A virtual display method, device and equipment for cultural and creative products

Abstract

The present invention relates to the field of data processing technology, and in particular to a method, device and equipment for virtual display of cultural and creative products, the method comprising: determining the peak date of attention to cultural and creative products according to user operation information on cultural and creative products; determining the user's hot spot following enthusiasm according to the peak date and the new date of cultural and creative products; determining the similarity between different users according to the hot spot following enthusiasm between different users, and obtaining the user's similarity matrix; determining the first target cultural and creative product that is hot in the cultural and creative products according to the product information of the cultural and creative products and the text of the hot events within the second predetermined time period; selecting the target user; determining the damping coefficient of the target clustering algorithm; using the target clustering algorithm with the damping coefficient and the similarity matrix to recommend the second target cultural and creative product to the current user, and virtually displaying the second target cultural and creative product. The present invention improves the accuracy of the clustering results and improves the accuracy of personalized recommendations for users.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method, device and equipment for virtual display of cultural and creative products. Background Art

[0002] Cultural and creative products are products that integrate creative elements such as culture, art, and design into daily necessities or souvenirs to form commodities with unique cultural connotations and market value. Cultural and creative products can not only inherit and promote culture, but also promote economic development and enhance brand image. Cultural and creative products can be modeled through three-dimensional technology (3D) and then virtually displayed. Virtual display allows users to see product details and understand cultural connotations, and can also enhance users' desire to buy and promote purchasing behavior.

[0003] In some scenarios, in order to make the user's online experience better, the clustering algorithm based on message passing (Affinity Propagation, AP) is used to analyze the user's behavioral characteristics to cluster the users, and to make personalized recommendations for the current user by analyzing the user behavior of other users in the same cluster as the current user. The damping coefficient in the AP clustering algorithm is used to control the convergence speed and stability of the AP clustering algorithm. At present, the damping coefficient is often set in advance manually. However, in actual applications, some cultural and creative products attract users by taking advantage of hot spots. Users who are more familiar with hot spots often do not choose to buy or pay attention to such cultural and creative products that take advantage of hot spots. In this way, it is easy to classify such users into the category of not following hot spots, making the pre-set damping coefficient no longer applicable to the current situation, thereby affecting the accuracy of the clustering results, resulting in low accuracy of personalized recommendations for users. Summary of the invention

[0004] In order to solve the technical problem of low accuracy in personalized recommendations for users, the purpose of the present invention is to provide a method, device and equipment for virtual display of cultural and creative products. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present invention provides a method for virtual display of cultural and creative products, including: determining a peak date of user attention to the cultural and creative product based on user operation information on the cultural and creative product within a first predetermined time period; determining the user's hot spot following enthusiasm based on the peak date and the new release date of the cultural and creative product; determining the similarity between different users based on the hot spot following enthusiasm between different users to obtain a user similarity matrix; determining a first target cultural and creative product that is riding on the hot spot among cultural and creative products based on product information of the cultural and creative product and the text of the hot event within a second predetermined time period; selecting users whose hot spot following enthusiasm is greater than a first threshold as target users for following the hot spot; determining a damping coefficient of a target clustering algorithm based on the peak date of the target user's attention to the first target cultural and creative product, the new release date of the first target cultural and creative product, and the target user's attention on the peak date; using the target clustering algorithm with the damping coefficient and the similarity matrix to recommend a second target cultural and creative product to the current user, and virtually displaying the second target cultural and creative product.

[0006] Optionally, determining the peak date of user's attention to the cultural and creative product based on the user's operation information on the cultural and creative product within the first predetermined time period includes: determining the user's residence time on the page of the cultural and creative product, the first number of times the user zooms in on the cultural and creative product, the second number of times the user rotates the cultural and creative product, and the type of target operation the user performs on the cultural and creative product in the operation information; determining the user's attention to the cultural and creative product based on the residence time, the first number, the second number, and the type of target operation the user performs on the cultural and creative product; and determining the date corresponding to the maximum value in the attention as the peak date.

[0007] Optionally, determining the user's enthusiasm for following hot spots based on the peak date and the new release date of the cultural and creative products includes: determining a peak date sequence of the user's attention to the cultural and creative products based on the peak date, and determining a new release date sequence of the cultural and creative products based on the new release date of the cultural and creative products; calculating the Pearson correlation between the peak date sequence and the new release date sequence to obtain the user's enthusiasm for following hot spots.

[0008] Optionally, determining the similarity between different users based on the hotspot following enthusiasm between different users includes: calculating the absolute value of a first difference between the hotspot following enthusiasm of different users; normalizing the reciprocal of the absolute value of the first difference and a first sum of a preset value to obtain the similarity between different users.

[0009] Optionally, determining the first target cultural and creative product of the cultural and creative product based on the product information of the cultural and creative product and the text of the hot event within the second predetermined time period includes: determining a first word segmentation sequence of the product information of the cultural and creative product and a second word segmentation sequence of the title text of the hot event within the second predetermined time period; calculating the text similarity between the first word segmentation sequence and the second word segmentation sequence; determining the hot event corresponding to the maximum value in the text similarity as the target hot event of the cultural and creative product; determining a keyword set of the product information of the cultural and creative product, and the full text of the target hot event; determining the possibility of the cultural and creative product to take advantage of the hot topic based on the number of times each keyword in the keyword set appears in the full text; and selecting the cultural and creative product whose possibility is less than a second threshold as the first target cultural and creative product to take advantage of the hot topic.

[0010] Optionally, determining the possibility of a cultural and creative product taking advantage of a hot topic based on the number of times each keyword in the keyword set appears in the full text includes: averaging the number of times each keyword appears in the full text to obtain an average number of times; and normalizing the average number of times to obtain the possibility of the cultural and creative product taking advantage of a hot topic.

[0011] Optionally, determining the damping coefficient of the target clustering algorithm based on the peak date of target users' attention to the first target cultural and creative product, the update date of the first target cultural and creative product, and the target users' attention on the peak date includes: determining the influence of the first target cultural and creative product on the clustering results based on the peak date of target users' attention to the first target cultural and creative product, the update date of the first target cultural and creative product, and the target users' attention on the peak date; averaging the influence of all the first target cultural and creative products on the clustering results to obtain an average influence; and determining the average influence as the damping coefficient of the target clustering algorithm.

[0012] Optionally, determining the degree of influence of the first target cultural and creative product on the clustering results based on the peak date of target users' attention to the first target cultural and creative product, the date on which the first target cultural and creative product is updated, and the target users' attention on the peak date includes: calculating the second difference between the peak date of target users' attention to the first target cultural and creative product and the date on which the first target cultural and creative product is updated, and calculating the second sum of the target users' attention on the peak date and a preset value; superimposing the second difference and the second sum of each target user to obtain a superimposed value; and normalizing the superimposed value to obtain the degree of influence of the first target cultural and creative product on the clustering results.

[0013] In a second aspect, an embodiment of the present invention provides a virtual display device for cultural and creative products, including: a determination module, which is used to determine the peak date of users' attention to cultural and creative products based on the user's operation information on the cultural and creative products within a first predetermined time period; the determination module is also used to determine the user's hot spot following enthusiasm based on the peak date and the new release date of the cultural and creative products; the determination module is also used to determine the similarity between different users based on the hot spot following enthusiasm between different users to obtain the user's similarity matrix; the determination module is also used to determine the first target cultural and creative product that is riding on the hot spot among the cultural and creative products based on the product information of the cultural and creative products and the text of the hot event within the second predetermined time period; the selection module is used to select users whose hot spot following enthusiasm is greater than a first threshold as target users for following the hot spot; the determination module is also used to determine the damping coefficient of the target clustering algorithm based on the peak date of the target user's attention to the first target cultural and creative product, the new release date of the first target cultural and creative product, and the target user's attention on the peak date; the display module is used to recommend a second target cultural and creative product to the current user using the target clustering algorithm of the damping coefficient and the similarity matrix, and to virtually display the second target cultural and creative product.

[0014] In a third aspect, an embodiment of the present invention provides a virtual display device for cultural and creative products, comprising: a processor and a memory; wherein the memory is used to store computer programs that can be run on the processor; and the processor is used to execute the program stored in the memory to implement the steps of the virtual display method of cultural and creative products mentioned in the first aspect.

[0015] The present invention has the following beneficial effects: first, the peak date of user's attention to the cultural and creative product is determined according to the user's operation information on the cultural and creative product within a first predetermined time period, and then the user's hot spot following enthusiasm is determined according to the peak date and the new release date of the cultural and creative product; then the similarity between different users is determined according to the hot spot following enthusiasm between different users, and the user's similarity matrix is ​​obtained; secondly, according to the product information of the cultural and creative product and the text of the hot event within a second predetermined time period, the first target cultural and creative product that is riding on the hot spot among the cultural and creative products is determined; and users whose hot spot following enthusiasm is greater than a first threshold are selected as target users for following the hot spot; then the damping coefficient of the target clustering algorithm is determined according to the peak date of the target user's attention to the first target cultural and creative product, the new release date of the first target cultural and creative product, and the target user's attention on the peak date; finally, the target clustering algorithm with the damping coefficient and the similarity matrix are used to recommend the second target cultural and creative product to the current user, and the second target cultural and creative product is virtually displayed.

[0016] In this way, the embodiment of the present invention can calculate the user's enthusiasm for following hot spots according to the new release date of cultural and creative products and the peak date of user attention to cultural and creative products, and obtain the similarity between different users. Then, the cultural and creative products that are riding on the hot spots and the users who may follow the hot spots are obtained by analyzing. The damping coefficient in the target clustering algorithm is updated in this way, so that the clustering results of the target clustering algorithm are more accurate, so as to recommend cultural and creative products to users based on the similarity matrix between the target clustering algorithm and the users. In this way, the accuracy of the clustering results is improved, and the accuracy of personalized recommendations for users is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A flowchart of a method for virtual display of cultural and creative products provided by one embodiment of the present invention;

[0019] Figure 2 A schematic diagram of a correlation curve between a new release date of a cultural and creative product and a peak date of user attention to the cultural and creative product provided by an embodiment of the present invention;

[0020] Figure 3 A schematic diagram of the module composition of a virtual display device for cultural and creative products provided by one embodiment of the present invention;

[0021] Figure 4 A schematic structural diagram of a virtual display device for cultural and creative products provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the virtual display method, device and equipment of a cultural and creative product proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0023] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0024] The following is a detailed description of a method, device and equipment for virtual display of cultural and creative products provided by the present invention in conjunction with the accompanying drawings. Embodiment 1

[0025] See also Figure 1 , which shows a flow chart of a virtual display method for cultural and creative products provided by an embodiment of the present invention, including:

[0026] S101, determining a peak date when users pay attention to a cultural and creative product according to operation information of users on the cultural and creative product within a first predetermined time period.

[0027] Specifically, 3D modeling software can be used to create high-precision 3D models for cultural and creative products and conduct virtual displays. Users can view product details in all directions through operations such as rotation and zooming. Users can place orders to purchase or collect cultural and creative products they like.

[0028] Furthermore, the first predetermined time period can be determined according to actual conditions, and in the embodiment of the present invention, it can be within one year after the cultural and creative product is launched. Operation information includes but is not limited to the purchase, collection, zoom-in operation, rotation operation of the cultural and creative product, the user's stay time on the page of the cultural and creative product, the first number of times the user zooms in on the cultural and creative product, the second number of times the user rotates the cultural and creative product, and the type of target operation performed by the user on the cultural and creative product. The type of target operation performed by the user on the cultural and creative product includes but is not limited to the type of purchase and collection of the cultural and creative product by the user.

[0029] Furthermore, cultural and creative products are related to hot events, and some users will buy or pay more attention to currently popular cultural and creative products because of following hot topics. These users tend to be highly sensitive to current hot topics. Another part of users who do not follow hot topics only buy or pay more attention to cultural and creative products based on their personal preferences. In addition, in order to attract attention and increase sales, some cultural and creative products may deliberately associate with hot topics, and there is a phenomenon of riding on hot topics, while users who are more familiar with hot news often do not choose to buy or pay attention to such hot topic products. When classifying users, it is easy to classify such users into the category of not following hot topics, which will have an adverse effect on the accuracy of the clustering results of cultural and creative product users. The embodiment of the present invention analyzes the influence range of cultural and creative products that may ride on hot topics, updates the damping coefficient in the target clustering algorithm, and makes the clustering results more accurate. When a user follows a hot topic, it means that the user will pay attention to the cultural and creative product immediately after the cultural and creative product is updated. Therefore, the date of the new release of the cultural and creative product is correlated with the time when the user pays attention to the cultural and creative product. By analyzing the degree of attention of the user to the cultural and creative product, the peak date of the user's attention to a single cultural and creative product is determined.

[0030] Furthermore, after a new cultural and creative product is launched, the user's operations of viewing the cultural and creative product on that day, such as purchase, collection, zooming in and rotating, can show the user's attention to the cultural and creative product on that day. When a user views the detailed information of a cultural and creative product for a long time on a certain day, interactive operations such as zooming in and rotating are performed. And when the collection and purchase operations are performed on the same day, it means that the user pays more attention to a certain cultural and creative product on that day. Therefore, the peak date is determined by analyzing the user's operation information on the cultural and creative product. As an optional embodiment of the present invention, first determine the user's stay time on the page of the cultural and creative product in the operation information, the first number of times the user zooms in on the cultural and creative product, the second number of times the user rotates the cultural and creative product, and the type of target operation performed by the user on the cultural and creative product; then determine the user's attention to the cultural and creative product based on the stay time, the first number, the second number, and the type of target operation performed by the user on the cultural and creative product; finally determine that the date corresponding to the maximum value in the attention is the peak date.

[0031] Specifically, the embodiment of the present invention uses the following formula to calculate the user's attention to the cultural and creative product:

[0032]

[0033] In the above formula, Indicates the user's attention to the cultural and creative product on the tth day. C[] indicates the reference features in the operation information of the user's operation on the cultural and creative product on the tth day after the cultural and creative product is launched, including: the stay time on the cultural and creative product page on the tth day (min), the first number of times the cultural and creative product is enlarged on the tth day , and the second number of rotation operations on the cultural and creative product on the tth day The longer the user stays and the more frequently he / she operates on the cultural and creative product on the same day, the more attention the user pays to the cultural and creative product on that day. The larger the value, the larger the value. B[] represents the type of operation behavior on the cultural and creative product on the tth day, including whether the user purchases or collects the cultural and creative product on that day. Indicates the value corresponding to the user's purchase and collection of the cultural and creative product. If the user both purchases and collects the cultural and creative product on the same day, then The value is 2. If the user purchases or collects the cultural and creative product on the same day, The value is 1. If the user neither purchased nor collected the cultural and creative product on that day, The value is 0. Represents the normalization function, which is used to Perform normalization.

[0034] Furthermore, after determining the user's attention to each cultural and creative product in the manner described in the above embodiment of the present invention, the date corresponding to the maximum value of the attention of each cultural and creative product is determined as the day when the user pays the most attention to each cultural and creative product, that is, the peak date. Thus, the peak date of the user's attention to each cultural and creative product can be obtained for all the cultural and creative products launched in the past year. .

[0035] S102, determining the user's enthusiasm for following hot topics based on the peak date and the release date of the cultural and creative products.

[0036] Specifically, when users are highly motivated to follow hot topics, they will pay more attention to cultural and creative products as soon as they are released. At this time, the release date of cultural and creative products and the peak date of users’ attention to the cultural and creative products are more correlated. For example, if cultural and creative product A is released on February 3, the peak date of users’ attention to the cultural and creative product is February 25; if cultural and creative product B is released on March 10, the peak date of users’ attention to the cultural and creative product is March 12. When users are less motivated to follow hot topics, they will pay attention to cultural and creative products based on their preferences. Therefore, the peak date of users’ attention may be a long time after the cultural and creative product is released, that is, the popularity of the cultural and creative product has passed before users purchase or pay attention to it. For example, if Figure 2 As shown, Figure 2 A schematic diagram of a correlation curve between the release date of a cultural and creative product and the peak date of user attention to the cultural and creative product provided by an embodiment of the present invention is shown in FIG. Figure 2 As shown in the figure, the trends between the launch day of cultural and creative products and the peak day of user attention are roughly similar, and the two are correlated.

[0037] Furthermore, when determining the user's enthusiasm for following hot spots, as an optional embodiment of the present invention, first, the peak date sequence of the user's attention to cultural and creative products is determined based on the peak date, and the new release date sequence of cultural and creative products is determined based on the new release date of the cultural and creative products; then the Pearson correlation of the peak date sequence and the new release date sequence is calculated to obtain the user's enthusiasm for following hot spots.

[0038] Specifically, the embodiment of the present invention may use the following formula to calculate the user's hotspot following enthusiasm:

[0039]

[0040] In the above formula, Indicates the hotspot following enthusiasm of user v. Represents the peak date sequence of user v's attention to each cultural and creative product i. represents the sequence of new release dates of various cultural and creative products that user v follows. R[] is the Pearson correlation coefficient. At this time, when the correlation between the new release date of a cultural and creative product and the peak date of attention to the cultural and creative product is greater, that is, when R is closer to 1, it means that the user is more active in following hot spots, that is, The bigger.

[0041] S103, determining the similarity between different users according to the hot spot following enthusiasm between different users, and obtaining a user similarity matrix.

[0042] Specifically, when clustering is performed using a clustering algorithm, it is first necessary to create a similarity matrix for each user's purchase based on the similarities between users. In the present invention, the more similar the user's features are, the smaller the difference in the users' hotspot following enthusiasm is. Therefore, the similarity between users is represented by the difference in hotspot following enthusiasm between two users.

[0043] Furthermore, when determining the similarity between different users, as an optional embodiment of the present invention, the absolute value of the first difference between the hotspot following enthusiasm of different users is first calculated; then the absolute value of the first difference and the inverse of the first sum of the preset values ​​are normalized to obtain the similarity between different users.

[0044] Specifically, the preset value can be determined according to actual conditions, and in the embodiment of the present invention, the value is 1. The embodiment of the present invention specifically uses the following formula to calculate the similarity between any two users:

[0045]

[0046] In the above formula, Represents the similarity between user v and user w. represents the difference between the hotspot following enthusiasm of user v and user w. represents the similarity between user v and user w. The similarity here represents whether any two users pay attention to cultural and creative products based on hot spots. At this time, the similarity matrix of all users can be constructed. Represents the normalization function, which is used to Perform normalization.

[0047] S104, determining a first target cultural and creative product among the cultural and creative products that is riding on the hot topic based on the product information of the cultural and creative product and the text of the hot topic event within the second predetermined time period.

[0048] Specifically, the product information of cultural and creative products refers to the detailed information that introduces the functions of cultural and creative products. The text of hot events includes but is not limited to the title of the hot time and the main text. Among them, the damping coefficient in the target clustering algorithm is used to control the convergence speed and stability of the algorithm. The target often uses an artificial setting to set the damping coefficient. In practical applications, cultural and creative products that take advantage of hot spots will cause some users to be misclassified into categories that do not follow hot spots. Users who do not pay attention to and purchase products that take advantage of hot spots in real time are misjudged as being insensitive to hot spots. This makes the pre-set damping coefficient no longer applicable, thereby interfering with the accuracy of the clustering results. Therefore, the damping coefficient in the clustering algorithm can be determined by calculating the degree of influence of cultural and creative products that take advantage of hot spots on the clustering results.

[0049] Furthermore, cultural and creative products are the product of hot events, and cultural and creative products that take advantage of hot events often superficially use hot events, such as simply using the name of the hot event, the main characters, etc. as part of the product name or as a product introduction, to attract consumers' attention. Therefore, by calculating the correlation between the product function of each cultural and creative product and the hot event, the possibility of each cultural and creative product taking advantage of the hot event is determined, and each cultural and creative product that takes advantage of the hot event is determined.

[0050] Furthermore, when determining the first target cultural and creative product to take advantage of the hot topic, as an optional embodiment of the present invention, first determine a first word segmentation sequence of the product information of the cultural and creative product and a second word segmentation sequence of the title text of the hot event within a second predetermined time period; then calculate the text similarity between the first word segmentation sequence and the second word segmentation sequence; then determine that the hot event corresponding to the maximum value in the text similarity is the target hot event of the cultural and creative product; secondly determine a keyword set of the product information of the cultural and creative product, and the full text of the target hot event; and determine the possibility of the cultural and creative product taking advantage of the hot topic based on the number of times each keyword in the keyword set appears in the full text; finally, select a cultural and creative product whose possibility is less than the second threshold as the first target cultural and creative product to take advantage of the hot topic.

[0051] Specifically, the product information of the cultural and creative product can be a product introduction text that introduces the functions of the cultural and creative product. The second predetermined time period can be determined according to actual conditions. In the embodiment of the present invention, the value is taken as the last three months of the new release date of the cultural and creative product. The first word segmentation sequence and the second word segmentation sequence can be obtained by using the jieba word segmentation tool. The hot event can be hot news, etc.

[0052] Furthermore, when calculating the text similarity between the first word segmentation sequence and the second word segmentation sequence, the following formula can be used:

[0053]

[0054] In the above formula, Indicates the text similarity between the first word segmentation sequence and the second word segmentation sequence of the jth hot event. represents the first participle sequence, It represents the second segmentation sequence of the jth hot event, and jieba means using the jieba segmentation tool.

[0055] Through the above embodiment of the present invention, the text similarity between a certain cultural and creative product and each hot news can be obtained. The hot event corresponding to the maximum value is the target hot event of the cultural and creative product.

[0056] Furthermore, when determining the keyword set of the product information of cultural and creative products, the text of the product information of cultural and creative products can be obtained through the weighted technology (TermFrequency-InverseDocumentFrequency, TF-IDF) used for information retrieval and text mining, and the keyword set of the product information of cultural and creative products can be obtained by segmenting the words. Then the full text of the target hot event is obtained. At this time, the more frequently the keywords of the product information of the cultural and creative products appear in the full text of the target hot event, the more it indicates that the design concept of the cultural and creative products is combined with the hot event. The stronger the correlation between the two, the less likely the cultural and creative products are to take advantage of the popularity. When calculating the possibility of cultural and creative products taking advantage of the popularity, first average the number of times each keyword appears in the full text to obtain the average number of times; then normalize the average number of times to obtain the possibility of cultural and creative products taking advantage of the popularity.

[0057] Specifically, the embodiment of the present invention can use the following formula to calculate the possibility of a cultural and creative product riding on the trend:

[0058]

[0059] In the above formula, It indicates the possibility of cultural and creative products taking advantage of the popularity. The number of times the wth keyword in the keyword set of the product information of cultural and creative product i appears in the full text of the hot event corresponding to the cultural and creative product. W represents the number of keywords in the keyword set of the product information of cultural and creative product i. Represents the normalization function, which is used to Perform normalization.

[0060] Furthermore, the second threshold value can be determined according to actual conditions. In the embodiment of the present invention, the value is 0.5. Cultural and creative products with a value less than 0.5 are the first target products for taking advantage of the hot topics.

[0061] S105: Select users whose hotspot following enthusiasm is greater than a first threshold as target users for following the hotspot.

[0062] Specifically, the first threshold can be determined according to actual conditions, and in the embodiment of the present invention, the value is 0.2. Users with a value greater than 0.2 are target users who follow hot spots.

[0063] S106, determining a damping coefficient of a target clustering algorithm according to a peak date when target users pay attention to the first target cultural and creative product, a new release date of the first target cultural and creative product, and the attention degree of target users on the peak date.

[0064] Specifically, the longer the time interval between the peak date of the cultural and creative product followed by the users who follow the hot spot and the new release date of the cultural and creative product, the greater the user's attention on the peak date. The smaller it is, the greater the influence of the cultural and creative products on the clustering results. The target clustering algorithm can be the AP clustering algorithm.

[0065] Furthermore, when determining the damping coefficient of the target clustering algorithm, as an optional embodiment of the present invention, the degree of influence of the first target cultural and creative product on the clustering results is first determined based on the peak date of the target user's attention to the first target cultural and creative product, the date of the new release of the first target cultural and creative product, and the target user's attention on the peak date; then, the degrees of influence of all the first target cultural and creative products on the clustering results are averaged to obtain the average influence degree; finally, the average influence degree is determined as the damping coefficient of the target clustering algorithm.

[0066] Specifically, when determining the degree of influence of the first target cultural and creative product on the clustering results, as an optional embodiment of the present invention, the second difference between the peak date of attention to the first target cultural and creative product and the new release date of the first target cultural and creative product is first calculated, as well as the second sum of the target user's attention on the peak date and the preset value; then the second difference and the second sum of each target user are superimposed to obtain a superimposed value; finally, the superimposed value is normalized to obtain the degree of influence of the first target cultural and creative product on the clustering results.

[0067] Specifically, the preset value can be determined according to actual conditions, and in the embodiment of the present invention, the value is 1. In the embodiment of the present invention, the following formula is used to calculate the influence of the first target cultural and creative product on the clustering result:

[0068]

[0069] In the above formula, Indicates the influence of the first target cultural and creative product on the clustering results. It indicates the peak date when the target user v who may follow the hot topic pays attention to a single first-target cultural and creative product that is riding on the hot topic. Indicates the release date of a single cultural and creative product that is riding on the trend. It indicates the time interval between the peak date when target users pay attention to the cultural and creative product and the release date of the new cultural and creative product that is riding on the trend. The attention level of target user v following a hotspot on the peak date t. V represents the number of target users following a hotspot. Represents the normalization function, which is used to Perform normalization. The larger the value is, the greater the impact of the first target cultural and creative products on the clustering results.

[0070] Furthermore, the influence of each first target cultural and creative product on the clustering result calculated according to the above embodiment of the present invention is used to reflect the influence of the overall cultural and creative products on the clustering result through the mean. In the embodiment of the present invention, the influence of all first target cultural and creative products is averaged using the following formula:

[0071]

[0072] In the above formula, Indicates the average impact. Indicates the number of cultural and creative products that are the first target of the hot topic. It indicates the influence of the k-th target cultural and creative product on the clustering results.

[0073] Furthermore, in order to reduce the impact of users being misclassified on the clustering results, the damping coefficient is determined according to the degree of influence of the cultural and creative products riding on the hot spots on the clustering results. When there are relatively few users affected by the misclassification, that is, when the degree of influence of the cultural and creative products riding on the hot spots on the clustering results is small, the damping coefficient is reduced so that the clustering algorithm can converge faster. This is because the influence of the misclassified users on the overall clustering results is relatively small at this time, and there is no need to excessively reduce the convergence speed to ensure stability. On the contrary, when the degree of influence of the cultural and creative products riding on the hot spots on the clustering results is large, the damping coefficient is increased to make the clustering algorithm more cautious when processing data, so as to avoid excessive deviations in the clustering results due to the misclassification of a large number of users. Therefore, in the embodiment of the present invention, the damping coefficient in the AP clustering algorithm is set .

[0074] S107, using the target clustering algorithm of the damping coefficient and the similarity matrix to recommend the second target cultural and creative product to the current user, and virtually displaying the second target cultural and creative product.

[0075] Specifically, when recommending the second target cultural and creative product to the user, the target clustering algorithm of the damping coefficient and the similarity matrix are used to cluster the users to obtain the clustering results; based on the user behavior of other users in the clustering results, the second target cultural and creative product is recommended to the current user in the clustering results, and the second target cultural and creative product is displayed.

[0076] More specifically, the target clustering algorithm may be an AP algorithm, and the damping coefficient of the AP algorithm is set to be the value calculated in the above embodiment of the present invention. Then, the AP algorithm is used to calculate the user's attractiveness and belongingness based on the similarity matrix. Adjust the update speed of attraction and belonging. Set the number of iterations ,When the iteration conditions are met, the algorithm outputs the clustering results. The clustering results here are clustering users according to the hot spots followed by users. Among them, the clustering results can divide users into Y categories, and make personalized recommendations for the current user based on the behaviors of other users in the cluster where the current user is located. This is because people in the same cluster have the same hobbies, so according to the clustering results, cultural and creative products that other users like can be recommended to the current user.

[0077] Furthermore, the user at the cluster center of the cluster where the current user is located has a low enthusiasm for following hot spots, indicating that the users in the cluster do not like to pay attention to cultural and creative products related to hot spots. Therefore, the cultural and creative products that are purchased the most by other people in the cluster are virtually displayed on the homepage where the user uses cultural and creative products. This cultural and creative product is not a hot cultural and creative product. The user at the cluster center of the cluster where the current user is located has a high enthusiasm for following hot spots, indicating that the users in the cluster like to pay attention to cultural and creative products related to hot spots. Therefore, the cultural and creative products that are purchased the most by other people in the cluster are virtually displayed on the homepage where the user uses cultural and creative products. This cultural and creative product is a hot cultural and creative product.

[0078] The embodiment of the present invention can calculate the user's enthusiasm for following hot spots according to the new release date of cultural and creative products and the peak date of the user's attention to cultural and creative products, and obtain the similarity between different users. Then, the cultural and creative products that are riding on the hot spots and the users who may follow the hot spots are analyzed. The damping coefficient in the target clustering algorithm is updated in this way, so that the clustering results of the target clustering algorithm are more accurate, so as to recommend cultural and creative products to users based on the similarity matrix between the target clustering algorithm and the users. In this way, the accuracy of the clustering results is improved, and the accuracy of personalized recommendations for users is improved. Embodiment 2

[0079] Corresponding to the virtual display method of cultural and creative products provided in the above embodiment, based on the same technical concept, an embodiment of the present invention further provides a virtual display device of cultural and creative products, which is used to execute the virtual display method of cultural and creative products. Figure 3 A schematic diagram of the module composition of a virtual display device for cultural and creative products provided by an embodiment of the present invention is shown in FIG. Figure 3The virtual display device 300 of cultural and creative products includes: a determination module 301, which is used to determine the peak date of users' attention to cultural and creative products according to the operation information of users on cultural and creative products in the first predetermined time period; the determination module 301 is also used to determine the user's hot spot following enthusiasm according to the peak date and the new date of the cultural and creative products; the determination module 301 is also used to determine the similarity between different users according to the hot spot following enthusiasm between different users, and obtain the user similarity matrix; the determination module 301 is also used to determine the hot spot event text in the second predetermined time period according to the product information of the cultural and creative products, Determine a first target cultural and creative product that is riding on the hot spot among cultural and creative products; a selection module 302 is used to select users whose hot spot following enthusiasm is greater than a first threshold as target users who follow the hot spot; the determination module 301 is also used to determine the damping coefficient of the target clustering algorithm according to the peak date when the target user pays attention to the first target cultural and creative product, the new date of the first target cultural and creative product, and the attention of the target user on the peak date; the display module 303 is used to recommend a second target cultural and creative product to the current user by using the target clustering algorithm of the damping coefficient and the similarity matrix, and to virtually display the second target cultural and creative product.

[0080] The embodiment of the present invention can calculate the user's enthusiasm for following hot spots according to the new release date of cultural and creative products and the peak date of the user's attention to cultural and creative products, and obtain the similarity between different users. Then, the cultural and creative products that are riding on the hot spots and the users who may follow the hot spots are analyzed. The damping coefficient in the target clustering algorithm is updated in this way, so that the clustering results of the target clustering algorithm are more accurate, so as to recommend cultural and creative products to users based on the similarity matrix between the target clustering algorithm and the users. In this way, the accuracy of the clustering results is improved, and the accuracy of personalized recommendations for users is improved.

[0081] Optionally, determination module 301 is also used to determine in the operation information the user's dwell time on the page of the cultural and creative product, the first number of times the user zooms in on the cultural and creative product, the second number of times the user rotates the cultural and creative product, and the type of target operation the user performs on the cultural and creative product; determine the user's attention to the cultural and creative product based on the dwell time, the first number, the second number, and the type of target operation the user performs on the cultural and creative product; and determine the date corresponding to the maximum value in the attention as the peak date.

[0082] Optionally, the determination module 301 is also used to determine the peak date sequence of the user's attention to the cultural and creative products based on the peak date, and to determine the new release date sequence of the cultural and creative products based on the new release date of the cultural and creative products; calculate the Pearson correlation between the peak date sequence and the new release date sequence to obtain the user's enthusiasm for following hot spots.

[0083] Optionally, the determination module 301 is further used to calculate the absolute value of the first difference between the hotspot following positivity of different users; normalize the reciprocal of the absolute value of the first difference and the first sum of a preset value to obtain the similarity between different users.

[0084] Optionally, the determination module 301 is also used to determine a first word segmentation sequence of the product information of the cultural and creative product and a second word segmentation sequence of the title text of the hot event within a second predetermined time period; calculate the text similarity between the first word segmentation sequence and the second word segmentation sequence; determine the hot event corresponding to the maximum value in the text similarity as the target hot event of the cultural and creative product; determine a keyword set of the product information of the cultural and creative product, and the full text of the target hot event; determine the possibility of the cultural and creative product to take advantage of the hot topic based on the number of times each keyword in the keyword set appears in the full text; select the cultural and creative product whose possibility is less than the second threshold as the first target cultural and creative product to take advantage of the hot topic.

[0085] Optionally, the determination module 301 is further used to average the number of times each keyword appears in the full text to obtain an average number of times; the average number of times is normalized to obtain the possibility of the cultural and creative product taking advantage of the hot topic.

[0086] Optionally, determination module 301 is also used to determine the influence of the first target cultural and creative product on the clustering results based on the peak date of target users' attention to the first target cultural and creative product, the date of the first target cultural and creative product's new release, and the target users' attention on the peak date; averaging the influence of all first target cultural and creative products on the clustering results to obtain an average influence; and determining the average influence as the damping coefficient of the target clustering algorithm.

[0087] Optionally, the determination module 301 is also used to calculate the second difference between the peak date of attention to the first target cultural and creative product and the new release date of the first target cultural and creative product, and to calculate the second sum of the target user's attention on the peak date and a preset value; superimpose the second difference and the second sum of each target user to obtain a superimposed value; and normalize the superimposed value to obtain the degree of influence of the first target cultural and creative product on the clustering results. Embodiment 3

[0088] Corresponding to the virtual display method of cultural and creative products provided in the above embodiment, based on the same technical concept, an embodiment of the present invention further provides a virtual display device for cultural and creative products, and the virtual display device for cultural and creative products is used to execute the virtual display method for cultural and creative products. Figure 4 A schematic diagram of a virtual display device for cultural and creative products provided by an embodiment of the present invention is shown in FIG. Figure 4The virtual display device of cultural and creative products may have relatively large differences due to different configurations or performances, and may include one or more processors 401 and memory 402, the memory 402 is used to store computer programs that can be run on the processor 401, and the processor 401 is used to execute the program stored in the memory 402 to achieve the above Figure 1 The various steps in the method embodiment. Among them, the memory 402 can be a temporary storage or a permanent storage. The application stored in the memory 402 may include one or more modules (not shown in the figure), and each module may include a series of computer executable instructions in the virtual display device of the cultural and creative products.

[0089] Furthermore, the processor 401 may be configured to communicate with the memory 402 to execute a series of computer executable instructions in the memory 402 on the virtual display device of the cultural and creative product. The virtual display device of the cultural and creative product may also include one or more power supplies 403, one or more wired or wireless network interfaces 404, one or more input and output interfaces 405, and one or more keyboards 406.

[0090] Specifically in this embodiment, the virtual display device of cultural and creative products includes a processor, a communication interface, a memory and a communication bus; wherein the processor, the communication interface and the memory communicate with each other through the bus; the memory is used to store computer programs; the processor is used to execute the programs stored in the memory to achieve the above Figure 1 The various steps in the method embodiment are similar to those in the method embodiment, and have the beneficial effects of the above method embodiments. To avoid repetition, the embodiments of the present invention will not be described in detail here.

[0091] It should be noted that the virtual display device for cultural and creative products provided in the embodiment of the present invention and the virtual display method for cultural and creative products provided in the embodiment of the present invention are based on the same application concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned virtual display method for cultural and creative products, and has the same or similar beneficial effects, and the repetitive parts will not be repeated.

[0092] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0093] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A virtual display method for cultural and creative products, characterized in that: The virtual display method of the cultural and creative products comprises: Determining a peak date when users pay attention to the cultural and creative product according to the operation information of the users on the cultural and creative product within the first predetermined time period; Determining the user's enthusiasm for following hot topics according to the peak date and the new release date of the cultural and creative product; Determine the similarity between different users based on their hot spot following enthusiasm and obtain the user similarity matrix; Determine a first target cultural and creative product among the cultural and creative products according to the product information of the cultural and creative product and the text of the hot event within the second predetermined time period; Selecting users whose hotspot following enthusiasm is greater than a first threshold as target users for following the hotspot; Determining the damping coefficient of the target clustering algorithm according to the peak date when the target user pays attention to the first target cultural and creative product, the date when the first target cultural and creative product is updated, and the attention degree of the target user on the peak date includes: determining the influence of the first target cultural and creative product that takes advantage of the hot spot on the clustering result according to the peak date when the target user pays attention to the first target cultural and creative product, the date when the first target cultural and creative product is updated, and the attention degree of the target user on the peak date; determining the damping coefficient according to the influence of the first target cultural and creative product that takes advantage of the hot spot on the clustering result; The target clustering algorithm of the damping coefficient and the similarity matrix are used to recommend a second target cultural and creative product to the current user, and the second target cultural and creative product is virtually displayed.

2. The virtual display method of cultural and creative products according to claim 1, characterized in that: Determining the peak date of user attention to the cultural and creative product according to the user's operation information on the cultural and creative product within the first predetermined time period includes: Determine, in the operation information, the user's stay time on the page of the cultural and creative product, the first number of times the user zooms in on the cultural and creative product, the second number of times the user rotates the cultural and creative product, and the type of target operation performed by the user on the cultural and creative product; Determining the user's attention to the cultural and creative product according to the dwell time, the first number of times, the second number of times, and the type of target operation performed by the user on the cultural and creative product; The date corresponding to the maximum value of the attention degree is determined as the peak date.

3. The virtual display method of cultural and creative products according to claim 1, characterized in that: The determining of the user's hot spot following enthusiasm according to the peak date and the new release date of the cultural and creative product comprises: Determining a peak date sequence of users' attention to the cultural and creative products according to the peak dates, and determining a new release date sequence of the cultural and creative products according to the new release dates of the cultural and creative products; The Pearson correlation is calculated for the peak date sequence and the new date sequence to obtain the user's hot spot following enthusiasm.

4. The virtual display method of cultural and creative products according to claim 1, characterized in that: Determining the similarity between different users according to the hotspot following enthusiasm between different users includes: Calculate the absolute value of the first difference between the hotspot following positivity of different users; The reciprocal of the first sum of the absolute value of the first difference and the preset value is normalized to obtain the similarity between the different users.

5. The method for virtual display of cultural and creative products according to claim 1, characterized in that: The determining, based on the product information of the cultural and creative product and the text of the hot event within the second predetermined time period, of a first target cultural and creative product among the cultural and creative products to take advantage of the hot event includes: Determine a first word segmentation sequence of product information of the cultural and creative product and a second word segmentation sequence of a title text of a hot event within a second predetermined time period; Calculating text similarity between the first word segmentation sequence and the second word segmentation sequence; Determine the hot event corresponding to the maximum value in the text similarity as the target hot event of the cultural and creative product; Determine a keyword set of product information of the cultural and creative product, and the full text of the target hot event; Determining the possibility of the cultural and creative product riding on the trend according to the number of times each keyword in the keyword set appears in the full text; The cultural and creative product whose probability is less than the second threshold is selected as the first target cultural and creative product for riding on the hot topic.

6. The method for virtual display of cultural and creative products according to claim 5, characterized in that: The method of determining the possibility of the cultural and creative product riding on the trend according to the number of times each keyword in the keyword set appears in the full text includes: Averaging the number of times each keyword appears in the full text to obtain an average number of times; The average number of times is normalized to obtain the possibility of the cultural and creative product taking advantage of the hot topic.

7. The method for virtual display of cultural and creative products according to any one of claims 1 to 6, characterized in that: The step of determining the damping coefficient according to the influence of the first target cultural and creative product on the clustering result includes: Averaging the influence of all the first target cultural and creative products on the clustering results to obtain an average influence; The average influence degree is determined as a damping coefficient of the target clustering algorithm.

8. The method for virtual display of cultural and creative products according to claim 7, characterized in that: The determining the influence of the first target cultural and creative product on the clustering result according to the peak date when the target user pays attention to the first target cultural and creative product, the new release date of the first target cultural and creative product, and the attention degree of the target user on the peak date includes: Calculating a second difference between a peak date when a target user pays attention to a first target cultural and creative product and a new release date of the first target cultural and creative product, and calculating a second sum of the target user's attention on the peak date and a preset value; Superimposing the second difference value and the second sum value of each target user to obtain a superimposed value; The superposition value is normalized to obtain the influence of the first target cultural and creative product on the clustering result.

9. A virtual display device for cultural and creative products, characterized in that: The device is used to execute the method according to claim 1, and the virtual display device of the cultural and creative product comprises: A determination module, configured to determine a peak date when users pay attention to the cultural and creative product according to operation information of the user on the cultural and creative product within a first predetermined time period; The determination module is further used to determine the user's enthusiasm for following hot spots according to the peak date and the new release date of the cultural and creative product; The determination module is further used to determine the similarity between different users according to the hotspot following enthusiasm between different users, and obtain a user similarity matrix; The determination module is further used to determine a first target cultural and creative product among the cultural and creative products that is riding on the trend according to the product information of the cultural and creative product and the text of the hot event within the second predetermined time period; A selection module, configured to select users whose hotspot following enthusiasm is greater than a first threshold as target users for following the hotspot; The determination module is further configured to determine a damping coefficient of a target clustering algorithm according to a peak date when the target user pays attention to the first target cultural and creative product, a new release date of the first target cultural and creative product, and a degree of attention of the target user on the peak date; A display module is used to recommend a second target cultural and creative product to the current user by using the target clustering algorithm of the damping coefficient and the similarity matrix, and to virtually display the second target cultural and creative product.

10. A virtual display device for cultural and creative products, characterized in that: The virtual display device of the cultural and creative product includes: a processor and a memory; wherein the memory is used to store a computer program running on the processor; the processor is used to execute the program stored in the memory to implement the steps of the virtual display method of the cultural and creative product as described in any one of claims 1-8.

Images