a cultural preference tag "Miao silver jewelry enthusiast" is generated. Generate consumption habit tags based on consumption data (purchase frequency and purchase price). For example, if a user purchases frequently and at a high price, a consumption habit tag "high-frequency, high-priced consumer" is generated. Integrate registration information, cultural preference tags, and consumption habit tags to form a comprehensive user portrait. For example, a user portrait may include: Basic information: Age_25, Gender_Female, Occupation_Teacher, Region_Beijing; Cultural preference: Miao silver jewelry enthusiasts, interested in Miao silver jewelry, and prefer traditional style Miao silver jewelry; Consumption habits: high-frequency purchasing users, high-price consumers.

[0024] Convert the user portrait into a feature vector and calculate the similarity between the target user and other users. You can use methods such as cosine similarity and Euclidean distance. Set a similarity threshold (first threshold, such as 0.7) to filter out users with a similarity greater than the first threshold. Obtain the products purchased by these similar users within a preset time period (such as the past three months) as candidate products for recommendation. Extract product feature information such as cultural elements, style characteristics, materials, craftsmanship and price. Calculate the matching degree between the cultural preferences and consumption habits in the user portrait and the product feature information. You can use methods such as cosine similarity. Set a matching threshold (second threshold, such as 0.7) to filter out products with a matching degree greater than the second threshold as candidate products for recommendation. Count the number of times the target user browses ethnic cultural products within a preset time period (such as the past month). If the number of browsing times is large, increase the weight of the second product; if the number of browsing times is small, increase the weight of the first product. Calculate the weighted scores of the first product and the second product, and determine the recommendation order based on the weighted scores. Recommend the products to the target user in the recommended order. Receive feedback from target users on recommended products, and based on the user feedback, allow users to customize the products, such as choosing different materials, colors, patterns, etc.

[0025] By integrating the target user's registration information, behavior data and consumption data, a comprehensive user profile is constructed. This includes not only the user's basic information (such as age, gender, occupation, region), but also cultural preferences and consumption habits, so that the recommendation system can understand the user's needs and preferences more accurately. The characteristic information of ethnic cultural products, including cultural elements, style characteristics, materials, craftsmanship and price, is extracted from the product database, so that the recommendation system can more comprehensively evaluate the matching degree between the product and the user profile, thereby improving the accuracy of the recommendation. According to the matching degree between the user profile and the product characteristic information, the ethnic cultural products that best meet the user's interests and needs are recommended to the user. This personalized recommendation can improve user participation, increase the user's stay time and interaction frequency on the platform. Provide online customization function, allowing users to choose the material, color, pattern, etc. of the product according to their preferences, further meet the user's personalized needs and improve user satisfaction. According to the number of times the target user browses the ethnic cultural products within the preset time period, the weight of the first product and the second product is dynamically adjusted. This dynamic adjustment mechanism enables the recommendation system to adjust the recommended content in a timely manner according to the user's current behavior and interest changes, and improve the timeliness and adaptability of the recommendation. By calculating the similarity between users, users with high similarity are identified, and the products they have purchased are obtained for recommendation. This method not only considers the user's historical behavior, but also draws on the purchase experience of similar users, further improving the accuracy and diversity of recommendations. The first and second products are weighted according to the weights to determine the recommendation order. This weighted processing method can comprehensively consider the user's historical behavior and current interests, generate a more reasonable recommendation list, and improve the user experience. Products are sorted according to the weighted scores to ensure that the recommended product order is more in line with the user's current needs and interests, and improve the user's purchase conversion rate.

[0026] Optionally, generating a cultural preference label according to the behavior data includes: Determining a second number of times the target user browses each national cultural product category according to the first number of times the target user browses each national cultural product page; Determining a second time that the target user stays in each ethnic cultural product category according to the first time that the target user stays on each ethnic cultural product page; Generate a first cultural preference sub-label according to a first target ethnic cultural product category, wherein the first target ethnic cultural product category is an ethnic cultural product category whose second number is greater than a third threshold and whose second time is greater than a fourth threshold; generating a second cultural preference sub-label according to a second target ethnic cultural product category, wherein the second target ethnic cultural product category is the ethnic cultural product category to which the target ethnic cultural product belongs, and the target ethnic cultural product is the ethnic cultural product corresponding to the keyword whose search frequency exceeds the fifth threshold; Generate a third cultural preference sub-label based on the cultural elements and style characteristics of the national cultural products in the favorites; A cultural preference tag is generated according to the first cultural preference subtag, the second cultural preference subtag, and the third cultural preference subtag.

[0027] Record the first number of times the target user browses each ethnic cultural product page. Based on the first number, determine the total number of times the target user browses each ethnic cultural product category. For example, count the number of times the user browses the Miao silver jewelry page, the number of times the user browses the Tibetan thangka page, etc. Record the first time the target user stays on each ethnic cultural product page. Based on the first time, determine the total stay time of the target user on each ethnic cultural product category page. For example, count the total stay time of the user on the Miao silver jewelry page, the total stay time on the Tibetan thangka page, etc. Set the third threshold of the number of views and the fourth threshold of the stay time. For example, set the third threshold to 10 times and the fourth threshold to 10 minutes. Set the ethnic cultural product category with a number of views greater than the third threshold and a stay time greater than the fourth threshold as the first target ethnic cultural product category, and generate a first cultural preference sub-label. For example, if the user browses the Miao silver jewelry page more than 10 times and the stay time exceeds 10 minutes, generate a sub-label "Miao silver jewelry enthusiast". Record the target user's search keywords, identify keywords related to ethnic culture, and count the search frequency of each keyword. Set the fifth threshold of search frequency. For example, the fifth threshold is set to 5 times. The ethnic cultural product category corresponding to the keyword whose search frequency exceeds the fifth threshold is used as the second target ethnic cultural product category, and a second cultural preference sub-tag is generated. For example, if the number of times the user searches for "Miao silver jewelry" exceeds 5 times, a sub-tag "interested in Miao silver jewelry" is generated. The product information in the favorites of the target user is retrieved, and the cultural elements and style characteristics of the collected ethnic cultural products are extracted. A third cultural preference sub-tag is generated based on the cultural elements and style characteristics of the ethnic cultural products in the favorites. For example, if the user collects a number of Miao silver jewelry products with traditional style, a sub-tag "prefer traditional style Miao silver jewelry" is generated. The first cultural preference sub-tag, the second cultural preference sub-tag and the third cultural preference sub-tag are integrated into a comprehensive cultural preference tag. The generated cultural preference tags are optimized and integrated to avoid repeated and redundant tags. For example, "Miao silver jewelry lovers" and "interested in Miao silver jewelry" are merged into "Miao silver jewelry lovers", and "prefer traditional style Miao silver jewelry" and "interested in traditional style Miao silver jewelry" are merged into "prefer traditional style Miao silver jewelry".

[0028] By analyzing the number of views, dwell time, search keywords and favorites of users on the platform, the interests and preferences of users are captured from multiple dimensions. This method can more comprehensively reflect the real interests of users than single-dimensional behavior analysis. The analysis of the number of views and dwell time can dynamically identify the changes in users' interests in different ethnic cultural products, timely capture users' current interests, and improve the timeliness and accuracy of recommendations. By setting thresholds for the number of views and dwell time, the categories of cultural products that users are really interested in are screened out to avoid misjudging users' interests due to accidental browsing behavior, and improve the accuracy and reliability of labels. The first cultural preference sub-label, the second cultural preference sub-label and the third cultural preference sub-label are combined to generate a comprehensive cultural preference label to ensure the comprehensiveness and accuracy of the label and avoid errors caused by a single behavior. The generated cultural preference label provides a solid foundation for personalized recommendations, enabling the recommendation system to more accurately recommend ethnic cultural products that meet users' interests and improve the relevance and personalization of recommendations. By analyzing the frequency of search keywords, the categories of cultural products that users actively search for are identified, further enhancing the pertinence and personalization of recommendations, because users' search behavior usually reflects their strong interests and needs. Through accurate cultural preference tags, the recommendation system can recommend national cultural products that best suit users' interests, improve user satisfaction and engagement, and increase user stay time and interaction frequency on the platform. Combined with users' cultural preference tags, online customization functions are provided, allowing users to choose product materials, colors, patterns, etc. according to their preferences, further meeting users' personalized needs and improving user satisfaction.

[0029] Optionally, generating a consumption habit label according to the consumption data includes: Generate a first consumption habit sub-tag according to the target purchase frequency of the target user within a third preset time period; Generating a second consumption habit sub-label according to the median and maximum values ​​of the purchase prices of the target user within a third preset time period; A consumption habit tag is generated according to the first consumption habit sub-tag and the second consumption habit sub-tag.

[0030] Collect the target user's purchase records within the third preset time period, including the number and time of purchases. Calculate the target user's purchase frequency within the time period, that is, the number of purchases divided by the length of the time period (such as days, months, etc.). Generate a first consumer habit sub-tag based on the purchase frequency. For example: if the purchase frequency is higher than the preset high frequency threshold (such as more than 3 purchases per month), generate a tag "high frequency purchase user"; if the purchase frequency is lower than the preset low frequency threshold (such as less than 1 purchase per month), generate a tag "low frequency purchase user"; if the purchase frequency is between the two, generate a tag "medium frequency purchase user". Collect the target user's purchase records within the third preset time period, including the price of each purchase. Calculate the median and maximum purchase prices. Generate a second consumer habit sub-tag based on the median and maximum purchase prices. For example: if the median and maximum of the purchase price are both higher than the preset high price threshold (e.g. the median is more than 500 yuan and the maximum is more than 1,000 yuan), the label "high-priced consumer" is generated; if the median and maximum of the purchase price are both lower than the preset low price threshold (e.g. the median is less than 200 yuan and the maximum is less than 500 yuan), the label "low-priced consumer" is generated; if the median and maximum of the purchase price are between the two, the label "medium-priced consumer" is generated. Integrate the first consumption habit sub-label and the second consumption habit sub-label into a comprehensive consumption habit label. For example: if the user is a "high-frequency purchase user" and a "high-priced consumer", the comprehensive label "high-frequency high-priced consumer" is generated. If the user is a "low-frequency purchase user" and a "low-priced consumer", the comprehensive label "low-frequency low-priced consumer" is generated. If the user is a "medium-frequency purchase user" and a "medium-priced consumer", the comprehensive label "medium-frequency medium-priced consumer" is generated.

[0031] By analyzing the purchase frequency and purchase price, the user's consumption habits can be accurately identified, providing a more accurate basis for personalized recommendations. Combining the median and maximum values ​​of the purchase frequency and purchase price, the user's consumption behavior is analyzed from multiple dimensions to generate more comprehensive consumption habit labels. The generated consumption habit labels can help the recommendation system more accurately recommend products that meet the user's consumption habits, and improve the relevance and personalization of the recommendations. Through accurate consumption habit labels, the recommendation system can recommend products that are more in line with the user's consumption ability and frequency, thereby improving user satisfaction and purchase conversion rate. By analyzing the user's consumption data, the platform can better understand market trends and user needs, and provide data support for product procurement, inventory management, and marketing strategies.

[0032] Optionally, calculating the similarity between users according to user portraits includes: Converting the registration information of the user portrait into a first feature vector, converting the cultural preference label of the user portrait into a second feature vector, and converting the consumption habit label of the user portrait into a third feature vector; The first feature vector, the second feature vector and the third feature vector are combined into a comprehensive feature vector, and the similarity between the comprehensive feature vector of the target user and the comprehensive feature vectors of other users is calculated, where the other users are users other than the target user in the national cultural product platform.

[0033] Extract registration information from user portraits, including age, gender, occupation, region, etc. Convert this basic information into numerical or vector form. For example: Age: Use the value directly (such as 25). Gender: Use binary encoding (female = 1, male = 0). Occupation and region: Use one-hot encoding. Assuming that the occupations include "teacher", "doctor", "engineer", etc., and the regions include "Beijing", "Shanghai", "Guangzhou", etc., the encoding examples are as follows: Teacher: [1, 0, 0]; Doctor: [0, 1, 0]; Engineer: [0, 0, 1]; Beijing: [1, 0, 0]; Shanghai: [0, 1, 0]; Guangzhou: [0, 0, 1].

[0034] These codes are combined into the first feature vector. For example, the registration information feature vector of user A is: [25, 1, [1, 0, 0], [1, 0, 0]].

[0035] Extract cultural preference labels from user portraits, such as "Hmong silver jewelry enthusiast", "interested in Miao silver jewelry", "prefer traditional style Miao silver jewelry". Convert these labels into numerical or vector form, using the bag of words model or TF-IDF (Term Frequency-Inverse Document Frequency). For example: Miao silver jewelry enthusiasts: 1; Interested in Miao silver jewelry: 1; Prefer traditional style Miao silver jewelry: 1.

[0036] These codes are combined into the second feature vector. For example, the cultural preference feature vector of user A is: [1, 1, 1].

[0037] Extract consumption habit labels from user portraits, such as "high-frequency purchase users" and "high-price consumers". Convert these labels into numerical or vector form, and use simple binary encoding. For example: High frequency purchasing users: 1; High-price consumers: 1.

[0038] These codes are combined into the third feature vector. For example, the consumption habit feature vector of user A is: [1, 1].

[0039] Combine the first feature vector, the second feature vector, and the third feature vector into a comprehensive feature vector. For example, the comprehensive feature vector of user A is: [25, 1, [1, 0, 0], [1, 0, 0], 1, 1, 1, 1, 1]. Select an appropriate similarity calculation method. Common methods include cosine similarity, Euclidean distance, Jaccard similarity, etc. Use the selected similarity calculation method to calculate the similarity between the comprehensive feature vector of the target user and the comprehensive feature vectors of other users.

[0040] By converting registration information, cultural preferences, and consumption habits into feature vectors, a comprehensive user feature representation is generated, providing richer information for similarity calculation. Using appropriate similarity calculation methods, the similarity between users can be accurately calculated to improve the accuracy and relevance of recommendations. By identifying similar users, the recommendation system can learn from the purchase behavior of similar users and recommend products that are more in line with their interests and consumption habits to target users, thereby enhancing the personalization of recommendations. Through accurate similarity calculations, the recommendation system can recommend products that better meet users' needs, thereby improving user satisfaction and purchase conversion rates. By analyzing the similarity between users, the platform can better understand the characteristics and behavior patterns of user groups, and provide data support for product procurement, inventory management, and marketing strategies.

[0041] Optionally, determining the matching degree between the target user and the feature information according to the user portrait includes: Converting the feature information into a fourth feature vector, calculating a first matching degree between the second feature vector and the fourth feature vector, and calculating a second matching degree between the third feature vector and the fourth feature vector; The first matching degree and the second matching degree are weighted to determine a matching degree between the target user and the feature information.

[0042] Extract characteristic information of ethnic cultural products from the product database, including cultural elements, style characteristics, materials, craftsmanship and prices. Convert these characteristic information into numerical or vector form. For example: Cultural elements: Miao silver jewelry = 1, Tibetan thangka = 2, Dai brocade = 3; Style characteristics: traditional = 1, modern = 2, simple = 3; Material: Silver = 1, Silk = 2, Wood = 3; Technology: handmade = 1, machine = 2; Price: Use a numeric value (such as 500 yuan).

[0043] Combine these codes into the fourth feature vector. For example, the feature vector of product P is: [1, 1, 1, 1, 500].

[0044] Calculate the similarity between the user's cultural preference feature vector (second feature vector) and the product's feature vector (fourth feature vector) as the first matching degree. You can use methods such as cosine similarity, Euclidean distance, or Jaccard similarity to calculate the similarity. Calculate the similarity between the user's consumption habit feature vector (third feature vector) and the product's feature vector (fourth feature vector) as the second matching degree. You can also use methods such as cosine similarity, Euclidean distance, or Jaccard similarity. Set the weight of cultural preference matching to w1 and the weight of consumption habit matching to w2. For example, w1=0.6, w2=0.4. Use the weighted average to calculate the comprehensive matching degree: comprehensive matching degree=w1×first matching degree+w2×second matching degree. For example, if the first matching degree is 0.8 and the second matching degree is 0.7, the comprehensive matching degree is: comprehensive matching degree=0.6×0.8+0.4×0.7=0.76. Set a matching degree threshold, such as 0.7, to determine the matching degree between the user and the product. According to the comprehensive matching degree, select products with matching degree greater than the threshold, and recommend products with high matching degree to target users.

[0045] By combining the matching degree of cultural preferences and consumption habits, the matching degree between users and products is comprehensively evaluated to improve the accuracy and relevance of recommendations. By setting different weights, the importance of cultural preferences and consumption habits can be adjusted according to actual conditions, making the matching degree calculation more flexible and more in line with actual needs. Through comprehensive matching, the recommendation system can more accurately recommend products that meet user interests and consumption habits, and enhance the personalization of recommendations. Through accurate matching calculation, the recommendation system can recommend products that better meet users' needs, improve user satisfaction and purchase conversion rate. By analyzing the matching degree between users and products, the platform can better understand user needs and market trends, and provide data support for product procurement, inventory management and marketing strategies.

[0046] Optionally, determining the weights of the first product and the second product according to the number of times the target user browses the ethnic cultural product within the second preset time period includes: Calculate the difference between the third time the target user browses the national cultural product within the second preset time period and a sixth threshold, and calculate the ratio of the difference to the sixth threshold; When the difference is greater than zero, reducing the basic weight of the first product according to the proportion to obtain a first weight of the first product, and increasing the basic weight of the second product according to the proportion to obtain a second weight of the second product; When the difference is less than zero, the basic weight of the first product is increased according to the proportion to obtain the first weight of the first product, and the basic weight of the second product is reduced according to the proportion to obtain the second weight of the second product.

[0047] Count the number of times the target user browses ethnic cultural products in the second preset time period, and record it as the third number. Set a sixth threshold to determine whether the user's browsing behavior is frequent. For example, the sixth threshold can be set to 10 times. Calculate the difference between the third number and the sixth threshold, and record it as the difference. Difference = third number - sixth threshold. Calculate the ratio of the difference to the sixth threshold = difference / sixth threshold. Set the basic weights of the first product and the second product. For example, the basic weight of the first product is 0.5, and the basic weight of the second product is 0.5. If the difference is greater than zero, it means that the number of times the user browses ethnic cultural products in the second preset time period exceeds the sixth threshold, indicating that the user has a high interest in ethnic cultural products. The weight of the second product should be increased and the weight of the first product should be reduced. Reduce the basic weight of the first product in proportion to obtain the first weight of the first product: first weight = basic weight of the first product - (basic weight of the first product × ratio). Increase the basic weight of the second product in proportion to obtain the second weight of the second product: second weight = basic weight of the second product + (basic weight of the second product × ratio). If the difference is less than zero, it means that the number of times the user browsed the national cultural products in the second preset time period is lower than the sixth threshold, indicating that the user has a low interest in national cultural products. The weight of the first product should be increased and the weight of the second product should be reduced. Increase the basic weight of the first product in proportion to obtain the first weight of the first product: first weight = basic weight of the first product + (basic weight of the first product × proportion). Reduce the basic weight of the second product in proportion to obtain the second weight of the second product: second weight = basic weight of the second product − (basic weight of the second product × proportion).

[0048] Dynamically adjusting product weights according to the user's browsing behavior within a specific time period enables the recommendation system to flexibly respond to changes in user interests and improve the timeliness and adaptability of recommendations. Weight adjustment is based on the user's current browsing behavior, which can more accurately reflect the user's immediate interests, thereby providing more personalized recommendations. By considering the number of times the user browses within the second preset time period, the impact of the user's current behavior on the recommendation results is emphasized, making the recommendation closer to the user's immediate needs. When the user browses a lot, the weight of the second product (a product with a high matching degree) is increased to ensure that the recommendation is more in line with the user's current interests; when the user browses a little, the weight of the first product (a product purchased by similar users) is increased to ensure that the recommendation is more valuable. The user's current browsing behavior immediately affects the recommendation results, and the user can feel the response of the recommendation system, enhancing the user's sense of participation and satisfaction. By dynamically adjusting the weights, the recommendation system can more accurately recommend products that users are interested in, improving user satisfaction and purchase conversion rate. By calculating the weights and performing weighted processing, a more reasonable recommendation order is generated to ensure that the recommended product order is more in line with the user's current needs and interests. Dynamically adjusting weights can avoid recommending a large number of irrelevant products to users, reduce information overload, and improve the practicality and effectiveness of recommendations. By analyzing users' browsing behavior data, the platform can better understand user needs and market trends, and provide data support for product procurement, inventory management and marketing strategies.

[0049] Optionally, performing weighted processing on the first product and the second product according to the weight to determine the recommendation order includes: The first target product that overlaps between the first product and the second product is recommended as the most prioritized product; Using the similarity as the first matching value of the first product, using the matching degree as the second matching value of the second product, calculating the total score of the first product according to the first matching value and the first weight, and calculating the total score of the second product according to the second matching value and the second weight; The second target product is recommended in descending order of total score, and the second target product is the product between the first product and the second product except the first target product.

[0050] Identify the overlapping products in the first and second products, that is, the products that appear in the purchase records of similar users and in the list of highly matched products. Take these overlapping products as the top priority products, because these products are consistent with the user's historical behavior and current interests, and have a high recommendation value. Take the similarity between the user and similar users as the first matching value of the first product. The higher the similarity, the more consistent the first product is with the user's historical behavior and the recommendations of similar users. Take the matching degree between the user and the product as the second matching value of the second product. The higher the matching degree, the more consistent the second product is with the user's current interests and preferences. Total score of the first product: Calculate the total score of the first product based on the first matching value and the first weight. The formula is: Total score of the first product = first matching value × first weight. Total score of the second product: Calculate the total score of the second product based on the second matching value and the second weight. The formula is: Total score of the second product = second matching value × second weight. Sort the other products (second target products) in the first and second products except the overlapping products from high to low according to the total score. Generate the final recommendation list, ensuring that the overlapping products are at the top and the other products are sorted according to the total score.

[0051] The first target product that overlaps between the first product and the second product is taken as the most preferred product to be recommended, ensuring that the recommended product is consistent with the user's historical behavior (products purchased by similar users) and the user's current interest (products with high matching degree), thereby improving the relevance and accuracy of the recommendation. Prioritizing the recommendation of overlapping products can enhance the user's trust in the recommendation system, because these products meet the conditions of multiple recommendation dimensions at the same time, and users are more likely to accept and purchase them. The similarity is taken as the first matching value of the first product, and the matching degree is taken as the second matching value of the second product, and the total score of each product is calculated through weighted processing. This method comprehensively considers the user's historical behavior and current interest, making the recommendation more comprehensive and accurate. The weight is dynamically adjusted according to the user's browsing behavior within the second preset time period, ensuring that the recommendation result can timely reflect the user's current interest changes, and improving the timeliness and adaptability of the recommendation. The second target product is recommended in order of total score from high to low, ensuring that the recommended product order is more consistent with the user's current needs and interests, and improving the practicality and effectiveness of the recommendation. Through reasonable sorting, it is possible to avoid recommending a large number of irrelevant products to users, reduce information overload, and improve user satisfaction and purchase conversion rate. By accurately calculating the total score and optimizing the recommendation order, the recommendation system can more accurately recommend products that meet the user's interests and needs, thereby improving user satisfaction and engagement. The user's current browsing behavior immediately affects the recommendation results, and the user can feel the response of the recommendation system, which enhances the user's sense of engagement and satisfaction.

[0052] This embodiment also discloses a system for online customization and recommendation of personalized ethnic cultural products. Figure 2is a module diagram of a system for online customization and recommendation of personalized ethnic cultural products disclosed in an embodiment of the present application, such as Figure 2 As shown, the system includes a label module 201, a portrait module 202, a first recommendation module 203, a second recommendation module 204, an execution module 205 and a customization module 206, wherein: The tag module 201 is configured to obtain the registration information, behavior data and consumption data of the target user, generate a cultural preference tag according to the behavior data, and generate a consumption habit tag according to the consumption data, wherein the registration information includes age, gender, occupation, and region, the behavior data includes browsing tracks, search keywords, and product information in favorites on the platform, and the consumption data includes purchase frequency and purchase price; A portrait module 202, configured to construct a user portrait of the target user according to the registration information, the cultural preference tag and the consumption habit tag; A first recommendation module 203 is configured to calculate the similarity between users according to the user portraits, determine a first user whose similarity is greater than a first threshold, and obtain a first product purchased by the first user within a first preset time period; The second recommendation module 204 is configured to extract characteristic information of ethnic cultural products from the product database, the characteristic information including cultural elements, style characteristics, material, craftsmanship and price, determine the matching degree between the target user and the characteristic information according to the user portrait, and determine the second product with a matching degree greater than a second threshold; An execution module 205 is configured to determine the weights of the first product and the second product according to the number of times the target user browses ethnic cultural products within a second preset time period, perform weighted processing on the first product and the second product according to the weights to determine a recommendation order, and make recommendations according to the recommendation order; The customization module 206 is configured to receive the target user's feedback information on the recommended product for customization.

[0053] Optionally, the label module 201 is configured to: Determining a second number of times the target user browses each national cultural product category according to the first number of times the target user browses each national cultural product page; Determining a second time that the target user stays in each ethnic cultural product category according to the first time that the target user stays on each ethnic cultural product page; Generate a first cultural preference sub-label according to a first target ethnic cultural product category, wherein the first target ethnic cultural product category is an ethnic cultural product category whose second number is greater than a third threshold and whose second time is greater than a fourth threshold; generating a second cultural preference sub-label according to a second target ethnic cultural product category, wherein the second target ethnic cultural product category is the ethnic cultural product category to which the target ethnic cultural product belongs, and the target ethnic cultural product is the ethnic cultural product corresponding to the keyword whose search frequency exceeds the fifth threshold; Generate a third cultural preference sub-label based on the cultural elements and style characteristics of the national cultural products in the favorites; A cultural preference tag is generated according to the first cultural preference subtag, the second cultural preference subtag, and the third cultural preference subtag.

[0054] Optionally, the label module 201 is configured to: Generate a first consumption habit sub-tag according to the target purchase frequency of the target user within a third preset time period; Generating a second consumption habit sub-label according to the median and maximum values ​​of the purchase prices of the target user within a third preset time period; A consumption habit tag is generated according to the first consumption habit sub-tag and the second consumption habit sub-tag.

[0055] Optionally, the first recommendation module 203 is configured to: Converting the registration information of the user portrait into a first feature vector, converting the cultural preference label of the user portrait into a second feature vector, and converting the consumption habit label of the user portrait into a third feature vector; The first feature vector, the second feature vector and the third feature vector are combined into a comprehensive feature vector, and the similarity between the comprehensive feature vector of the target user and the comprehensive feature vectors of other users is calculated, where the other users are users other than the target user in the national cultural product platform.

[0056] Optionally, the second recommendation module 204 is configured to: Converting the feature information into a fourth feature vector, calculating a first matching degree between the second feature vector and the fourth feature vector, and calculating a second matching degree between the third feature vector and the fourth feature vector; The first matching degree and the second matching degree are weighted to determine a matching degree between the target user and the feature information.

[0057] Optionally, the execution module 205 is configured to: Calculate the difference between the third time the target user browses the national cultural product within the second preset time period and a sixth threshold, and calculate the ratio of the difference to the sixth threshold; When the difference is greater than zero, reducing the basic weight of the first product according to the proportion to obtain a first weight of the first product, and increasing the basic weight of the second product according to the proportion to obtain a second weight of the second product; When the difference is less than zero, the basic weight of the first product is increased according to the proportion to obtain the first weight of the first product, and the basic weight of the second product is reduced according to the proportion to obtain the second weight of the second product.

[0058] Optionally, the execution module 205 is configured to: The first target product that overlaps between the first product and the second product is recommended as the most prioritized product; Using the similarity as the first matching value of the first product, using the matching degree as the second matching value of the second product, calculating the total score of the first product according to the first matching value and the first weight, and calculating the total score of the second product according to the second matching value and the second weight; The second target product is recommended in descending order of total score, and the second target product is the product between the first product and the second product except the first target product.

[0059] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0060] This embodiment also discloses an electronic device, referring to Figure 3 The electronic device may include: at least one processor 301 , at least one communication bus 302 , a user interface 303 , a network interface 304 , and at least one memory 305 .

[0061] The communication bus 302 is used to realize the connection and communication between these components.

[0062] The user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0063] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0064] Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301, and it can be implemented separately through a chip.

[0065] Among them, the memory 305 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may also be optionally at least one storage device located away from the aforementioned processor 301. As Figure 3 As shown, the memory 305 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program of a method for online customization and recommendation of personalized ethnic cultural products.

[0066] exist Figure 3In the electronic device shown, the user interface 303 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 301 can be used to call the application program stored in the memory 305 for the method of online customized recommendation of personalized ethnic cultural products. When executed by one or more processors 301, the electronic device executes one or more methods in the above-mentioned embodiments.

[0067] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0068] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0069] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0070] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0071] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0072] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory 305 and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory 305 includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk.

[0073] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification, those skilled in the art will easily think of other embodiments of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The description and examples are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for online customization and recommendation of personalized ethnic cultural products, characterized in that: Applied to the national cultural product platform, the method includes: Obtain the target user's registration information, behavior data and consumption data, generate a cultural preference tag based on the behavior data, and generate a consumption habit tag based on the consumption data. The registration information includes age, gender, occupation, and region. The behavior data includes browsing history, search keywords, and product information in favorites on the platform. The consumption data includes purchase frequency and purchase price. Constructing a user profile of the target user according to the registration information, the cultural preference tag, and the consumption habit tag; Calculate the similarity between users based on the user portraits, determine a first user whose similarity is greater than a first threshold, and obtain a first product purchased by the first user within a first preset time period; Extracting characteristic information of ethnic cultural products from a product database, the characteristic information including cultural elements, style characteristics, material, craftsmanship and price, determining a matching degree between the target user and the characteristic information according to the user portrait, and determining a second product with a matching degree greater than a second threshold; Determine the weights of the first product and the second product according to the number of times the target user browses ethnic cultural products within a second preset time period, perform weighted processing on the first product and the second product according to the weights to determine a recommendation order, and make recommendations according to the recommendation order; Receive feedback information from the target user on the recommended product for customization.

2. The method for online customization and recommendation of personalized ethnic cultural products according to claim 1, characterized in that: Generating a cultural preference label according to the behavior data includes: Determining a second number of times the target user browses each national cultural product category according to the first number of times the target user browses each national cultural product page; Determining a second time that the target user stays in each ethnic cultural product category according to the first time that the target user stays on each ethnic cultural product page; Generate a first cultural preference sub-label according to a first target ethnic cultural product category, wherein the first target ethnic cultural product category is an ethnic cultural product category whose second number is greater than a third threshold and whose second time is greater than a fourth threshold; generating a second cultural preference sub-label according to a second target ethnic cultural product category, wherein the second target ethnic cultural product category is the ethnic cultural product category to which the target ethnic cultural product belongs, and the target ethnic cultural product is the ethnic cultural product corresponding to the keyword whose search frequency exceeds the fifth threshold; Generate a third cultural preference sub-label based on the cultural elements and style characteristics of the national cultural products in the favorites; A cultural preference tag is generated according to the first cultural preference subtag, the second cultural preference subtag, and the third cultural preference subtag.

3. The method for online customization and recommendation of personalized ethnic cultural products according to claim 1, characterized in that: Generating a consumption habit label according to the consumption data includes: Generate a first consumption habit sub-tag according to the target purchase frequency of the target user within a third preset time period; Generating a second consumption habit sub-label according to the median and maximum values ​​of the purchase prices of the target user within a third preset time period; A consumption habit tag is generated according to the first consumption habit sub-tag and the second consumption habit sub-tag.

4. The method for online customization and recommendation of personalized ethnic cultural products according to claim 1, characterized in that: Calculating the similarity between users based on user portraits includes: Converting the registration information of the user portrait into a first feature vector, converting the cultural preference label of the user portrait into a second feature vector, and converting the consumption habit label of the user portrait into a third feature vector; The first feature vector, the second feature vector and the third feature vector are combined into a comprehensive feature vector, and the similarity between the comprehensive feature vector of the target user and the comprehensive feature vectors of other users is calculated, where the other users are users other than the target user in the national cultural product platform.

5. The method for online customization and recommendation of personalized ethnic cultural products according to claim 4, characterized in that: Determining the matching degree between the target user and the characteristic information according to the user portrait includes: Converting the feature information into a fourth feature vector, calculating a first matching degree between the second feature vector and the fourth feature vector, and calculating a second matching degree between the third feature vector and the fourth feature vector; The first matching degree and the second matching degree are weighted to determine a matching degree between the target user and the feature information.

6. The method for online customization and recommendation of personalized ethnic cultural products according to claim 1, characterized in that: Determining the weights of the first product and the second product according to the number of times the target user browses the ethnic cultural product within the second preset time period includes: Calculate the difference between the third time the target user browses the national cultural product within the second preset time period and a sixth threshold, and calculate the ratio of the difference to the sixth threshold; When the difference is greater than zero, reducing the basic weight of the first product according to the proportion to obtain a first weight of the first product, and increasing the basic weight of the second product according to the proportion to obtain a second weight of the second product; When the difference is less than zero, the basic weight of the first product is increased according to the proportion to obtain the first weight of the first product, and the basic weight of the second product is reduced according to the proportion to obtain the second weight of the second product.

7. The method for online customization and recommendation of personalized ethnic cultural products according to claim 6, characterized in that: The step of performing weighted processing on the first product and the second product according to the weight to determine the recommendation order includes: The first target product that overlaps between the first product and the second product is recommended as the most prioritized product; Using the similarity as the first matching value of the first product, using the matching degree as the second matching value of the second product, calculating the total score of the first product according to the first matching value and the first weight, and calculating the total score of the second product according to the second matching value and the second weight; The second target product is recommended in descending order of total score, and the second target product is the product between the first product and the second product except the first target product.

8. A system for online customization and recommendation of personalized ethnic cultural products, characterized in that: It includes a label module, a portrait module, a first recommendation module, a second recommendation module, an execution module and a customization module, among which: A label module is configured to obtain registration information, behavior data and consumption data of target users, generate cultural preference labels based on the behavior data, and generate consumption habit labels based on the consumption data, wherein the registration information includes age, gender, occupation, and region, the behavior data includes browsing tracks, search keywords, and product information in favorites on the platform, and the consumption data includes purchase frequency and purchase price; A portrait module, configured to construct a user portrait of the target user according to the registration information, the cultural preference tag and the consumption habit tag; A first recommendation module, configured to calculate the similarity between users according to the user portraits, determine a first user whose similarity is greater than a first threshold, and obtain a first product purchased by the first user within a first preset time period; A second recommendation module is configured to extract characteristic information of ethnic cultural products from a product database, the characteristic information including cultural elements, style characteristics, material, craftsmanship and price, determine a matching degree between the target user and the characteristic information according to the user portrait, and determine a second product whose matching degree is greater than a second threshold; an execution module, configured to determine weights of the first product and the second product according to the number of times the target user browses ethnic cultural products within a second preset time period, perform weighted processing on the first product and the second product according to the weights to determine a recommendation order, and make recommendations according to the recommendation order; The customization module is configured to receive the target user's feedback information on the recommended product for customization.

9. An electronic device, characterized in that: It includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is executed.

Images