Intelligent advertising delivery optimization method and system

Abstract

The present invention discloses an intelligent advertising delivery optimization method and system, which relates to the field of advertising delivery technology. The present invention dynamically adjusts the advertising delivery strategy through comprehensive analysis of user behavior data, device type and network status, significantly improves the accuracy and personalization of advertisements, and combines historical interest and potential interest tags to ensure that the advertising content is highly matched with user needs, which can not only meet current interests but also guide potential needs. Through pre-selected models and optimization processing, advertising delivery achieves diversified and personalized display, greatly improving the user's click-through rate and interactivity; in addition, the resolution, bit rate and layout of the advertisement are automatically adjusted according to the device type and network status to ensure that the advertisement can be quickly loaded and displayed in the best state, reduce freezes and loading delays, and improve the user's viewing experience; through this series of intelligent optimizations, the advertisement can instantly adapt to the user's current usage environment to achieve smooth and efficient advertising display.

Description

Technical Field

[0001] The present invention relates to the technical field of advertisement delivery, and in particular to an intelligent advertisement delivery optimization method and system. Background Art

[0002] With the rapid development of the digital advertising market, intelligent advertising delivery systems have gradually become an important tool for advertisers to accurately reach target users. Most existing intelligent advertising delivery systems rely on pre-set rules or offline trained models to distribute advertisements. These systems usually make advertising delivery decisions based on static information such as users' historical behaviors and interest preferences, and rarely consider dynamic changes such as the user's device type and network conditions when the advertisement is displayed. Therefore, although these systems have certain advantages in advertising relevance, in actual use, the user's device performance and network environment often change, resulting in the inability to effectively optimize the display effect of the advertisement, which ultimately affects the user's advertising viewing experience and the advertiser's delivery effect.

[0003] In addition, traditional advertising delivery systems lack the ability to dynamically respond to real-time data and are unable to instantly adjust advertising content and display methods based on the user's current device and network conditions. For example, when the network speed is slow or the device performance is low, the system may still deliver high-resolution or complex layout ads, causing the ads to load slowly or even fail to display normally. This static advertising delivery model limits the flexibility of the advertising system and the improvement of user experience, and cannot meet the needs of the modern advertising market for efficient and personalized delivery. Therefore, there is an urgent need for an intelligent advertising delivery optimization system that can dynamically adjust according to real-time changes in user devices and environments. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides an intelligent advertising delivery optimization method and system, which solves the problems in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a smart advertising delivery optimization method, comprising:

[0006] Step 1: Based on user authorization, collect the user's historical behavior information, generate corresponding historical behavior interest tags according to the historical behavior information, and predetermine potential interest tags different from the historical behavior interest tags according to the user's age, gender characteristics, and geographic location, and then determine the user's advertising set through the historical behavior interest tags or the potential interest tags;

[0007] Step 2: Based on the user's authorization, collect the user's currently used device information, and determine the specific device type used by the user and the current network speed of the device according to the device information, wherein the device information includes the device type and the network speed;

[0008] Step 3: Determine an ad to be delivered in the delivery ad set through the pre-selection model and mark it as a pre-delivery ad;

[0009] Step 4: Optimize the pre-delivered advertisements according to the device type and network speed, and deliver the optimized advertisements.

[0010] As a further solution of the present invention: in step 1, the method of predetermining potential interest tags different from historical behavior interest tags based on the user's age, gender characteristics, and geographic location is:

[0011] First, by collecting the user's age information, find the corresponding category from the preset age interest tag classification according to the user's age information, and find the tag different from the historical behavior interest tag according to the interest tag of this category as the user's potential interest tag;

[0012] Then, according to the user's gender characteristics, a corresponding category is found from the preset gender interest tag classification, and according to the interest tags of this category, a tag different from the historical behavior interest tag is found as the user's potential interest tag;

[0013] Finally, based on the user's geographic location and local lifestyle, cultural background and consumption habits, tags different from historical behavior interest tags are found from the preset geographic location interest tag classification as potential interest tags for the user.

[0014] As a further solution of the present invention: the specific method of determining the advertisement set of the user through the historical behavior interest tag or the potential interest tag is:

[0015] The advertisements to be delivered are summarized, one or more advertisement tag words are defined for each advertisement according to its content, and the advertisement tag words are compared one by one with the historical behavior interest tags and the potential interest tags. If an interest tag identical to the advertisement tag word is found in the historical behavior interest tags or the potential interest tags, the advertisement is included in the delivery advertisement set, thereby determining the delivery advertisement set for the user.

[0016] As a further solution of the present invention: in the step 2, the specific method of determining the specific device type of the user and the current network status of the device according to the device information is:

[0017] AS1: divides device types into three categories: mobile phones, tablets, and computers;

[0018] AS2: Identify device information through User-Agent identification technology to determine the specific device type in the device information;

[0019] AS3: Next, use the Network Information API to get the device's current network speed.

[0020] As a further solution of the present invention: in step 3, the specific method of determining a placement of advertisement by the pre-selected model is:

[0021] BS1: Divide the delivered advertising set into a behavioral interest advertising subset and a potential interest advertising subset; and mark them as and ;

[0022] BS2: Set a delivery probability value , where the probability value is the default value;

[0023] BS3: Then generate a random number ,Compare and the probability of delivery The size of the behavior interest advertising subset or the potential interest advertising subset is determined according to the comparison result to deliver an advertisement; wherein, ;

[0024] BS31: If , from the Behavioral Interest-Based Advertising subset An advertisement is randomly selected to be placed;

[0025] BS32: If , from the Interest-Based Ads subset An advertisement is randomly selected from the list to be delivered.

[0026] As a further solution of the present invention: the method of dividing the delivered advertisement set into a behavioral interest advertisement set and a potential interest advertisement set is as follows;

[0027] Summarize the advertisements to be delivered, define one or more advertisement tag words for each advertisement according to the content of each advertisement, compare the advertisement tag words with the historical behavior interest tags and the potential interest tags one by one, and if an interest tag identical to the advertisement tag word is found in the historical behavior interest tags, classify the advertisement into the behavior interest advertisement subset;

[0028] If an interest tag identical to the advertisement tag word is found in the potential interest tags, the advertisement is classified into a potential interest advertisement subset.

[0029] As a further solution of the present invention: in the step 4, the specific method of optimizing the pre-delivered advertisement according to the device type and network speed is:

[0030] CS1: Define device type and network speed: Device type , the network speed is recorded as ,in, Represents a mobile phone, Represented as a flat plate, Represented as a computer;

[0031] CS2: Define available resolutions for pre-launch ads And the corresponding bit rate ,in, , is the total number of available resolutions;

[0032] CS3: Set the maximum resolution supported by each device type to ;

[0033] for : ;

[0034] for : ;

[0035] for : ;

[0036] in, is the default value;

[0037] CS4: Then calculate each available resolution using the following formula Minimum network speed required ;

[0038]

[0039] In the formula, is the redundancy factor;

[0040] CS5: And according to the user's network speed Determine the highest available resolution using the following formula :

[0041]

[0042] CS6: Select the maximum resolution supported by the device type. and the highest available resolution supported by the device The minimum value in:

[0043]

[0044] CS7: Then determine the layout format of the pre-launched ad based on the device type, as follows:

[0045]

[0046] In the formula, They are respectively represented as vertical screen layout, adaptive layout and horizontal screen layout.

[0047] An intelligent advertising delivery optimization system, comprising:

[0048] The information collection module is used to collect the user's historical behavior information and the user's current device information;

[0049] An advertisement set determination module is used to generate a corresponding historical behavior interest tag according to the historical behavior information, and predetermine a potential interest tag different from the historical behavior interest tag according to the user's age, gender characteristics, and geographic location, and then determine the user's advertisement set through the historical behavior interest tag or the potential interest tag;

[0050] The device identification module is used to determine the specific device type used by the user and the current network speed of the device based on the device information;

[0051] A pre-delivery advertisement module is used to determine a delivery advertisement in the delivery advertisement set through a pre-selection model and mark it as a pre-delivery advertisement;

[0052] The optimization module is used to optimize the pre-delivered advertisements according to the device type and network speed, and deliver the optimized advertisements.

[0053] The specific contents of the optimization module include:

[0054] Define the device type and network speed: Device Type , the network speed is recorded as N, where, Represents a mobile phone, Represented as a flat plate, Represented as a computer;

[0055] Defines the resolutions available for pre-order ads And the corresponding bit rate ,in, , n is the total number of available resolutions;

[0056] Set the maximum resolution supported by each device type to ;

[0057] for : ;

[0058] for : ;

[0059] for : ;

[0060] in, is the default value;

[0061] Then calculate each available resolution using the following formula Minimum network speed required ;

[0062]

[0063] In the formula, is the redundancy factor;

[0064] And according to the user's network speed Determine the highest available resolution using the following formula :

[0065]

[0066] Select the maximum resolution supported by the device type. and the highest available resolution supported by the device The minimum value in:

[0067]

[0068] CS7: Then determine the layout format of the pre-launched ad based on the device type, as follows:

[0069]

[0070] In the formula, Represented as vertical screen layout, adaptive layout and horizontal screen layout respectively

[0071] The delivery module is used to deliver the pre-delivered advertisements after optimization.

[0072] The present invention provides a smart advertising delivery optimization method and system. Compared with the prior art, it has the following beneficial effects:

[0073] The present invention dynamically adjusts the advertising delivery strategy through comprehensive analysis of user behavior data, device type and network status, significantly improving the accuracy and personalization of advertising. It combines historical behavior interests with potential interest tags to ensure that the advertising content is highly matched with user needs, which can not only meet current interests but also guide potential needs. Through pre-selection models and optimization processing, advertising delivery achieves diversified and personalized display, greatly improving user click-through rate and interactivity.

[0074] In addition, the resolution, bit rate and layout of ads are automatically adjusted according to the device type and network conditions to ensure that ads can be loaded quickly and displayed in the best state, reducing freezes and loading delays, and improving the user's viewing experience; through this series of intelligent optimizations, ads can instantly adapt to the user's current usage environment to achieve smooth and efficient ad display. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] The present invention will be further described below in conjunction with the accompanying drawings.

[0076] Figure 1 A flowchart of the steps of an intelligent advertising delivery optimization method of the present invention;

[0077] Figure 2 This is a structural framework diagram of an intelligent advertising delivery optimization system of the present invention. DETAILED DESCRIPTION

[0078] 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.

[0079] Embodiment 1

[0080] See also Figure 1 ,The present invention provides an advertising delivery management method based on artificial intelligence, comprising:

[0081] Step 1: Collect the user's historical behavior information, generate corresponding historical behavior interest tags according to the historical behavior information, and predetermine potential interest tags different from the historical behavior interest tags according to the user's age, gender characteristics, and geographic location, and then determine the user's advertising set through the historical behavior interest tags or the potential interest tags;

[0082] The method of generating corresponding historical behavior interest tags according to historical behavior information is as follows:

[0083] Collect and preprocess the user's online historical behavior information, then classify the historical behavior information and extract the keywords of interest in each category through natural language processing (NLP) ， Generate historical behavior interest tags based on the extracted interest keywords;

[0084] Specifically, the classification of historical behavior information includes the following:

[0085] Browsing history information: content or web pages that a user has viewed on a website, app or social media platform;

[0086] Shopping record information: the types, brands, and price ranges of goods purchased by users on e-commerce platforms;

[0087] Video viewing history information: content watched and interactions of users on video platforms;

[0088] The method of predetermining potential interest tags different from historical behavior interest tags based on the user's age, gender characteristics, and geographic location is as follows:

[0089] First, by collecting the user's age information, find the corresponding category from the preset age interest tag classification according to the user's age information, and find the tag different from the historical behavior interest tag according to the interest tag of this category as the user's potential interest tag;

[0090] Then, according to the user's gender characteristics, a corresponding category is found from the preset gender interest tag classification, and according to the interest tags of this category, a tag different from the historical behavior interest tag is found as the user's potential interest tag;

[0091] Finally, based on the user's geographic location, and based on the local lifestyle, cultural background, and consumption habits, find tags that are different from historical behavior interest tags from the preset geographic location interest tag classification as the user's potential interest tags;

[0092] Exemplarily, the preset age interest tag classification is divided according to age, and in the divided age groups, professional personnel pre-fill interest tags that match the age group for each age group as interest tags for the category; for example, the age group of 18-25 corresponds to a class in an age interest tag, and there are multiple interest tags preset in this class; in the preset gender interest tags, professional personnel pre-fill interest tags that match the gender for people of different genders as interest tags for the analogy; in the preset geographic location interest tags, professional personnel pre-fill interest tags that match the geographic location according to the local lifestyle, cultural background and consumption habits as interest tags for the analogy;

[0093] The specific method of determining the advertisement set for the user through the historical behavior interest tag or the potential interest tag is:

[0094] Summarize the advertisements to be delivered, define one or more advertisement tag words for each advertisement according to the content of each advertisement, compare the advertisement tag words with the historical behavior interest tags and the potential interest tags one by one, and if an interest tag identical to the advertisement tag word is found in the historical behavior interest tags or the potential interest tags, include the advertisement in the delivered advertisement set, thereby determining the delivered advertisement set of the user;

[0095] Step 2: Collect the device information currently used by the user, and determine the specific device type used by the user and the current network speed of the device according to the device information, wherein the device information includes the device type and the network speed;

[0096] The specific method of determining the specific device type of the user and the current network status of the device according to the device information is:

[0097] AS1: divides device types into three categories: mobile phones, tablets, and computers;

[0098] AS2: Identify device information through User-Agent identification technology to determine the specific device type in the device information;

[0099] AS3: Then use the Network Information API to get the current network speed of the device;

[0100] It should be noted that modern browsers support the Network Information API, which can provide users with network connection information, including network type (such as Wi-Fi, 4G, 5G, etc.) and estimated downlink bandwidth. Through this API, you can obtain the user's network status, such as network speed, connection type, latency, etc.

[0101] Specifically, the collection permissions in steps 1 and 2 need to be clearly informed to the user. If the user does not grant permissions, the subsequent processes cannot be carried out.

[0102] Step 3: Determine an ad to be delivered in the delivery ad set through the pre-selection model and mark it as a pre-delivery ad;

[0103] The specific method of determining an advertisement to be placed by the pre-selection model is:

[0104] BS1: Divide the delivered advertising set into a behavioral interest advertising subset and a potential interest advertising subset; and mark them as and ;

[0105] The method of dividing the delivered advertisement set into the behavioral interest advertisement set and the potential interest advertisement set is:

[0106] Summarize the advertisements to be delivered, define one or more advertisement tag words for each advertisement according to the content of each advertisement, compare the advertisement tag words with the historical behavior interest tags and the potential interest tags one by one, and if an interest tag identical to the advertisement tag word is found in the historical behavior interest tags, classify the advertisement into the behavior interest advertisement subset;

[0107] If an interest tag identical to the advertisement tag word is found in the potential interest tags, the advertisement is classified into a potential interest advertisement subset;

[0108] BS2: Set a delivery probability value , where the delivery probability value Set up by professional staff;

[0109] BS3: Then generate a random number ,Compare and the probability of delivery The size of the behavior interest advertising subset or the potential interest advertising subset is determined according to the comparison result to deliver an advertisement; wherein, ;

[0110] BS31: If , from the Behavioral Interest-Based Advertising subset An advertisement is randomly selected to be placed;

[0111] BS32: If , from the Interest-Based Ads subset An advertisement is randomly selected to be placed;

[0112] The pre-selection model uses a combination of random numbers and delivery probabilities to balance the selection between behavioral interest ads and potential interest ads. This mechanism can instantly bring about diversified ad displays, so that users no longer receive only a single type of ad content, thereby avoiding ad fatigue and maintaining user interest in ads. At the same time, the ads delivered are closely integrated with users' actual behaviors and potential needs, immediately improving the relevance of ads, making the ad content more in line with users' current situations and interests, and making it easier for users to click and interact.

[0113] Step 4: Optimizing the pre-delivered advertisements according to the device type and network speed, and delivering the optimized advertisements;

[0114] The specific method of optimizing the pre-delivered advertisement according to the device type and network speed is:

[0115] CS1: Define device type and network speed: Device type , the network speed is recorded as ,in, Represents a mobile phone, Represented as a flat plate, Represented as a computer;

[0116] CS2: Define available resolutions for pre-launch ads And the corresponding bit rate ,in, and Each specific parameter value is selected by professional staff, among which, ; is the total number of available resolutions, which generally does not exceed 5;

[0117] In this embodiment, the available resolution It is 480P, and the corresponding bit rate is 1Mbps;

[0118] Available resolutions It is 720P, and the corresponding bit rate is 2.5Mbps;

[0119] Available resolutions For 1080P, the corresponding bit rate is 5Mbps;

[0120] Available resolutions For 4k, the corresponding bit rate is 15Mbps;

[0121] CS3: Set the maximum resolution supported by each device type to ;

[0122] for (cell phone): ;

[0123] for (flat): ;

[0124] for (computer): ;

[0125] in, is a preset value, and the specific parameters are selected by professional staff. 1080p, 2k and 4k respectively;

[0126] CS4: Then calculate each available resolution using the following formula Minimum network speed required ;

[0127]

[0128] In the formula, is the redundancy factor, taking into account network fluctuations and other overhead;

[0129] In this embodiment, , the calculation results are: N1=1.5, N2=3.75, N3=7.5, N4=22.5;

[0130] CS5: And according to the user's network speed Determine the highest available resolution using the following formula :

[0131]

[0132] In this embodiment, when the user's network speed is At 6Mbps, determine the highest resolution available yes 720P;

[0133] CS6: Select the maximum resolution supported by the device type. and the highest available resolution supported by the device The minimum value in:

[0134]

[0135] CS7: Then determine the layout format of the pre-launched ad based on the device type, as follows:

[0136]

[0137] In the formula, Represented as vertical screen layout, adaptive layout and horizontal screen layout respectively;

[0138] In this implementation, if the device type currently used by the user is a tablet, the maximum resolution supported by the selected device type is and the highest available resolution supported by the device The minimum value in If the resolution is 720P, the pre-launched ads are adjusted to 720P resolution and delivered to the user's device through adaptive layout format for display;

[0139] By detecting the device type and network speed, the system can instantly optimize the resolution, bit rate and layout of ads, ensuring that ads can be displayed in the best state on different devices, significantly shortening the loading time of ads and making the user experience smoother. At the same time, the resolution of ads is adjusted according to the user's network status to avoid freezes or low-quality displays caused by unstable networks, so that ads can be played quickly and maintain high visual effects in any environment, thereby immediately enhancing the viewing experience and appeal of ads.

[0140] Embodiment 2

[0141] See also Figure 2 In the specific implementation process, this embodiment is based on the first embodiment and is different from the first embodiment in that this embodiment provides an intelligent advertising delivery optimization system, including:

[0142] The information collection module is used to collect the user's historical behavior information and the user's current device information;

[0143] An advertisement set determination module is used to generate a corresponding historical behavior interest tag according to the historical behavior information, and predetermine a potential interest tag different from the historical behavior interest tag according to the user's age, gender characteristics, and geographic location, and then determine the user's advertisement set through the historical behavior interest tag or the potential interest tag;

[0144] The device identification module is used to determine the specific device type used by the user and the current network speed of the device based on the device information;

[0145] A pre-delivery advertisement module is used to determine a delivery advertisement in the delivery advertisement set through a pre-selection model and mark it as a pre-delivery advertisement;

[0146] An optimization module, used for optimizing the pre-delivered advertisements according to the device type and network speed, and delivering the optimized advertisements;

[0147] The specific contents of the optimization module include:

[0148] Define the device type and network speed: Device Type , the network speed is recorded as ,in, Represents a mobile phone, Represented as a flat plate, Represented as a computer;

[0149] Defines the resolutions available for pre-order ads And the corresponding bit rate ,in, , is the total number of available resolutions;

[0150] Set the maximum resolution supported by each device type to ;

[0151] for (cell phone): ;

[0152] for (flat): ;

[0153] for (computer): ;

[0154] in, is the default value;

[0155] Then calculate each available resolution using the following formula Minimum network speed required ;

[0156]

[0157] In the formula, is the redundancy factor;

[0158] And according to the user's network speed Determine the highest available resolution using the following formula :

[0159]

[0160] Select the maximum resolution supported by the device type. and the highest available resolution supported by the device The minimum value in:

[0161]

[0162] CS7: Then determine the layout format of the pre-launched ad based on the device type, as follows:

[0163]

[0164] In the formula, Represented as vertical screen layout, adaptive layout and horizontal screen layout respectively;

[0165] A delivery module is used to deliver the optimized pre-delivered advertisements;

[0166] Embodiment 3

[0167] The specific implementation process of this embodiment includes all the implementation processes of the above three groups of embodiments.

[0168] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0169] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A smart advertising delivery optimization method, characterized in that: include: Step 1: Based on user authorization, collect the user's historical behavior information, generate corresponding historical behavior interest tags according to the historical behavior information, and predetermine potential interest tags different from the historical behavior interest tags according to the user's age, gender characteristics, and geographic location, and then determine the user's advertising set through the historical behavior interest tags or the potential interest tags; Step 2: Based on the user's authorization, collect the user's currently used device information, and determine the specific device type used by the user and the current network speed of the device according to the device information, wherein the device information includes the device type and the network speed; Step 3: Determine an ad to be delivered in the delivery ad set through the pre-selection model and mark it as a pre-delivery ad; Step 4: Optimizing the pre-delivered advertisements according to the device type and network speed, and delivering the optimized advertisements; In the step 4, the specific method of optimizing the pre-delivered advertisement according to the device type and network speed is: CS1: Define device type and network speed: Device type , the network speed is recorded as ,in, Represents a mobile phone, Represented as a flat plate, Represented as a computer; CS2: Define available resolutions for pre-launch ads And the corresponding bit rate ,in, , is the total number of available resolutions; CS3: Set the maximum resolution supported by each device type to ; for : ; for : ; for : ; in, is the default value; CS4: Then calculate each available resolution using the following formula Minimum network speed required ; In the formula, is the redundancy factor; CS5: And according to the user's network speed Determine the highest available resolution using the following formula : CS6: Select the maximum resolution supported by the device type. and the highest available resolution supported by the device The minimum value in: CS7: Then determine the layout format of the pre-launched ad based on the device type, as follows: In the formula, They are respectively represented as vertical screen layout, adaptive layout and horizontal screen layout.

2. The intelligent advertising delivery optimization method according to claim 1, characterized in that: In the step 1, the method of predetermining potential interest tags different from historical behavior interest tags based on the user's age, gender characteristics, and geographic location is: First, by collecting the user's age information, find the corresponding category from the preset age interest tag classification according to the user's age information, and find the tag different from the historical behavior interest tag according to the interest tag of this category as the user's potential interest tag; Then, according to the user's gender characteristics, a corresponding category is found from the preset gender interest tag classification, and according to the interest tags of this category, a tag different from the historical behavior interest tag is found as the user's potential interest tag; Finally, based on the user's geographic location and local lifestyle, cultural background and consumption habits, tags different from historical behavior interest tags are found from the preset geographic location interest tag classification as potential interest tags for the user.

3. The intelligent advertising delivery optimization method according to claim 1, characterized in that: The specific method of determining the advertisement set for the user through the historical behavior interest tag or the potential interest tag is: The advertisements to be delivered are summarized, one or more advertisement tag words are defined for each advertisement according to its content, and the advertisement tag words are compared one by one with the historical behavior interest tags and the potential interest tags. If an interest tag identical to the advertisement tag word is found in the historical behavior interest tags or the potential interest tags, the advertisement is included in the delivery advertisement set, thereby determining the delivery advertisement set for the user.

4. The intelligent advertising delivery optimization method according to claim 1, characterized in that: In step 2, the specific method of determining the user's specific device type and the current network status of the device according to the device information is: AS1: divides device types into three categories: mobile phones, tablets, and computers; AS2: Identify device information through User-Agent identification technology to determine the specific device type in the device information; AS3: Next, use the Network Information API to get the device's current network speed.

5. The intelligent advertising delivery optimization method according to claim 1, characterized in that: In step 3, the pre-selection model determines a specific method for placing an advertisement as follows: BS1: Divide the delivered advertising set into a behavioral interest advertising subset and a potential interest advertising subset; and mark them as and ; BS2: Set a delivery probability value , where the probability value is the default value; BS3: Then generate a random number ,Compare and the probability of delivery The size of the behavior interest advertising subset or the potential interest advertising subset is determined according to the comparison result to deliver an advertisement; wherein, ; BS31: If , from the Behavioral Interest-Based Advertising subset An advertisement is randomly selected to be placed; BS32: If , from the Interest-Based Ads subset An advertisement is randomly selected from the list to be delivered.

6. The intelligent advertising delivery optimization method according to claim 5, characterized in that: The method of dividing the delivered advertisement set into the behavioral interest advertisement set and the potential interest advertisement set is: Summarize the advertisements to be delivered, define one or more advertisement tag words for each advertisement according to the content of each advertisement, compare the advertisement tag words with the historical behavior interest tags and the potential interest tags one by one, and if an interest tag identical to the advertisement tag word is found in the historical behavior interest tags, classify the advertisement into the behavioral interest advertisement subset middle; If an interest tag identical to the advertisement tag word is found in the potential interest tags, the advertisement is classified into a potential interest advertisement subset. middle.

7. An intelligent advertising delivery optimization system, characterized in that: include: The information collection module is used to collect the user's historical behavior information and the user's current device information; An advertisement set determination module is used to generate a corresponding historical behavior interest tag according to the historical behavior information, and predetermine a potential interest tag different from the historical behavior interest tag according to the user's age, gender characteristics, and geographic location, and then determine the user's advertisement set through the historical behavior interest tag or the potential interest tag; The device identification module is used to determine the specific device type used by the user and the current network speed of the device based on the device information; A pre-delivery advertisement module is used to determine a delivery advertisement in the delivery advertisement set through a pre-selection model and mark it as a pre-delivery advertisement; An optimization module, used for optimizing the pre-delivered advertisements according to the device type and network speed, and delivering the optimized advertisements; A delivery module is used to deliver the optimized pre-delivered advertisements; The specific contents of the optimization module include: Define the device type and network speed: Device Type , the network speed is recorded as N, where D=1 represents a mobile phone, D=2 represents a tablet, and D=3 represents a computer; Defines the resolutions available for pre-order ads And the corresponding bit rate ,in, , n is the total number of available resolutions; Set the maximum resolution supported by each device type to ; for : ; for : ; for : ; in, is the default value; Then calculate each available resolution using the following formula Minimum network speed required ; In the formula, is the redundancy factor; And according to the user's network speed Determine the highest available resolution using the following formula : Select the maximum resolution supported by the device type. and the highest available resolution supported by the device The minimum value in: CS7: Then determine the layout format of the pre-launched ad based on the device type, as follows: In the formula, They are respectively represented as vertical screen layout, adaptive layout and horizontal screen layout.

Images