Intelligent advertising system and method for advertising space

Abstract

The invention discloses an intelligent advertising system and an intelligent advertising method, relates to the technical field of advertising, and solves the problems of advertising load, advertisement lagging and poor advertising effect caused by the fact that corresponding related network features are not confirmed in advance. According to the invention, through multi-dimensional parameter monitoring (CPU occupancy rate, load, memory and power consumption) in the pre-regulation and control stage, the optimal frame rate is dynamically locked in combination with the hardware characteristics of the advertisement space, the optimal matching between the advertisement playing effect and the equipment performance is realized, the hardware overload is avoided while the picture quality is ensured, and the equipment operation stability is improved; and in combination with real-time bandwidth monitoring data, the system can accurately calculate the frame rate adjustment amplitude according to the ratio of a fluctuation value to a characteristic variable value, and millisecond-level dynamic response is realized. The adaptive adjustment mechanism not only ensures the video fluency, but also avoids the image quality loss caused by excessive frame drop, and can maintain stable user experience in different network environments.

Description

Technical Field

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

[0002] Ad space refers to a specific location or area in various media platforms or physical spaces that is specifically used to display advertising content; The application with publication number CN105872637A discloses an on-screen display-based advertising delivery method and a smart TV, the method comprising determining whether the current playback channel is a carousel channel, and if so, requesting advertising materials for advertising delivery from a server after receiving a notification of the display screen; after receiving the advertising materials provided by the server, displaying the advertising materials on a reserved advertising position on the on-screen display until the on-screen display disappears; the smart TV comprises an advertising delivery system capable of implementing the advertising delivery method; the inventive method and the smart TV utilize the on-screen display of a carousel channel for advertising delivery, and utilizing the on-screen display for advertising delivery can achieve the beneficial effect of not significantly reducing the user experience and significantly increasing the exposure rate of the delivered advertisements.

[0003] The ad slot is an intelligent delivery process for ad content. Generally, the ad content is delivered directly based on the original device parameters of the ad slot. However, this delivery process is relatively rough. First, different ad slots have different parameter characteristics. When delivering ads, the frame rate characteristics of the ads can be adjusted based on the relevant characteristics of the corresponding ad slot, so that the ad slot can achieve the best delivery effect. Second, when the internal bandwidth of the ad slot fluctuates, the corresponding ad frame rate will fluctuate. However, because the corresponding relevant characteristics are not confirmed in advance, the CPU still needs to confirm the characteristics, which causes the ad slot to be loaded. When the load is serious, it will cause the ad to be stuck, resulting in poor delivery effect. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides an intelligent advertising placement system and method, which solves the problem that the corresponding relevant network features are not confirmed in advance, resulting in advertising placement load, advertising freeze, and poor placement effect.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for intelligently placing advertisements, comprising the following steps: Step 1: Pre-regulate the frame rate of the advertisement to be placed in the specified advertisement slot. Based on the pre-regulated frame rate change stage, obtain the advertisement slot operation parameters associated with each frame rate change stage. Based on the operation interval associated with the operation item corresponding to the advertisement slot and the specific change characteristics, lock the optimal frame rate that needs to be regulated for the advertisement in the specified advertisement slot. The specific sub-steps are as follows: S11. Based on the advertisement and the determined advertisement position, the advertisement position is made to conduct a trial run for the advertisement. During the trial run, the frame rate of the advertisement is gradually adjusted from low to high, and the minimum and maximum frame rates are both characteristic values ​​preset for the advertisement. The frame rate is increased by 1 Hz for a period of 3 seconds during the playback phase, and the advertisement position operation parameters associated with the end of the 3-second playback phase are confirmed, and different operation parameters of the advertisement positions associated with different increase processes are recorded as different process parameters. S12, confirming the operation item associated with the operation parameter of the corresponding ad slot, and then confirming the operation interval associated with the operation item, wherein the operation interval is a preset interval, and then confirming the process characteristics of different improvement processes based on the confirmed process parameters, specifically in the following manner: S121, for a single group of improvement processes, comparing different process parameters with the operation range of the corresponding operation items, if all process parameters belong to the operation range, then the improvement process is marked as the process to be selected, otherwise, no marking is performed; S122, based on the calibrated process to be selected, identify the location of the corresponding process parameter in the operation interval, and divide each operation interval into ten equal parts, and calibrate the equal division ratio of the front end of the process parameter as QB i-k , mark the equal division ratio of the process parameter backend as HB i-k , where i represents different processes to be selected and k represents different process parameters; S123, multiple groups of QB associated with different process parameters in the same process to be selected i-k Perform variance processing, confirm the front-end variance, and then associate multiple groups of HBs associated with different process parameters i-k Perform variance processing to confirm the back-end variance, then perform mean processing on the confirmed front-end variance and back-end variance, and use the processed mean as the process feature associated with this improvement process; S13, based on different process characteristics associated with different improved processes, the improved process that satisfies the process characteristic ≤ Y1 is calibrated as the standard process, otherwise no calibration is performed, where Y1 is a preset value, and based on different frame rates associated with different standard processes, the frame rate with the maximum value is selected as the optimal frame rate; Step 2: Based on the confirmed best frame rate and other ads that were previously placed in this ad slot, other ads with the same frame rate as the best frame rate are recorded as feature ads. Based on the feature ad bandwidth and frame rate variation characteristics, the standard feature segment is locked, and the feature variable value is determined from the standard feature segment. The specific method is as follows: S21, using the confirmed best frame rate as a benchmark, identifying other advertisements with the same frame rate as the benchmark from the historical delivery data of the advertisement position, and recording the identified other advertisements as feature advertisements; S22, confirming the bandwidth fluctuation period associated with the characteristic advertisement during the playback process: when the bandwidth change value associated with the adjacent moments exceeds 2Mbps, the adjacent moments are recorded into the bandwidth fluctuation period, and the bandwidth fluctuation curve associated with the bandwidth fluctuation period is confirmed, and then the frame rate fluctuation curve associated with the bandwidth fluctuation period is confirmed from the playback process; Selecting the same-frequency characteristic segment from the confirmed bandwidth fluctuation curve and frame rate fluctuation curve: identifying the partial curve segments with the same trend in the same period from the two groups of curves, and marking the identified partial curve segments as the same-frequency characteristic segment; the trend includes: synchronous rise and synchronous fall; S23, confirm the initial moment and the end moment of the same-frequency feature segment, record the same-frequency feature segment belonging to the bandwidth fluctuation curve as the main segment, record the same-frequency feature segment belonging to the frame rate fluctuation curve as the secondary segment, confirm the change characteristics of adjacent moments in the main segment and the secondary segment: mark the numerical feature associated with the moment before the adjacent moment as Tz1, mark the numerical feature associated with the moment after the adjacent moment as Tz2, and use: (Tz2-Tz1) ÷ adjacent moment duration = change characteristics; S24, executing several analysis processes, and the time characteristics associated with each analysis process are different, and the specific method is as follows: S241, keeping the end time unchanged, extending the initial time back by one time, recording the extended time period as the processing time period, performing standard deviation processing on several groups of change features associated with the processing time period, and confirming the analysis features associated with the processing time period; S242, each time a processing period is executed, it is extended one time backward based on the position of the initial time of the previous processing period, and the same processing method of the processing period is used to confirm the analysis features associated with each different processing period, and the processing is stopped when the initial time and the end time of the corresponding processing period are adjacent time, and the processing process of the first analysis process is completed; S243, moving the end time forward by one time, and then using the same processing method as steps S241-S242 to move the initial time, and confirming the analysis features associated with each different processing period; S244, repeating step S243 until the initial time and the terminal time before the corresponding analysis process are processed are adjacent times; S25, selecting a minimum value from the plurality of groups of analysis features confirmed by the plurality of analysis processes, and recording the processing period associated with the minimum value as the standard period, recording the curve segment associated with the standard period in the bandwidth fluctuation curve as the primary standard feature segment, and recording the curve segment associated with the standard period in the frame rate fluctuation curve as the secondary standard feature segment; S26, performing ratio processing on the change characteristics associated with the primary standard characteristic segment and the secondary standard characteristic segment in the same period, confirming the characteristic ratio, wherein the characteristic ratio = the change characteristics within the primary characteristic segment ÷ the change characteristics within the secondary characteristic segment, performing average processing on several groups of characteristic ratios confirmed within the standard period, and locking the characteristic variation value; Step 3: According to the confirmed characteristic variable value, during the process of advertising delivery in this advertising position, the frame rate of the advertisement is adjusted in real time based on the fluctuation characteristics of the network bandwidth of the advertising position.

[0006] Preferably, an intelligent advertising delivery system includes: The frame rate confirmation end obtains the ad slot operation parameters associated with each frame rate change stage based on the pre-regulated frame rate change stage, and locks the optimal frame rate that needs to be regulated for the specified ad slot based on the operation range and specific change characteristics associated with the operation items corresponding to the ad slot; The characteristic variable value confirmation end records other advertisements with the same frame rate as the best frame rate as characteristic advertisements based on the confirmed best frame rate and other advertisements previously placed in this advertisement position, and then locks the standard characteristic segment based on the characteristic advertisement bandwidth and frame rate variation characteristics, and then determines the characteristic variable value from the standard characteristic segment; The frame rate delivery control end, according to the confirmed characteristic variable value, adjusts the frame rate of the advertisement in real time during the advertisement delivery process in this advertisement position based on the fluctuation characteristics of the network bandwidth of the advertisement position.

[0007] The present invention provides an intelligent advertising placement system and method. Compared with the prior art, it has the following beneficial effects: The present invention monitors multi-dimensional parameters (CPU occupancy, load, memory and power consumption) in the pre-regulation stage, and dynamically locks the optimal frame rate in combination with the hardware characteristics of the ad slot, thereby achieving the optimal match between the ad playback effect and the device performance, avoiding hardware overload while ensuring the picture quality, and improving the stability of device operation; Combined with real-time bandwidth monitoring data, the system can accurately calculate the frame rate adjustment range based on the ratio of the fluctuation value to the characteristic variable value, achieving millisecond-level dynamic response. This adaptive adjustment mechanism not only ensures video smoothness, but also avoids image quality loss caused by excessive frame drops, and can maintain a stable user experience in different network environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0009] 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.

[0010] First embodiment See also Figure 1 The present application provides a method for intelligently placing advertisements, comprising the following steps: Step 1: pre-regulate the frame rate of the advertisement to be placed in the specified ad slot, obtain the ad slot operation parameters associated with each frame rate change stage based on the pre-regulated frame rate change stage, and lock the optimal frame rate that needs to be regulated for the specified ad slot based on the operation range and specific change characteristics associated with the corresponding operation item of the ad slot. Specifically, the ad slot operation parameters include CPU occupancy, CPU load, memory usage, and power consumption. Different ad slots have different hardware differences. Therefore, in the specific delivery and display process of the same type of advertisements, it is necessary to adjust the specific frame rate of the advertisement based on the specific parameter attributes of different ad slots. Based on the parameter changes of the corresponding ad slot, select the relevant frame rate that is most suitable for resolution for the corresponding ad slot, so as to ensure the subsequent output effect, and the advertisement that needs to be placed in the corresponding ad slot has been fully determined in advance; The specific sub-steps for locking the optimal frame rate are: S11. Based on the advertisement and the determined advertisement position, the advertisement position is made to conduct a trial run for the advertisement. During the trial run, the frame rate of the advertisement is gradually adjusted from low to high. The minimum and maximum frame rates are both characteristic values ​​preset for the advertisement. The frame rate is increased by 1 Hz for a 3-second playback phase. The advertisement position operation parameters associated with the end of the 3-second playback phase are confirmed. Different operation parameters of the advertisement positions associated with different increase processes are recorded as different process parameters. Each increase process lasts for 3 seconds, and the frame rate of each increase process is increased by 1 Hz. In each increase process, the advertisement position parameters change according to the frame rate change of the corresponding advertisement. S12. Confirm the operation item associated with the operation parameter of the corresponding ad slot, and then confirm the operation interval associated with the operation item. The operation intervals are all preset intervals. Different operation items have different operation intervals. The preset interval of the CPU occupancy operation item is generally 20%-50%, and the preset interval of the CPU load operation item is generally 0.5-1.5. The memory usage is generally 50MB-200MB, and the power consumption is generally 20%-50%. Then, based on the confirmed process parameters, confirm the process characteristics of different improvement processes: S121. For a single group of improvement processes, the different process parameters are compared with the operation range of the corresponding operation items. If all the process parameters belong to the operation range, the improvement process is marked as the process to be selected. Otherwise (that is, in the corresponding improvement process, as long as there is a group of parameters that do not belong to the corresponding operation range, it is the opposite case), no marking is performed; S122, based on the calibrated process to be selected, identify the location of the corresponding process parameter in the operation interval, and divide each operation interval into ten equal parts (the equal division distances of different operation intervals are different), and calibrate the equal division ratio of the front end of the process parameter as QB i-k , mark the equal division ratio of the process parameter backend as HB i-k , where i represents different processes to be selected, and k represents different process parameters. For example, if the running item is CPU occupancy, the process parameter associated with the process to be selected is 30%, and 20%-50% is divided into ten equal parts. The ratio before 30% is 3.33, and the ratio after 30% is 6.66; S123, multiple groups of QB associated with different process parameters in the same process to be selected i-k Perform variance processing, confirm the front-end variance, and then associate multiple groups of HBs associated with different process parameters i-k Perform variance processing, confirm the back-end variance, and then perform mean processing on the confirmed front-end variance and back-end variance, and use the processed mean as the process feature associated with this improvement process. The variance processing method is relatively common in the prior art, so it will not be described in detail here; S13, based on different process characteristics associated with different improvement processes, the improvement process that satisfies the process characteristic ≤ Y1 is calibrated as the standard process, otherwise no calibration is performed, where Y1 is a preset value, generally 5, and its corresponding parameter is located in the middle of the preset interval, and the variance generated is generally small, based on different frame rates associated with different standard processes, the maximum frame rate is selected as the optimal frame rate, and the frame rate of the advertisement to be placed in this advertisement position is adjusted based on the optimal frame rate; Specifically, after the optimal frame rate associated with the corresponding ad slot is specifically confirmed, the advertisement can be delivered based on the associated optimal frame rate, which can not only ensure the specific delivery effect of the corresponding advertisement, but also ensure the normal operation of the corresponding ad slot; Step 2: In the past operation stage, there were time periods when the bandwidth fluctuated. When the bandwidth fluctuated, the data transmission speed was limited, and the advertising video data could not be transmitted to the playback device in a timely and complete manner. In order to ensure the basic playback smoothness of the video, the playback system will reduce the frame rate, that is, reduce the number of frames displayed per second. For example, an advertisement that could originally be played smoothly at 30 frames per second (fps) may drop to 15fps or even lower when the bandwidth is insufficient, and the picture will appear stuck and incoherent. Based on the confirmed best frame rate and other advertisements that have been delivered to this ad slot in the past, other advertisements with the same frame rate as the best frame rate are recorded as feature advertisements. Based on the characteristic advertisement bandwidth and frame rate change characteristics, the standard feature segment is locked, and the feature variable value is determined from the standard feature segment. The determination of the feature variable value is to achieve early changes to the advertisement in the subsequent bandwidth fluctuation process to ensure that the corresponding ad slot will not be overloaded due to the processing process. If the load is too heavy, it will be stuck or other situations will occur, which will affect the corresponding delivery and playback effect. The specific sub-steps for confirmation are: S21, using the confirmed best frame rate as a benchmark, identifying other advertisements with the same frame rate as the benchmark from the historical delivery data of the advertisement position, and recording the identified other advertisements as feature advertisements; S22, confirming the bandwidth fluctuation period associated with the characteristic advertisement during the playback process: when the bandwidth change value associated with the adjacent moments exceeds 2Mbps, the adjacent moments are recorded as the bandwidth fluctuation period, and the bandwidth fluctuation curve associated with the bandwidth fluctuation period is confirmed (the horizontal coordinate of this curve is the time feature, and the vertical coordinate is the bandwidth feature), and then confirming the frame rate fluctuation curve associated with the bandwidth fluctuation period from the playback process (the horizontal coordinate of this curve is the time feature, and the vertical coordinate is the frame rate feature); Selecting the same-frequency characteristic segment from the confirmed bandwidth fluctuation curve and frame rate fluctuation curve: identifying the partial curve segments with the same trend in the same period from the two groups of curves, and marking the identified partial curve segments as the same-frequency characteristic segment; the trend includes: synchronous rise and synchronous fall; S23, confirm the initial moment and the end moment of the same-frequency feature segment, record the same-frequency feature segment belonging to the bandwidth fluctuation curve as the main segment, record the same-frequency feature segment belonging to the frame rate fluctuation curve as the secondary segment, and confirm the change characteristics of adjacent moments in the main segment and the secondary segment: mark the numerical feature associated with the moment before the adjacent moment as Tz1, and mark the numerical feature associated with the moment after the adjacent moment as Tz2, using: (Tz2-Tz1) ÷ adjacent moment duration = change characteristics, the adjacent moment duration is generally one second, that is, the interval between two adjacent moments; S24. Execute several analysis processes, and the time characteristics associated with each analysis process are different (that is, the time periods are not exactly the same): S241, keeping the end time unchanged, extending the initial time back by one time, recording the extended time period as the processing time period, performing standard deviation processing on several groups of change features associated with the processing time period, and confirming the analysis features associated with the processing time period; S242, each time a processing period is executed, it is extended one time backward based on the position of the initial time of the previous processing period, and the same processing method of the processing period is used to confirm the analysis features associated with each different processing period, and it stops when the initial time and the end time of the corresponding processing period are adjacent times (no need to confirm the analysis features at the adjacent times), and the processing process of the first analysis process is completed; S243, moving the end time forward by one time, and then using the same processing method as the first analysis process (steps S241 and S242), moving the initial time, and confirming the analysis features associated with each different processing period; S244, repeating step S243 until the initial moment and the terminal moment before the corresponding analysis process are adjacent moments (that is, the moments are adjacent, so relevant processing is required, and the numerical features of the processing are small and have no processing significance); Example: Assuming there are five moments t1-t5, the confirmed change characteristics are: main segment {1, 1, 1.5, 3}, secondary segment {2, 3, 4, 5}, the first processing period is t1-t4, then the associated change characteristics are the main segment {1, 1, 1.5} and the secondary segment {2, 3, 4}, the second processing period is t2-t4, the associated change characteristics are the main segment {1, 1} and the secondary segment {2, 3}, the third processing period is t2-t3, then the analysis process is completed, and the adjacent periods of t2-t3 do not need to be processed.

[0011] S25, selecting a minimum value from the confirmed groups of analysis features, and recording the processing period associated with the minimum value as the standard period, recording the curve segment associated with the standard period in the bandwidth fluctuation curve as the main standard feature segment, and recording the curve segment associated with the standard period in the frame rate fluctuation curve as the secondary standard feature segment; S26, performing ratio processing on the change characteristics associated with the primary standard characteristic segment and the secondary standard characteristic segment in the same period, confirming the characteristic ratio, wherein the characteristic ratio = the change characteristics in the primary characteristic segment ÷ the change characteristics in the secondary characteristic segment (all belonging to the same period), performing average processing on several groups of characteristic ratios confirmed in the standard period, and locking the characteristic variation value; Specifically, after the standard time period is confirmed, the corresponding curve segment can be selected from the corresponding bandwidth fluctuation curve and frame rate fluctuation curve, and then the change feature ratio of the same period can be identified from the two groups of curve segments, and then the average is confirmed based on the corresponding change feature ratio to complete the confirmation of the corresponding feature variable value. For example, the confirmed standard time period is t1-t3, the feature in the main segment is {1, 1}, and the feature in the secondary segment is {2, 3}. The feature ratio is: 0.5 (1÷2) and 0.33 (1÷3). Then 0.5 and 0.33 are averaged to confirm the required feature variable value. Step 3: According to the confirmed characteristic variable value, during the process of advertising delivery in this advertising position, the frame rate of the advertisement is adjusted in real time based on the fluctuation characteristics of the network bandwidth of the advertising position to ensure the normal operation of the advertising position and the normal output of this advertisement. The specific sub-steps of real-time adjustment are: The network bandwidth during the ad delivery process is monitored in real time. When the monitored network bandwidth fluctuates by more than 2Mbps compared to the value at the previous moment, the adjustment process is executed: Confirm the fluctuation value of network bandwidth between adjacent moments. The fluctuation value = network bandwidth at the next moment - network bandwidth at the previous moment. The next moment and the previous moment are adjacent moments. Based on the determined fluctuation value, confirm the frame rate value that needs to be adjusted, and its frame rate value = fluctuation value ÷ characteristic variable value. According to the confirmed frame rate value, adjust the frame rate based on the original frame rate of the advertisement. If the frame rate value is positive, adjust it upward based on the original frame rate. If the frame rate value is negative, adjust it downward based on the original frame rate.

[0012] Second embodiment An intelligent advertising placement system, comprising: The frame rate confirmation end obtains the ad slot operation parameters associated with each frame rate change stage based on the pre-regulated frame rate change stage, and locks the optimal frame rate that needs to be regulated for the specified ad slot based on the operation range and specific change characteristics associated with the operation items corresponding to the ad slot; The characteristic variable value confirmation end records other advertisements with the same frame rate as the best frame rate as characteristic advertisements based on the confirmed best frame rate and other advertisements previously placed in this advertisement position, and then locks the standard characteristic segment based on the characteristic advertisement bandwidth and frame rate variation characteristics, and then determines the characteristic variable value from the standard characteristic segment; The frame rate delivery control end, according to the confirmed characteristic variable value, adjusts the frame rate of the advertisement in real time during the advertisement delivery process in this advertisement position based on the fluctuation characteristics of the network bandwidth of the advertisement position.

[0013] 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.

[0014] 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 method for intelligently placing advertisements, characterized in that: The following steps are involved: Step 1: pre-regulate the frame rate of the advertisement to be placed in the specified advertisement slot, obtain the advertisement slot operation parameters associated with each frame rate change stage based on the pre-regulated frame rate change stage, and lock the optimal frame rate that needs to be regulated for the specified advertisement slot based on the operation interval associated with the operation item corresponding to the advertisement slot and the specific change characteristics; Step 2: Based on the confirmed best frame rate and other advertisements that were previously placed in this ad slot, other advertisements with the same frame rate as the best frame rate are recorded as characteristic advertisements, and then based on the characteristic advertisement bandwidth and frame rate variation characteristics, the standard characteristic segment is locked, and then the characteristic variable value is determined from the standard characteristic segment; Step 3: According to the confirmed characteristic variable value, during the process of advertising delivery in this advertising position, the frame rate of the advertisement is adjusted in real time based on the fluctuation characteristics of the network bandwidth of the advertising position.

2. The method for intelligently placing advertisements according to claim 1, characterized in that: In step 1, the specific sub-steps of locking the optimal frame rate are: S11. Based on the advertisement and the determined advertisement position, the advertisement position is made to conduct a trial run for the advertisement. During the trial run, the frame rate of the advertisement is gradually adjusted from low to high, and the minimum and maximum frame rates are both characteristic values ​​preset for the advertisement. The frame rate is increased by 1 Hz for a period of 3 seconds during the playback phase, and the advertisement position operation parameters associated with the end of the 3-second playback phase are confirmed, and different operation parameters of the advertisement positions associated with different increase processes are recorded as different process parameters. S12, confirming the operation item associated with the operation parameter of the corresponding ad slot, and then confirming the operation interval associated with the operation item, wherein the operation interval is a preset interval, and then confirming the process characteristics of different improvement processes based on the confirmed process parameters; S13. Based on different process characteristics associated with different improved processes, the improved process that satisfies the process characteristic ≤ Y1 is calibrated as the standard process, otherwise no calibration is performed, where Y1 is a preset value, and based on different frame rates associated with different standard processes, the frame rate with the maximum value is selected as the optimal frame rate.

3. The method for intelligently placing advertisements according to claim 2, characterized in that: In step S12, the specific method of confirming the process characteristics of different improvement processes is: S121, for a single group of improvement processes, comparing different process parameters with the operation range of the corresponding operation items, if all process parameters belong to the operation range, then the improvement process is marked as the process to be selected, otherwise, no marking is performed; S122, based on the calibrated process to be selected, identify the location of the corresponding process parameter in the operation interval, and divide each operation interval into ten equal parts, and calibrate the equal division ratio of the front end of the process parameter as QB i-k , mark the equal division ratio of the process parameter backend as HB i-k , where i represents different processes to be selected and k represents different process parameters; S123, multiple groups of QB associated with different process parameters in the same process to be selected i-k Perform variance processing, confirm the front-end variance, and then associate multiple groups of HBs associated with different process parameters i-k Variance processing is performed to confirm the back-end variance, and then the confirmed front-end variance and back-end variance are averaged, and the average obtained by processing is used as the process feature associated with this improvement process.

4. The method for intelligently placing advertisements according to claim 1, characterized in that: In the step 2, the specific method of determining the characteristic variable value from the standard characteristic segment is: S21, using the confirmed best frame rate as a benchmark, identifying other advertisements with the same frame rate as the benchmark from the historical delivery data of the advertisement position, and recording the identified other advertisements as feature advertisements; S22, confirming the bandwidth fluctuation period associated with the characteristic advertisement during the playback process: when the bandwidth change value associated with the adjacent moments exceeds 2Mbps, the adjacent moments are recorded into the bandwidth fluctuation period, and the bandwidth fluctuation curve associated with the bandwidth fluctuation period is confirmed, and then the frame rate fluctuation curve associated with the bandwidth fluctuation period is confirmed from the playback process; Select the same-frequency characteristic segment from the confirmed bandwidth fluctuation curve and frame rate fluctuation curve: identify the partial curve segments with the same trend in the same period from the two groups of curves, and mark the identified partial curve segments as the same-frequency characteristic segment; Its trends include: synchronous rise and synchronous fall; S23, confirm the initial moment and the end moment of the same-frequency feature segment, record the same-frequency feature segment belonging to the bandwidth fluctuation curve as the main segment, record the same-frequency feature segment belonging to the frame rate fluctuation curve as the secondary segment, confirm the change characteristics of adjacent moments in the main segment and the secondary segment: mark the numerical feature associated with the moment before the adjacent moment as Tz1, mark the numerical feature associated with the moment after the adjacent moment as Tz2, and use: (Tz2-Tz1) ÷ adjacent moment duration = change characteristics; S24, executing a plurality of analysis processes, wherein the time characteristics associated with each analysis process are different; S25, selecting a minimum value from the plurality of groups of analysis features confirmed by the plurality of analysis processes, and recording the processing period associated with the minimum value as the standard period, recording the curve segment associated with the standard period in the bandwidth fluctuation curve as the primary standard feature segment, and recording the curve segment associated with the standard period in the frame rate fluctuation curve as the secondary standard feature segment; S26. Perform ratio processing on the change characteristics associated with the main standard feature segment and the secondary standard feature segment in the same period to confirm the feature ratio, where the feature ratio = the change characteristics within the main feature segment ÷ the change characteristics within the secondary feature segment. Perform average processing on several groups of feature ratios confirmed within the standard time period to lock the feature change value.

5. The method for intelligently placing advertisements according to claim 4, characterized in that: In step S24, the specific method of executing several analysis processes is as follows: S241, keeping the end time unchanged, extending the initial time back by one time, recording the extended time period as the processing time period, performing standard deviation processing on several groups of change features associated with the processing time period, and confirming the analysis features associated with the processing time period; S242, each time a processing period is executed, it is extended one time backward based on the position of the initial time of the previous processing period, and the same processing method of the processing period is used to confirm the analysis features associated with each different processing period, and the processing is stopped when the initial time and the end time of the corresponding processing period are adjacent time, and the processing process of the first analysis process is completed; S243, moving the end time forward by one time, and then using the same processing method as steps S241-S242 to move the initial time, and confirming the analysis features associated with each different processing period; S244, repeat step S243 until the initial moment and the final moment before the corresponding analysis process are processed are adjacent moments.

6. The method for intelligently placing advertisements according to claim 1, characterized in that: In step 3, the specific method of adjusting the frame rate of the advertisement in real time is: The network bandwidth during the ad delivery process is monitored in real time. When the monitored network bandwidth fluctuates by more than 2Mbps compared to the value at the previous moment, the adjustment process is executed: Confirm the fluctuation value of network bandwidth between adjacent moments, where the fluctuation value = network bandwidth at the next moment - network bandwidth at the previous moment; Based on the determined fluctuation value, the frame rate value that needs to be adjusted is confirmed, and its frame rate value = fluctuation value ÷ characteristic variable value. According to the confirmed frame rate value, the frame rate is adjusted based on the original frame rate of the advertisement.

7. The method for intelligently placing advertisements according to claim 6, characterized in that: The latter moment and the previous moment are adjacent moments. If the frame rate value is a positive value, the frame rate is adjusted upward based on the original frame rate. If the frame rate value is a negative value, the frame rate is adjusted downward based on the original frame rate.

8. An intelligent advertising placement system, the placement system operates according to the intelligent advertising placement method according to any one of claims 1 to 7, characterized in that: include: The frame rate confirmation end obtains the ad slot operation parameters associated with each frame rate change stage based on the pre-regulated frame rate change stage, and locks the optimal frame rate that needs to be regulated for the specified ad slot based on the operation range and specific change characteristics associated with the operation items corresponding to the ad slot; The characteristic variable value confirmation end records other advertisements with the same frame rate as the best frame rate as characteristic advertisements based on the confirmed best frame rate and other advertisements previously placed in this advertisement position, and then locks the standard characteristic segment based on the characteristic advertisement bandwidth and frame rate variation characteristics, and then determines the characteristic variable value from the standard characteristic segment; The frame rate delivery control end, according to the confirmed characteristic variable value, adjusts the frame rate of the advertisement in real time during the advertisement delivery process in this advertisement position based on the fluctuation characteristics of the network bandwidth of the advertisement position.

Images