Telecommunication service propaganda pushing method based on artificial intelligence

Abstract

The invention relates to the technical field of business propaganda and pushing, in particular to a telecommunication business propaganda and pushing method based on artificial intelligence. Comprising the following steps: S1, collecting behavior data, preference data and telecommunication service data of a user, and sending the collected data to a database for storage; s2, obtaining a push value of a telecommunication service product through multi-dimensional analysis of the user information and the telecommunication service product information; s3, performing comprehensive analysis on the user active time, the login frequency and the response behavior to realize accurate pushing of related telecommunication services; s4, obtaining real-time feedback of the user to the pushed content by providing a user feedback entrance; through analysis of the time period and the activity degree, pushing is carried out at the active time of the user, the opening rate and the click rate of pushing are improved, the pushing frequency and the pushing interval are adjusted, and excessive interference to the user is avoided; the invalid pushing frequency is reduced, pushing in a low-response-rate channel and an inactive time period is avoided, and resource use is optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of service promotion and push technology, and in particular to a telecommunication service promotion and push method based on artificial intelligence. Background Art

[0002] Telecom business promotion and push has important strategic significance in the modern communications industry. It is not only an engine for business growth, but also an important means to optimize user experience and enhance market competitiveness. The importance of telecom business promotion and push is not only reflected in the direct growth of business revenue, but also in its long-term impact on user relationship maintenance, brand value enhancement and market competitiveness enhancement.

[0003] At present, telecommunication business promotion is mainly delivered through telephone and text message push. Although this method has the advantage of wide coverage, it usually adopts fixed telecommunication business push, and has problems such as single content and high repetition rate. At the same time, due to the lack of real-time analysis and dynamic adjustment of user behavior, the pushed content often does not match user needs, resulting in a decline in promotional effects and even invalid push information. Summary of the invention

[0004] To this end, the present invention provides an artificial intelligence-based telecommunication service promotion and push method to overcome the problems mentioned in the above background technology.

[0005] To achieve the above object, the present invention provides a telecommunication service promotion and push method based on artificial intelligence, comprising the following steps:

[0006] S1: Collect user behavior data, preference data and telecommunications service data, and send the collected data to the database for storage;

[0007] S2: Obtain the push value of the telecommunication service product through multi-dimensional analysis of user information and telecommunication service product information, and recommend telecommunication services to users based on the push value; the specific analysis steps are as follows:

[0008] Construct a cone with the values ​​of browsing evaluation value Tβ, price evaluation value Qij and function-related value Gij. Use the browsing evaluation value as the diameter of the cone, the function-related value as the height of the cone, and the price evaluation value as the tangent. Cut a small cone vertically downward from the vertex of the cone. The height of the cut cone is the value of the price evaluation value. Substitute the values ​​of browsing evaluation value Tβ, price evaluation value Qij and function evaluation value Gij into the formula The volume of the cut cone is calculated, which is the comprehensive matching value Pij between the user and the recommended telecommunication service; That is, the tangent length Qij does not exceed the height Gij of the cone; the values ​​of the comprehensive matching value Pij and the feedback coefficient λ are substituted into the formula Vij=Pij×(1+λ) to calculate the push value of the telecommunication service product;

[0009] A product recommendation threshold is set, and the product push value of each telecommunication service is compared and analyzed with the set product recommendation threshold; when the product push value is greater than or equal to the set product recommendation threshold, the telecommunication service is marked as a related telecommunication service, and the recommendation frequency is increased or it is displayed preferentially in the recommendation list; when the product push value is less than the set product recommendation threshold, the recommendation frequency of the telecommunication service is reduced, and for irrelevant telecommunication service products, the recommendation frequency is reduced or they are temporarily removed from the recommendation list;

[0010] S3: Comprehensive analysis of user active time, login frequency and response behavior to achieve accurate push of relevant telecommunications services;

[0011] S4: By providing a user feedback portal, we can obtain real-time feedback from users on pushed content and ensure that the pushed content matches the user's interests and needs.

[0012] As a preferred implementation of the present invention, user behavior data collection includes: user basic information, browsing service time, user's current telecommunications service price and user's current telecommunications service function; user preference data collection includes: user login frequency, interaction records and active time; telecommunications service data includes the number of times the user likes or dislikes.

[0013] As a preferred embodiment of the present invention, the specific analysis steps of browsing evaluation values ​​are as follows:

[0014] Get the time when the user last browsed the telecommunications service and mark it as T 浏览 ; Mark the current browsing time as Tij, where i = 1, 2, 3 ... I, I is a positive integer, I represents the total number of users, i represents the number of any one of them; where j = 1, 2, 3 ... J, J is a positive integer, J represents the total number of telecommunications service product types, j represents the serial number of any one of them; the current browsing time Tij and the time when the user last browsed the telecommunications service T 浏览 Substitute the value of into the formula The browsing evaluation value Tβ is calculated, where e is a natural constant and β1 is a set weight factor;

[0015] A browsing evaluation value threshold is set, and the user's browsing evaluation value is compared and analyzed with the set browsing evaluation threshold. When the browsing evaluation value is greater than or equal to the set browsing evaluation threshold, the user is marked as an active user and precise push is performed; when the browsing evaluation value is less than the set browsing evaluation value, the user is marked as a normal user and no additional telecommunications services are pushed.

[0016] As a preferred embodiment of the present invention, the specific analysis steps of the price evaluation value are:

[0017] Get the price of the telecommunications service product currently used by the active user and mark it as Q 当前 ; The price of the telecommunications service product recommended to active users is marked as Q 推荐 ; Set the current price of telecom service products for active users to Q 当前 and recommended prices for telecom business products 推荐 Substitute the value of into the formula The user's price evaluation value Qij is calculated, where β2 is the set weight factor.

[0018] As a preferred embodiment of the present invention, the specific analysis steps of the function-related value are:

[0019] Get the total number of active users' current telecom service functions as G 总 ; Set the number of parts of the telecommunication service functions recommended to active users that are identical or similar to the functions provided by the telecommunication service to G 相 Similar; the total number of telecommunication service functions required by the user G 总 and the number of similar functions G 相 Substituting similar values ​​into the formula The function-related value Gij of the telecommunication service is calculated.

[0020] As a preferred embodiment of the present invention, the specific analysis steps of the feedback coefficient are:

[0021] Get the user's feedback on the recommended product, where the number of likes for the recommended product is recorded as D 赞 ; The number of clicks on the recommended product is recorded as D 踩 ; Record the time point of the user's like and record it as Cn, where n = 1, 2, 3...N, N is a positive integer, N represents the total number of time points of all the user's likes, and n represents the time point of any one of the likes; Record the time point of the user's dislike and record it as Cm, where m = 1, 2, 3...M, M is a positive integer, M represents the total number of time points of the user's dislike behavior, and m represents the time point of any one of the dislikes; Get the current time point and record it as C1; Record the number of likes D of the user for the recommended product 赞, click count D 踩 , the values ​​of the like time point Cn, the dislike time point Cm and the current time point C1 are substituted into the set formula group The feedback value λn is calculated, where α1 and α2 are the set weight factors, α1>0 and α2<0, α3 is the set time decay coefficient and α3>0; the like feedback value λ 赞 and the tap feedback value λ 踩 Substitute the value of into the formula Calculate the total feedback value λ 总 ; The total feedback value λ 总 Substitute into the feedback coefficient formula The feedback coefficient λ is calculated.

[0022] As a preferred implementation of the present invention, the specific steps of analyzing the user's active time are as follows:

[0023] Divide a day into multiple time periods, namely early morning, morning, afternoon and evening; obtain the cumulative number of logins of users in each time period and mark them as K 凌晨 , K 上午 , K 下午 and K 晚上 , and sum up the browsing time corresponding to the cumulative number of logins in each time period to obtain the cumulative time F of each time period 凌晨 、F 上午 、F 下午 and F 晚上 ; The cumulative number of logins K in each time period 凌晨 , K 上午 , K 下午 , K 晚上 and the cumulative duration of each time period F 凌晨 、F 上午 、F 下午 、F 晚上 Substitute into the formula group Calculate the user's activity H in each time period, where a1 and a2 are the set weight factors; compare the user's activity H in all time periods 凌晨 , H 上午 , H 下午 , H 晚上 , sort them from high to low according to the activity, select the time period with the highest activity as the user's recommended time period, and mark it as H 推荐 After obtaining the user's recommended time period, push telecommunication services based on the time point when the user logs in most frequently during the recommended time period.

[0024] As a preferred implementation of the present invention, the specific steps of analyzing the user response behavior are:

[0025] Get the user's push response data and record the total number of pushes as R 总 ; Get the number of times the user opens the push notification after receiving it, which is R 打开 ; Get the number of times the user clicks a link or button after opening the push notification, and record it as R 点击 ; Get the number of times a user opens a push notification, clicks to purchase a product, registers an account, or fills out a form, and record it as R 转化 ; The total number of pushes to the user R 总 , Number of times opened after push R 打开 , the number of times the push notification is opened and clicked R 点击 And the number of times the push notification was opened and clicked to purchase R 转化 Substitute the value of into the formula Calculate the user's push interval value R 间隔 , where a3, a4 and a5 are the set weight factors respectively.

[0026] As a preferred implementation of the present invention, the specific analysis steps of the push strategy are:

[0027] Count the response rates of users on different channels, setting the current channels to SMS, APP notifications, emails, and social media; get the total number of users who received the notification and record it as P 总 ; Get the number of users who click on the push content to interact (click, view, reply), and record it as P 响应 , using the formula Calculate the user's response rate P in each channel 响应率 ; Get the number of users who clicked on the push content link or button, and record it as P 点击 , using the formula Calculate the user's click rate P in each channel 点击率 ; Get the number of users who purchased goods, registered accounts, or filled out forms after clicking the push content link or button, and record it as P 转化 , using the formula Calculate the conversion rate P of users in each channel 转化率 ; The user's response rate P in each channel 响应率 , the click rate of users in each channel P 点击率 and the conversion rate of users in each channel 转化率 Substitute the values ​​into the formula Calculate the priority value P of each channel 短信 , P App , P 邮件 and P 社交媒体 , where a6, a7 and a8 are the set weight factors respectively; the priority values ​​of each channel P 短信 , P App, P 邮件 and P 社交媒体 Perform comparative analysis and select the channel with the largest priority value as the preferred channel. When the user does not respond on the preferred channel, select the channel with the second largest priority value as the suboptimal channel to continue pushing.

[0028] As a preferred embodiment of the present invention, the specific analysis steps for the real-time feedback of the pushed content are:

[0029] Set a feedback button under each push content, allowing users to like, dislike or choose "no longer receive this content";

[0030] When the user gives feedback and clicks "like" or "dislike", the user's likes or dislikes and the corresponding telecommunications services will be updated to the database in real time; when the user chooses "no longer receive this content", the telecommunications service will be immediately marked as "no longer recommended" and removed from the user's subsequent push list.

[0031] Beneficial effects of the present invention:

[0032] 1. By analyzing the user's browsing behavior, price matching degree and function relevance, it can accurately match user needs, effectively prevent users from receiving irrelevant or low-matching push notifications, and ensure that the recommended telecommunications service products are highly consistent with user needs; through dynamic evaluation of time decay and price ratio, it ensures that the recommendation results are optimized in real time as time changes or user behavior updates; by effectively screening highly compatible products, it improves users' trust and satisfaction with recommended content, and promotes users' choice and conversion of telecommunications services.

[0033] 2. Through time period analysis and activity calculation, push notifications are concentrated during user active time, which greatly improves the open rate and click rate of push notifications, and dynamically adjusts the push frequency and interval to avoid excessive interference to users; reduces the number of invalid push notifications, avoids push notifications in low-response channels and inactive time periods, and optimizes resource usage; links multiple channels such as SMS, APP notifications, emails, and social media to form full coverage and improve user conversion rates; overall, it significantly improves the publicity effect of telecommunications services, reduces costs, optimizes user experience, brings more traffic and higher conversion benefits to the business, and at the same time enhances users' recognition and trust in push content.

[0034] 3. Improve user participation and interaction frequency in pushed content through users' active likes or thumbs-up actions; allowing users to directly participate in the selection of pushed content can stimulate their enthusiasm and improve the overall push response rate; users can control the content they are not interested in, reduce interference, and improve their satisfaction with the push system; by no longer pushing unpopular content and providing customized options, reduce user loss due to too many irrelevant pushes. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0036] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0037] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0038] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0039] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0040] See also Figure 1 As shown, the present invention is a telecommunication service promotion and push method based on artificial intelligence, comprising the following steps:

[0041] S1: Collect user behavior data, preference data and telecommunication service data, and send the collected data to the database for storage; user behavior data collection includes: user basic information, browsing service time, user's current telecommunication service price and user's current telecommunication service function; user preference data collection includes: user login frequency, interaction record and active time; telecommunication service data includes the number of times the user likes or dislikes;

[0042] S2: Obtain the push value of the telecommunication service product through multi-dimensional analysis of user information and telecommunication service product information, and recommend telecommunication services to users based on the push value; the specific analysis steps are as follows:

[0043] Get the time when the user last browsed the telecommunications service and mark it as T 浏 Browse; mark the current browsing time as Tij, where i = 1, 2, 3 ... I, I is a positive integer, I represents the total number of users, i represents the number of any one of them; where j = 1, 2, 3 ... J, J is a positive integer, J represents the total number of telecommunications service product types, j represents the serial number of any one of them; the current browsing time Tij and the time when the user last browsed the telecommunications service T 浏览 Substitute the value of into the formula The browsing evaluation value Tβ is calculated, where e is a natural constant and β1 is a set weight factor. It can be seen from the formula that the larger the interval between the user's last browsing time and the current browsing time, the smaller the browsing evaluation value, that is, (Tij-T 浏览 ) is larger, the smaller the browsing evaluation value is; when the time interval between the user's last browsing time and the current browsing time is smaller, that is, the current browsing time is closer to the last browsing time, the larger the browsing evaluation value is; set a browsing evaluation value threshold, compare and analyze the user's browsing evaluation value with the set browsing evaluation threshold, when the browsing evaluation value is greater than or equal to the set browsing evaluation threshold, it means that the user is relatively active, and the user is marked as an active user, and accurate push is performed; when the browsing evaluation value is less than the set browsing evaluation value, the user is marked as a normal user, and no additional telecommunications services are pushed;

[0044] Get the price of the telecommunications service product currently used by the active user and mark it as Q 当前 ; Mark the price of the telecommunications service product recommended to the user as Q 推荐 ; Set the current price of telecom service products for active users to Q 当前 and recommended prices for telecom business products 推荐 Substitute the value of into the formula The user's price evaluation value Qij is calculated, where β2 is the set weight factor. As can be seen from the formula, when the price ratio of the current telecommunication service product of the active user is closer to the price of the recommended telecommunication service product, the greater the price evaluation value, the product will be recommended first.

[0045] Get the total number of active users' current telecom service functions as G 总 ; Set the number of parts of the telecommunication service functions recommended to active users that are identical or similar to the functions provided by the telecommunication service to G 相似 ; The total number of telecommunication service functions required by the user G 总 and the number of similar functions G 相似 Substitute the value of into the formula The function relevance value Gij of the telecommunication service is calculated; it can be seen from the formula that when the number of similar functions of the telecommunication service is larger than the total number of functions of the user's current telecommunication service, the function relevance value of the telecommunication service is larger; when the difference between the number of similar functions of the telecommunication service and the total number of functions of the user's current telecommunication service is larger, the function relevance value of the telecommunication service is smaller;

[0046] Construct a cone with the values ​​of browsing evaluation value Tβ, price evaluation value Qij and function-related value Gij. Use the browsing evaluation value as the diameter of the cone, the function-related value as the height of the cone, and the price evaluation value as the tangent. Cut a small cone vertically downward from the vertex of the cone. The height of the cut cone is the price evaluation value. Substitute the values ​​of browsing evaluation value Tβ, price evaluation value Qij and function evaluation value Gij into the set formula The volume of the cut cone is calculated, which is the comprehensive matching value Pij between active users and recommended telecommunication services; That is, the tangent length Qij does not exceed the height Gij of the cone; the values ​​of the comprehensive matching value Pij and the feedback coefficient λ are substituted into the set formula Vij=Pij×(1+λ) to calculate the push value of the telecommunication service product; it can be seen from the formula that the greater the comprehensive matching value, the greater the push value of the telecommunication service; the greater the feedback coefficient, the greater the push value of the telecommunication service; the specific analysis steps of the feedback coefficient λ are:

[0047] Get the user's feedback on the recommended product, where the number of likes for the recommended product is recorded as D 赞 ; The number of clicks on the recommended product is recorded as D 踩 ; Record the time point of the user's like and record it as Cn, where n = 1, 2, 3...N, N is a positive integer, N represents the total number of time points of all the user's likes, and n represents the time point of any one of the likes; Record the time point of the user's dislike and record it as Cm, where m = 1, 2, 3...M, M is a positive integer, M represents the total number of time points of the user's dislike behavior, and m represents the time point of any one of the dislikes; Get the current time point and record it as C1; Record the number of likes D of the user for the recommended product 赞 , click count D 踩 , the values ​​of the like time point Cn, the dislike time point Cm and the current time point C1 are substituted into the set formula group The feedback value λn is calculated, where α1 and α2 are the set weight factors, α1>0 and α2<0, α3 is the set time decay coefficient and α3>0; the like feedback value λ 赞 and the tap feedback value λ 踩 Substitute the value of into the formula Calculate the total feedback value λ总 ; The total feedback value λ 总 Substitute the set feedback coefficient formula The feedback coefficient λ is calculated;

[0048] A product recommendation threshold is set, and the product push value of each telecommunication service is compared and analyzed with the set product recommendation threshold; when the product push value is greater than or equal to the set product recommendation threshold, it indicates that the telecommunication service has a high degree of fit with the user, and the telecommunication service is marked as a related telecommunication service. For products marked as related telecommunication services, the recommendation frequency is increased or they are given priority in the recommendation list; when the product push value is less than the set product recommendation threshold, it indicates that the telecommunication service has a low degree of fit with the user, and the recommendation frequency of the telecommunication service is reduced. For irrelevant telecommunication service products, the recommendation frequency is reduced or they are temporarily removed from the recommendation list;

[0049] By analyzing the user's browsing behavior, price matching degree and function relevance, it can accurately match user needs, effectively prevent users from receiving irrelevant or low-matching push notifications, and ensure that the recommended telecommunications service products are highly compatible with user needs; through dynamic evaluation of time decay and price ratio, it ensures that the recommendation results are optimized in real time as time changes or user behavior updates; by effectively screening highly compatible products, it improves users' trust and satisfaction with recommended content, and promotes users' choice and conversion of telecommunications services;

[0050] S3: Comprehensively analyze user active time, login frequency and response behavior to implement different push strategies for related telecommunications services; the specific steps are as follows:

[0051] Divide a day into multiple time periods, namely early morning, morning, afternoon and evening; the early morning period is 0:00-6:00; the morning period is 6:00-12:00; the afternoon period is 12:00-18:00; the evening period is 18:00-24:00; obtain the cumulative number of logins of the user in each time period and mark them as K 凌晨 , K 上午 , K 下午 and K 晚上 , and sum up the browsing time corresponding to the cumulative number of logins in each time period to obtain the cumulative time F of each time period 凌晨 、F 上午 、F 下午 and F 晚上 ; The cumulative number of logins K in each time period 凌晨 , K 上午 , K 下午 , K 晚上 and the cumulative duration of each time period F 凌晨 、F 上午 、F下午 、F 晚上 Substitute the set formula group Calculate the user's activity H in each time period, where a1 and a2 are the set weight factors; compare the user's activity H in all time periods 凌晨 , H 上午 , H 下午 , H 晚上 , sort them from high to low according to the activity, select the time period with the highest activity as the user's recommended time period, and mark it as H 推荐 ; After obtaining the user's recommended time period, push telecommunication services based on the time point when the user logs in most frequently during the recommended time period;

[0052] Get the user's push response data and record the total number of pushes as R 总 ; Get the number of times the user opens the push notification after receiving it, which is R 打开 ; Get the number of times the user clicks a link or button after opening the push notification, and record it as R 点击 ; Get the number of times a user opens a push notification, clicks to purchase a product, registers an account, or fills out a form, and record it as R 转化 ; The total number of pushes to the user R 总 , Number of times opened after push R 打开 , the number of times the push notification is opened and clicked R 点击 And the number of times the push notification was opened and clicked to purchase R 转化 Substitute the value of into the formula Calculate the user's push interval value R 间隔 , where a3, a4 and a5 are the set weight factors respectively; from the formula, we can know that when the interval value is larger, the push frequency can be increased, and when the interval value is smaller, the push frequency can be reduced; the push frequency is once a day at most during the active time period, and the minimum push frequency is once a week;

[0053] Count the response rates of users on different channels, setting the current channels to SMS, APP notifications, emails, and social media; get the total number of users who received the notification and record it as P 总 ; Get the number of users who click on the push content to interact (click, view, reply), and record it as P 响应 , using the formula Calculate the user's response rate P in each channel 响应率 ; Get the number of users who clicked on the push content link or button, and record it as P 点击 , using the formula Calculate the user's click rate P in each channel 点击率 ; Get the number of users who purchased goods, registered accounts, or filled out forms after clicking the push content link or button, and record it as P转化 , using the formula Calculate the conversion rate P of users in each channel 转化率 ; The user's response rate P in each channel 响应率 , the click rate of users in each channel P 点击率 and the conversion rate of users in each channel 转化率 Substitute the value of into the set formula group Calculate the priority value P of each channel 短信 , P App , P 邮件 and P 社交媒体 , where a6, a7 and a8 are the set weight factors respectively; the priority values ​​of each channel P 短信 , P App , P 邮件 and P 社交媒体 Perform comparative analysis and select the channel with the largest priority value as the preferred channel. If the user does not respond on the preferred channel, select the channel with the second largest priority value as the suboptimal channel to continue pushing.

[0054] By analyzing time periods and calculating activity, push notifications are concentrated during user active hours, significantly increasing the open rate and click rate of push notifications, and dynamically adjusting push notification frequency and intervals to avoid excessive interference to users; reducing the number of invalid push notifications, avoiding push notifications in low-response channels and inactive time periods, and optimizing resource usage; linking multiple channels such as SMS, APP notifications, emails, and social media to form all-round coverage and improve user conversion rates; overall, the promotion effect of telecommunications services has been significantly improved, costs have been reduced, user experience has been optimized, more traffic and higher conversion revenue have been brought to the business, and at the same time, users' recognition and trust in push notification content have been enhanced;

[0055] S4: By providing a user feedback portal, obtain real-time feedback from users on the pushed content to ensure that the pushed content matches the user's interests and needs; the specific steps are as follows:

[0056] Set a feedback button under each push content, allowing users to like, dislike or choose "no longer receive this content";

[0057] When the user gives feedback and clicks "like" or "dislike", the user's behavior (like, dislike) and the corresponding telecommunication service ID and push content ID are updated to the database in real time; when the user chooses "no longer receive this content", the telecommunication service is immediately marked as "no longer recommended" and removed from the user's subsequent push list; after the user behavior data is updated to S1, the above steps are repeated to promote the telecommunication service to the user;

[0058] By encouraging users to actively click “likes” or “thumbs-up”, the user’s participation and interaction frequency in the pushed content can be increased; allowing users to directly participate in the selection of pushed content can stimulate their enthusiasm and improve the overall push response rate; users can control the content they are not interested in, reduce interference, and improve their satisfaction with the push system; by no longer pushing unpopular content and providing customized options, the loss of users due to too many irrelevant pushes can be reduced.

[0059] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0060] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A telecommunication service promotion and push method based on artificial intelligence, characterized in that: The following steps are involved: S1: Collect user behavior data, preference data, and telecommunications service data, and send them to the database for storage; S2: Analyze the user information and telecommunication service product information in multiple dimensions to obtain the push value of the telecommunication service product, and recommend telecommunication services to users based on the push value. The specific analysis steps are as follows: A cone is constructed with the values ​​of the browsing evaluation value, the price evaluation value and the function evaluation value, with the browsing evaluation value as the diameter of the cone, the function-related value as the height of the cone, and the price evaluation value as the tangent, and a small cone is cut vertically downward from the vertex of the cone, and the height of the cut cone is the value of the price evaluation value; the browsing evaluation value, the price evaluation value and the function evaluation value are normalized to obtain the volume of the cut cone, and the volume value of the cone volume is marked as the comprehensive matching value of the active user and the recommended telecommunication service; the comprehensive matching value and the feedback coefficient are normalized to obtain the push value of the telecommunication service product; Compare and analyze the product push value of each telecommunication service with the set product recommendation threshold: when the product push value is greater than or equal to the set product recommendation threshold, mark the telecommunication service as a related telecommunication service, increase the recommendation frequency or give priority to display in the recommendation list; when the product push value is less than the set product recommendation threshold, reduce the recommendation frequency of the telecommunication service, and for irrelevant telecommunication service products, reduce the recommendation frequency or temporarily remove them from the recommendation list; S3: Comprehensive analysis of user active time, login frequency and response behavior to achieve accurate push of relevant telecommunications services; S4: Provide a user feedback portal to obtain real-time feedback from users on pushed content and ensure that the pushed content matches user interests and needs.

2. The method for promoting telecommunication services based on artificial intelligence according to claim 1, characterized in that: User behavior data collection includes: user basic information, browsing time, user's current telecommunications service price and user's current telecommunications service function; user preference data collection includes: user login frequency, interaction records and active time; telecommunications service data includes the number of times the user likes or dislikes a post.

3. The method for promoting telecommunication services based on artificial intelligence according to claim 1, characterized in that: The detailed analysis of the browsing evaluation value is as follows: Obtain the time when the user last browsed the telecommunication service and mark it; mark the current browsing time, and normalize the values ​​of the current browsing time and the time when the user last browsed the telecommunication service to obtain a browsing evaluation value; A browsing evaluation value threshold is set, and the user's browsing evaluation value is compared and analyzed with the set browsing evaluation threshold. When the browsing evaluation value is greater than or equal to the set browsing evaluation threshold, the user is marked as an active user and precise push is performed; when the browsing evaluation value is less than the set browsing evaluation value, the user is marked as a normal user and no additional telecommunications services are pushed.

4. The method for promoting telecommunication services based on artificial intelligence according to claim 1, characterized in that: The specific analysis of the price assessment value is as follows: The prices of telecommunication service products currently used by active users are obtained and marked; the prices of telecommunication service products recommended to active users are marked; and the values ​​of the current prices of telecommunication service products of active users and the prices of recommended telecommunication service products are normalized to obtain the price evaluation value of the user.

5. The method for promoting telecommunication services based on artificial intelligence according to claim 1, characterized in that: The specific analysis of the function-related values ​​is: Obtain the total number of telecommunication service functions currently available to active users and mark them; obtain the number of telecommunication service functions recommended to active users that are identical or similar to the functional characteristics of the functions provided by the telecommunication services, and mark the number of those that are identical or similar in functional characteristics; The total number of telecommunication service functions and the number of similar functions required by active users are normalized to obtain the function-related value of the telecommunication service.

6. The method for promoting telecommunication services based on artificial intelligence according to claim 1, characterized in that: The specific analysis of the feedback coefficient is: Obtain user feedback on recommended products, mark the number of likes for recommended products; mark the number of dislikes for recommended products; record the time when users like and dislike, and mark the current time; normalize the values ​​of the number of likes, dislikes, time of likes, time of dislikes and current time of recommended products to obtain feedback values; normalize the values ​​of like feedback values ​​and dislike feedback values ​​to obtain a total feedback value; The total feedback value is normalized to obtain the feedback coefficient.

7. The method for promoting telecommunication services based on artificial intelligence according to claim 1, characterized in that: The specific analysis of user active time is as follows: Divide a day into multiple time periods, namely early morning, morning, afternoon and evening; obtain the cumulative number of logins of the user in each time period and mark them respectively; The browsing time corresponding to the cumulative number of logins in each time period is summed up to obtain the cumulative time of each time period, and the browsing time is marked respectively; Normalize the cumulative number of logins and the cumulative duration of each time period to obtain the user's activity in each time period; Compare the user's activity levels in all time periods, sort them from high to low, and select the time period with the highest activity level as the user's recommended time period; After obtaining the user's recommended time period, telecommunication services are pushed according to the time point when the user logs in most frequently in the recommended time period.

8. The method for promoting telecommunication services based on artificial intelligence according to claim 1, characterized in that: The specific analysis of user response behavior is as follows: Get the user's push response data and mark the total number of pushes; get the number of times the user opens the push after receiving it; get the number of times the user clicks on a link or button after opening the push, and mark it; Get the number of times a user opens a push notification, clicks to purchase a product, registers an account, or fills out a form, and mark it; The user's total number of push notifications, number of push opens, number of push clicks, and number of push click purchases are normalized to obtain the user's push interval value.

9. The method for promoting telecommunication services based on artificial intelligence according to claim 1, characterized in that: The specific analysis of the push strategy is as follows: Get the total number of users who received the notification and mark them; get the number of users who clicked on the push content to interact and mark them, normalize the total number of users who received the notification and the number of users who clicked on the push content to interact to get the user response rate in each channel; Obtain the number of users who clicked on the push content link or button, mark them, and normalize the number of users who clicked on the push content link or button and the total number of users who received the notification to obtain the click rate of users in each channel; The number of users who purchased goods, registered accounts or filled out forms after clicking on the push content link or button is obtained and marked, and the number of interactive users and the number of users who clicked on the push content link or button are normalized to obtain the conversion rate of users in each channel; the response rate of users in each channel, the click rate of users in each channel and the conversion rate of users in each channel are normalized to obtain the priority value of each channel; the priority values ​​of each channel are compared and analyzed, and the channel with the largest priority value is selected as the preferred channel. When the user does not respond on the preferred channel, the channel with the second largest priority value is selected as the suboptimal channel for continued push.

10. The method for promoting telecommunication services based on artificial intelligence according to claim 1, characterized in that: The specific analysis steps for real-time feedback of pushed content are as follows: Set a feedback button under each push content, allowing users to like, dislike or choose not to receive this content; When the user gives feedback and clicks like or dislike, the user's likes or dislikes and the corresponding telecommunications services will be updated to the database in real time; when the user chooses not to receive this content anymore, the telecommunications service will be immediately marked as no longer recommended and removed from the user's subsequent push list.

Images