Computer service platform for advertisement recommendation

Abstract

The invention relates to the technical field of computer service platforms, and discloses a computer service platform for advertisement recommendation, which comprises a data acquisition module, an analysis module, a platform recommendation module, an evaluation module and a comprehensive feedback module, the analysis module calculates a user characteristic index, an environment characteristic index and a comprehensive influence index according to the data acquired by the data acquisition module, and the platform recommendation module calculates an advertisement matching degree with platform advertisements according to the user characteristic index, the environment characteristic index and the comprehensive influence index; an evaluation module calculates an advertisement recommendation effect, a user feedback index and a recommendation comprehensive index, and a comprehensive feedback module generates a report for advertisement recommendation feedback according to the recommendation comprehensive index, so that the requirements and preferences of the user are accurately grasped, and the advertisement more conforming to the interest is recommended; and the advertisement recommendation effect is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of computer service platforms, and in particular to a computer service platform for advertisement recommendation. Background Art

[0002] The computer service platform is a platform built on cloud computing technology. It provides computing power, storage space, databases, applications and other IT resources to users through the Internet to support various business needs. Users can flexibly select and configure computing resources according to actual needs. As the business grows, users can easily expand computing resources without worrying about hardware limitations or maintenance issues. The platform provides multi-level security mechanisms, including data encryption, access control, firewalls, etc., to ensure the security and privacy protection of user data.

[0003] In today's era of information explosion, advertising is a bridge connecting products and consumers, and its importance is self-evident. However, when faced with new advertisements and new users, traditional advertising systems often find it difficult to effectively expose or recommend content of interest, which not only affects the advertising effect, but may also have a negative impact on user experience. What is more serious is that traditional advertising systems may cause the information users receive to gradually become monotonous, forming problems such as "information cocoons". These problems not only affect users' vision and cognition, but may also have an adverse impact on the overall development of society. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the deficiencies in the prior art, the present invention provides a computer service platform for advertisement recommendation. The advertisement recommendation system can more accurately grasp the needs and preferences of users through user characteristics, thereby recommending advertisement content that is more in line with their interests to users. Advertisement recommendations based on environmental characteristics can improve the accuracy and relevance of advertisements, enhance user experience and satisfaction, optimize advertisement effects and promote commercial conversions. It can also provide merchants with market insights and decision support, and enhance the adaptability and flexibility of advertisements.

[0006] (II) Technical solution

[0007] To achieve the above object, the present invention provides the following technical solution: a computer service platform for advertisement recommendation, characterized in that it includes a data collection module, an analysis module, a platform recommendation module, an evaluation module and a comprehensive feedback module, wherein the data collection module is used to collect user characteristic data and environmental characteristic data, the data collection module is connected to the analysis module via a network, and sends the collected user characteristic data and environmental characteristic data to the analysis module;

[0008] The analysis module calculates the user characteristic index Ytzs, the environmental characteristic index Htzs and the comprehensive impact index Zhyx according to the data collected by the data collection module, and sends the calculated user characteristic index Ytzs, the environmental characteristic index Htzs and the comprehensive impact index Zhyx to the platform recommendation module;

[0009] The platform recommendation module calculates the advertising matching degree Ggpp with the platform advertisements according to the user characteristic index Ytzs, the environmental characteristic index Htzs and the comprehensive impact index Zhyx, extracts the advertisements with higher matching degree for advertisement recommendation, and sends the data to the evaluation module.

[0010] The evaluation module calculates the advertising recommendation effect Tjxg and the user feedback index Yhfk, and calculates the recommendation comprehensive index Zhzb based on the advertising recommendation effect Tjxg and the user feedback index Yhfk, and sends the recommendation comprehensive index Zhzb to the comprehensive feedback module;

[0011] The comprehensive feedback module generates a report using the recommendation comprehensive index Zhzb for advertisement recommendation feedback.

[0012] Preferably, the data collection module collects multiple user feature data to form a data set, and the calculation expression of the data set is: {YHtz1, YHtz2, YHtz3, ..., YHtz n}, in the data set, YHtz1 represents the first user feature data collected by the data collection module, YHtz n It represents the last user characteristic data collected by the data collection module. At the same time, n represents that there are n data collected in total.

[0013] Preferably, the data acquisition module forms a data set with the collected multiple environmental feature data, and the calculation expression of the data set is: {HJtz1, HJtz2, HJtz3, ..., HJtz n}, in the data set, HJtz1 represents the first environmental feature data collected by the data collection module, HJtz n It represents the last environmental characteristic data collected by the data collection module. At the same time, n represents that there are n data collected in total.

[0014] Preferably, the calculation formula of the user characteristic index Ytzs is:

[0015]

[0016] In the above data set, GMjl represents the total purchase amount, Gt represents the number of purchases, LLjl represents the total browsing history, Lt represents the total browsing time, XQah represents hobbies, NLpa represents the overall preference for age, HYfx represents friends sharing, α 1 , α 2 , α 3 , α 4 , α 5 Represent the corresponding weight factors respectively.

[0017] Preferably, the calculation formula of the environmental characteristic index Htzs is:

[0018]

[0019] In the above formula, Jjbh represents seasonal weather changes, Dxlx represents the current popularity index, Zyxq represents the occupational market demand, Bdtj represents local recommendations, and β 1 , β 2 , β 3 , β 4 Represent the corresponding weight factors respectively.

[0020] Preferably, the calculation formula of the comprehensive impact index Zhyx is:

[0021]

[0022] In the above formula, Bz is the standard value of the comprehensive impact index.

[0023] Preferably, the calculation formula of the advertisement matching degree Ggpp is:

[0024]

[0025] In the above formula, GGzz is the advertising standard data existing in the platform.

[0026] Preferably, the advertisements whose advertisement matching degree Ggpp is greater than the minimum advertisement matching degree value are recommended.

[0027] Preferably, the calculation formula of the advertisement recommendation effect Tjxg is:

[0028]

[0029] In the above formula, Zscs represents the number of impressions, and Djcs represents the number of clicks. represents the click-through rate, γ 1 is the corresponding weight factor, Zhcs represents the number of conversions, is the conversion rate, γ 2 is the corresponding weight factor, Trcb represents the input cost, γ3 represents the corresponding weight factor, Qwxg represents the expected effect value;

[0030] The calculation formula of the user feedback index Yhfk is:

[0031]

[0032] In the above formula, the user feedback index Yhfk represents the user's satisfaction with the advertising, QWfk represents the expectation index, Represents the mean of the total value of user feedback ratings, Fksj i represents the feedback rating value of the i-th user, and y represents that y users participated in the feedback.

[0033] Preferably, the calculation formula of the recommended comprehensive index Zhzb is:

[0034] Zhzb=Tjxg*Yhfk

[0035] Compared with the prior art, the present invention provides a computer service platform for advertisement recommendation, which has the following beneficial effects:

[0036] 1. The present invention forms user characteristics by comprehensively analyzing the user's total purchase amount, purchase times, total browsing history, total browsing time, interests and hobbies, overall age preference and friends' sharing. The advertising recommendation system can more accurately grasp the user's needs and preferences, thereby recommending advertising content that is more in line with their interests to the user. This personalized recommendation method can significantly improve the click-through rate and conversion rate of advertisements. Combining weather changes, current popularity indexes, occupational market demand and local recommendations to form environmental characteristics for advertising recommendations can improve the accuracy and relevance of advertisements, enhance user experience and satisfaction, optimize advertising effects and promote commercial conversions. It can also provide merchants with market insights and decision-making support, and enhance the adaptability and flexibility of advertisements.

[0037] 2. The present invention calculates the recommendation comprehensive index by combining the advertising recommendation effect and the user feedback index, so that the system can more accurately grasp the needs and preferences of users, thereby recommending advertising content that is more in line with their interests. This precise matching can significantly improve the relevance and attractiveness of advertisements, thereby increasing users' click-through rate and willingness to buy. Merchants can reach target audiences more accurately, improve the return on investment of advertisements, and advertisers can better understand market trends and changes in consumer behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic diagram of the structural system of the present invention; DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] See also Figure 1 , a computer service platform for advertisement recommendation, comprising a data collection module, an analysis module, a platform recommendation module, an evaluation module and a comprehensive feedback module, wherein the data collection module is used to collect user characteristic data and environmental characteristic data, the data collection module is connected to the analysis module via a network, and sends the collected user characteristic data and environmental characteristic data to the analysis module;

[0041] The data collection module collects multiple user feature data to form a data set, and the calculation expression of the data set is: {YHtz1, YHtz2, YHtz3, ..., YHtz n}, in the data set, YHtz1 represents the first user feature data collected by the data collection module, YHtz n Represents the last user feature data collected by the data collection module. At the same time, n represents the total number of collected data;

[0042] The data collection module collects multiple environmental feature data to form a data set, and the calculation expression of the data set is: {HJtz1, HJtz2, HJtz3, ..., HJtz n}, in the data set, HJtz1 represents the first environmental feature data collected by the data collection module, HJtz n Represents the last environmental feature data collected by the data collection module. At the same time, n represents the total number of collected data;

[0043] The analysis module calculates the user characteristic index Ytzs, the environmental characteristic index Htzs and the comprehensive impact index Zhyx according to the data collected by the data collection module, and sends the calculated user characteristic index Ytzs, the environmental characteristic index Htzs and the comprehensive impact index Zhyx to the platform recommendation module;

[0044] The calculation formula of user characteristic index Ytzs is:

[0045]

[0046] In the above dataset, GMjl represents the total purchase amount, which reflects the user's spending power and willingness to buy. It helps to identify high-value customers and recommend higher-end or more expensive products to them. Gt represents the number of purchases, which indicates the user's loyalty and activity. Frequent purchasers may have a sustained interest in certain brands or product categories. LLjl represents the total browsing history, which shows the user's interest range and browsing habits, and helps to understand the user's potential needs and preferences. Lt represents the total browsing time, which reflects the user's attention to a certain product or page. Long browsing time may mean a higher interest. XQah represents interests and hobbies, which are directly related to the user's personality and preferences and are the key factors for recommendation. NLpa represents the overall preference for age. Users of different age groups may have different consumption habits and preferences. Understanding this helps to accurately locate the target group, and product advertisements suitable for that age group can be pushed according to the age group. HYfx represents friend sharing, which is a reflection of social influence. Friends' recommendations are often more convincing than advertisements. 1 , α 2 , α 3 , α 4 , α 5 Represent the corresponding weight factors respectively;

[0047] By comprehensively analyzing the user's total purchase amount, purchase frequency, total browsing history, total browsing time, hobbies, overall age preference, and friends' sharing, the ad recommendation system can more accurately grasp the user's needs and preferences, and recommend advertising content that better suits the user's interests. This personalized recommendation method can significantly increase the click-through rate and conversion rate of ads;

[0048] The calculation formula of environmental characteristic index Htzs is:

[0049]

[0050] In the above formula, Jjbh represents seasonal weather changes. Advertisements for related products or services are pushed based on weather changes, such as recommending umbrellas and snow boots on rainy days and sunscreen and outdoor equipment on sunny days, which can significantly improve users' shopping experience and satisfaction. Dxlx represents the current popularity index. By analyzing current popular trends, advertisers can timely launch products and services that meet market hotspots, attract user attention, and increase brand exposure. Zyxq represents occupational market demand. According to the demand characteristics of different occupational groups, advertisements that meet their occupational attributes and consumption habits are pushed to improve the accuracy and effectiveness of advertisements. Bdtj represents local recommendations. Through localized recommendations, users are provided with advertising information related to their area, such as local merchant discounts, community activities, etc., which can enhance users' dependence and stickiness on the platform. 1 , β 2 , β 3 , β4 Represent the corresponding weight factors respectively;

[0051] Combining weather changes, current popularity indexes, job market demand and local recommendations to form environmental characteristics for advertising recommendations can improve the accuracy and relevance of advertising, enhance user experience and satisfaction, optimize advertising effects and promote business conversion. It can also provide merchants with market insights and decision-making support, and enhance the adaptability and flexibility of advertising.

[0052] The calculation formula of the comprehensive impact index Zhyx is:

[0053]

[0054] In the above formula, Bz is the standard value of the comprehensive impact index;

[0055] The platform recommendation module calculates the advertising matching degree Ggpp with the platform advertisements based on the user characteristic index Ytzs, the environmental characteristic index Ytzs and the comprehensive impact index Zhyx, and extracts advertisements with higher matching degrees for advertisement recommendation;

[0056] The calculation formula of advertisement matching degree Ggpp is:

[0057]

[0058] In the above formula, GGzz is the advertising standard data existing in the platform;

[0059] Recommend the advertisements whose advertisement matching degree Ggpp is greater than the minimum advertisement matching degree, and send the data to the evaluation module;

[0060] The evaluation module calculates the advertising recommendation effect Tjxg and the user feedback index Yhfk, and calculates the recommendation comprehensive index Zhzb based on the advertising recommendation effect Tjxg and the user feedback index Yhfk, and sends the recommendation comprehensive index Zhzb to the comprehensive feedback module;

[0061] The calculation formula of advertising recommendation effect Tjxg is:

[0062]

[0063] In the above formula, Zscs represents the number of impressions, and Djcs represents the number of clicks. represents the click-through rate, γ 1 is the corresponding weight factor, Zhcs represents the number of conversions, is the conversion rate, γ 2 is the corresponding weight factor, Trcb represents the input cost, γ 3 represents the corresponding weight factor, Qwxg represents the expected effect value;

[0064] The calculation formula of user feedback index Yhfk is:

[0065]

[0066] In the above formula, the user feedback index Yhfk represents the user's satisfaction with the advertising, QWfk represents the expectation index, Represents the mean of the total value of user feedback ratings, Fksj i represents the feedback rating value of the i-th user, and y represents that there are y users who participated in the feedback;

[0067] The calculation formula of the recommended comprehensive index Zhzb is:

[0068] Zhzb=Tjxg*Yhfk

[0069] The comprehensive feedback module generates a report using the recommendation comprehensive index Zhzb for advertising recommendation feedback;

[0070] By combining the advertising recommendation effect and the user feedback index to calculate the comprehensive recommendation index, the system can more accurately grasp the user's needs and preferences, and recommend advertising content that is more in line with their interests. This precise matching can significantly improve the relevance and attractiveness of advertisements, thereby increasing users' click-through rates and willingness to buy. Merchants can reach their target audiences more accurately, improve the return on investment of advertisements, and advertisers can better understand market trends and changes in consumer behavior.

[0071] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A computer service platform for advertising recommendation, characterized in that: It includes a data collection module, an analysis module, a platform recommendation module, an evaluation module and a comprehensive feedback module. The data collection module is used to collect user feature data and environmental feature data. The data collection module is connected to the analysis module through a network and sends the collected user feature data and environmental feature data to the analysis module. The analysis module calculates the user characteristic index Ytzs, the environmental characteristic index Htzs and the comprehensive impact index Zhyx according to the data collected by the data collection module, and sends the calculated user characteristic index Ytzs, the environmental characteristic index Htzs and the comprehensive impact index Zhyx to the platform recommendation module; The platform recommendation module calculates the advertising matching degree Ggpp with the platform advertisements according to the user characteristic index Ytzs, the environmental characteristic index Htzs and the comprehensive impact index Zhyx, extracts the advertisements with higher matching degree for advertisement recommendation, and sends the data to the evaluation module. The evaluation module calculates the advertising recommendation effect Tjxg and the user feedback index Yhfk, and calculates the recommendation comprehensive index Zhzb based on the advertising recommendation effect Tjxg and the user feedback index Yhfk, and sends the recommendation comprehensive index Zhzb to the comprehensive feedback module; The comprehensive feedback module generates a report using the recommendation comprehensive index Zhzb for advertisement recommendation feedback.

2. The computer service platform for advertisement recommendation according to claim 1, characterized in that: The data acquisition module forms a data set with the collected multiple user feature data, and the calculation expression of the data set is: {YHtz1, YHtz2, YHtz3,..., YHtzn}, in which YHtz1 represents the first user feature data collected by the data acquisition module, YHtzn represents the last user feature data collected by the data acquisition module, and n represents that there are a total of n collected data.

3. The computer service platform for advertisement recommendation according to claim 1, characterized in that: The data acquisition module forms a data set with the collected multiple environmental feature data, and the calculation expression of the data set is: {HJtz1, HJtz2, HJtz3,..., HJtzn}, in which HJtz1 represents the first environmental feature data collected by the data acquisition module, HJtzn represents the last environmental feature data collected by the data acquisition module, and n represents that there are a total of n collected data.

4. The computer service platform for advertisement recommendation according to claim 2, characterized in that: The calculation formula of the user characteristic index Ytzs is: In the above data set, GMJ represents the total purchase amount, Gt represents the number of purchases, LLjl represents the total browsing history, Lt represents the total browsing time, XQah represents hobbies, NLpa represents the overall preference for age, HYfx represents friends sharing, α 1 , α 2 , α 3 , α 4 , α 5 Represent the corresponding weight factors respectively.

5. The computer service platform for advertisement recommendation according to claim 3, characterized in that: The calculation formula of the environmental characteristic index Htzs is: In the above formula, Jjbh represents seasonal weather changes, Dxlx represents the current popularity index, Zyxq represents the occupational market demand, Bdtj represents local recommendations, and β 1 , β 2 , β 3 , β 4 Represent the corresponding weight factors respectively.

6. The computer service platform for advertisement recommendation according to claim 5, characterized in that: The calculation formula of the comprehensive impact index Zhyx is: In the above formula, Bz is the standard value of the comprehensive impact index.

7. The computer service platform for advertisement recommendation according to claim 6, characterized in that: The calculation formula of the advertisement matching degree Ggpp is: In the above formula, GGzz is the advertising standard data existing in the platform.

8. The computer service platform for advertisement recommendation according to claim 7, characterized in that: The advertisements whose advertisement matching degree Ggpp is greater than the minimum advertisement matching degree value are recommended.

9. The computer service platform for advertisement recommendation according to claim 1, characterized in that: The calculation formula of the advertising recommendation effect Tjxg is: In the above formula, Zscs represents the number of impressions, and Djcs represents the number of clicks. represents the click-through rate, γ 1 is the corresponding weight factor, Zhcs represents the number of conversions, is the conversion rate, γ 2 is the corresponding weight factor, Trcb represents the input cost, γ 3 represents the corresponding weight factor, Qwxg represents the expected effect value, The calculation formula of the user feedback index Yhfk is: In the above formula, the user feedback index Yhfk represents the user's satisfaction with the advertising, QWfk represents the expectation index, represents the mean of the total value of user feedback ratings, Fksji represents the feedback rating value of the i-th user, and y represents that y users participated in the feedback.

10. The computer service platform for advertisement recommendation according to claim 9, characterized in that: The calculation formula of the recommended comprehensive index Zhzb is: Zhzb=Tjxg*Yhfk.

Images