Short video delivery effect analysis method and system based on big data, and storage medium
By obtaining the portrait information of users and videos, dividing user types and calculating the delivery reference value, the problem of insufficient classification analysis of target short video users in the prior art is solved, and the accuracy and intelligence of video delivery effect analysis is improved.
Patent Information
- Application Number
- CN202411227896.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-03
- Publication Date
- 2024-11-15
AI Technical Summary
The existing video delivery effect analysis methods lack classified analysis of target short video viewing users, resulting in a great impact on feedback information from non-target users and insufficient accuracy and reliability of analysis results.
By obtaining the image information of users and videos, dividing user types, calculating and assigning different values to different types of users, using specific calculation formulas to output and place reference values, reducing the influence of non-target users and improving the accuracy of analysis results.
The effective division of target users and non-target users is achieved, the accuracy and intelligence of video delivery effect analysis is improved, and the impact of non-target user interaction information on the analysis results is significantly reduced.
Smart Images

Figure CN118967230A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of video delivery effect analysis, and specifically to a short video delivery effect analysis method, system and storage medium based on big data. Background Art
[0002] Short videos, also known as short film videos, are a way of disseminating Internet video content. They are generally disseminated on new Internet media platforms, suitable for watching on the move and in short leisure states, with a duration ranging from a few seconds to a few minutes but generally controlled within three minutes, and are pushed at a high frequency. Due to the short content, they can be a separate film or a series of columns.
[0003] Existing improvements for video delivery effect analysis usually conduct a comprehensive analysis of the video's dissemination volume and user interaction data, thereby improving the accuracy of the effect analysis. For example, in a Chinese patent with invention publication number CN113327140A, a video advertising delivery effect intelligent analysis management system based on big data analysis is disclosed. This solution improves the accuracy of the analysis results corresponding to the video advertising delivery effect by obtaining and analyzing the dissemination effect, viewing interest effect and comment effect of video advertisements. When some short videos are delivered, the characteristics of the users who want to deliver them will be determined. For short videos with delivery targets, the interaction information of ordinary users and target users on the short videos has different weights in the delivery effect analysis. However, in the existing methods, there is a lack of a method for classifying and analyzing the viewing users of the targeted short videos, resulting in a greater impact of feedback information from non-target users on the analysis results, and the analysis results are not accurate enough. In view of this, it is necessary to improve the existing video delivery effect analysis. Summary of the invention
[0004] The present invention aims to solve at least one of the technical problems in the prior art to a certain extent. By improving the analysis of video delivery effects, it is used to solve the problem in the prior art that the feedback information of non-target users has a greater impact on the analysis results, resulting in insufficient accuracy and reliability of the analysis results due to the lack of a method for classifying and analyzing users who watch targeted short videos.
[0005] To achieve the above objectives, in a first aspect, the present invention provides a short video delivery effect analysis method based on big data, comprising:
[0006] Obtain user portraits and viewing information of a first number of analyzed users, and obtain target portraits and expected delivery information of analyzed videos;
[0007] Analyze user portraits based on target portraits, divide analyzed users into different types based on the analysis results;
[0008] Calculate the viewing rate based on the viewing information, calculate different user types based on the viewing rate and the viewing information, and output a delivery reference value for each user type;
[0009] Calculate the delivery reference value and output the delivery effective value; calculate and analyze the delivery effective value based on the expected delivery information, and output the delivery reference information based on the calculation and analysis results.
[0010] Further, obtaining user portraits and viewing information of the first number of analyzed users, and obtaining target portraits and expected delivery information of the analyzed videos include:
[0011] Mark the short videos that need to be analyzed as analysis videos, monitor the user's historical viewing records, and if the historical viewing records contain analysis videos, mark the user as an analysis user;
[0012] Acquire a first number of analysis users and viewing information of the analysis users for the analysis video, wherein the viewing information includes viewing time and interaction information;
[0013] Obtain a user profile of each analyzed user, wherein the user profile includes the user's age and the user's gender;
[0014] The target portrait and expected delivery information of the analyzed video are obtained, wherein the target portrait includes a target age range and a target gender; the expected delivery information includes an expected viewing rate and an expected delivery value.
[0015] Furthermore, the user types include common users and target users. The user portrait is analyzed based on the target portrait, and the analyzed users are divided based on the analysis results. Different user types are output, including:
[0016] Analyze the user portrait to determine the relationship between the user age and the target age range. If the target age range contains the user age, perform gender determination. If the target age range does not contain the user age, mark the analyzed user corresponding to the user portrait as a common user.
[0017] The gender judgment process includes: judging whether the user gender in the user portrait is the same as the target gender. If they are the same, marking the analysis user corresponding to the user portrait as the target user; if they are not the same, marking the analysis user corresponding to the user portrait as an ordinary user.
[0018] Furthermore, the viewing rate includes a target viewing rate and a normal viewing rate. The viewing rate is calculated based on the viewing information. Different user types are calculated based on the viewing rate and the viewing information. Outputting the delivery reference value for each user type includes:
[0019] Get the video duration of the analyzed video, calculate the ratio of each target user's viewing duration to the video duration, and mark it as the target viewing rate;
[0020] Calculate the ratio of each common user's viewing time to the video duration, and mark it as the common viewing rate;
[0021] Performing a first effect calculation process on the target user, and outputting a delivery reference value for the target user based on the processing result of the first effect calculation;
[0022] A second effect calculation process is performed on ordinary users, and a delivery reference value for ordinary users is output based on the processing result of the second effect calculation.
[0023] Furthermore, the first effect calculation includes:
[0024] Obtaining interactive information of each target user on the analyzed video, wherein the interactive information includes feedback information and comment information; the feedback information includes positive feedback information, negative feedback information and empty feedback information, and the comment information includes valid comments and empty comments;
[0025] Assign 1 to positive feedback information, -1 to negative feedback information, and 0 to null feedback information;
[0026] Assign 0 to empty comments and assign the same value as the feedback information in the interaction information to valid comments;
[0027] Calculate the assignment of the interactive information, the target viewing rate, and the expected viewing rate using the first assignment calculation formula, and output a delivery reference value for the target user;
[0028] The first assignment calculation formula is configured as: BU is the target user's delivery reference value, F is the value assigned to the feedback information, P is the value assigned to the comment information, BK is the target viewing rate, WK is the expected viewing rate, The round-up symbol.
[0029] Furthermore, the second effect calculation includes:
[0030] Get the interactive information of each ordinary user on the analyzed video, assign 1 to positive feedback information, assign 0 to negative feedback information and empty feedback information; assign 0 to comment information;
[0031] Calculate the assignment of the interactive information, the target viewing rate, and the expected viewing rate using the second assignment calculation formula, and output a delivery reference value for ordinary users;
[0032] The second assignment calculation formula is configured as: TU is the reference value for ordinary users, k is a constant, and TK is the normal viewing rate.
[0033] Further, the delivery reference information includes delivery high efficiency information, the delivery reference value is calculated, and the delivery effective value is output; the delivery effective value is calculated and analyzed based on the expected delivery information, and the delivery reference information is output based on the calculation and analysis results, including:
[0034] Calculate the sum of all users' delivery reference values and mark it as delivery effective value;
[0035] Compare the effective value of the delivery with the expected value of the delivery. If the effective value of the delivery is greater than or equal to the expected value of the delivery, output the delivery high efficiency information; if the effective value of the delivery is less than the expected value of the delivery, perform video analysis processing, and output the delivery reference information based on the video analysis processing.
[0036] Furthermore, the delivery reference information also includes video problem information and delivery problem information, and the video analysis process includes:
[0037] Marking a target viewing rate that is less than the expected viewing rate as an invalid viewing rate, and counting the invalid viewing rate to obtain an invalid number;
[0038] Calculate the ratio of the invalid number to the total number of target views, and mark it as invalid ratio;
[0039] When the invalid ratio is greater than or equal to the ratio threshold, the video problem information is output;
[0040] When the invalid ratio is less than the ratio threshold, the delivery problem information is output.
[0041] In a second aspect, the present invention also provides a short video delivery effect analysis system based on big data, including a user processing module, an effective calculation module and a delivery analysis module;
[0042] The user processing module is used to obtain user portraits and viewing information of a first number of analyzed users, and obtain target portraits and expected delivery information of analyzed videos;
[0043] The user processing module is also used to analyze the user portrait based on the target portrait, divide the analyzed users into different types based on the analysis results, and output different types of users;
[0044] The effective calculation module is used to calculate the viewing rate based on the viewing information, calculate different user types based on the viewing rate and the viewing information, and output a delivery reference value for each user type;
[0045] The delivery analysis module is used to calculate the delivery reference value and output the delivery effective value; calculate and analyze the delivery effective value based on the expected delivery information, and output the delivery reference information based on the calculation and analysis results.
[0046] In a third aspect, the present application provides a storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the method described above are executed.
[0047] Beneficial effects of the present invention: The present invention obtains the user portrait of the analyzed user and the target portrait of the analyzed video, and then analyzes the user portrait based on the target portrait to divide the analyzed user into target users and ordinary users; the advantage of this is that it can realize the division of target users and non-target users for short videos with target delivery groups, thereby improving the accuracy of subsequent effect analysis results;
[0048] The present invention also assigns different values to the interaction information of different types of users, and then uses different assignment calculation formulas to calculate the assignments, and outputs the delivery reference value of each user; finally, the delivery reference value is calculated and analyzed, and the delivery reference information is output based on the calculation and analysis results; the advantage of this is that different assignment rules are set for target users and ordinary users in the assignment stage, and a reduction coefficient k is set for ordinary users in the calculation formula, which can significantly reduce the influence of the interaction information of ordinary users on the final delivery effective value, and improve the intelligence and accuracy of the delivery effect analysis. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] Figure 1 is a functional block diagram of the system of the present invention;
[0050] Figure 2 is a flow chart of the steps of the method of the present invention;
[0051] Figure 3 It is a data flow diagram of the present invention. DETAILED DESCRIPTION
[0052] 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.
[0053] Example 1, first aspect, please refer to Figure 1 as well as Figure 3 As shown, the present invention provides a short video delivery effect analysis system based on big data, including a user processing module, an effective calculation module and a delivery analysis module;
[0054] The user processing module is used to obtain user portraits and viewing information of the first number of analyzed users, and obtain the target portrait and expected delivery information of the analyzed video; the user processing module is also used to analyze the user portrait based on the target portrait, divide the analyzed users based on the analysis results, and output different user types;
[0055] The user processing module is configured with a data acquisition strategy, which includes:
[0056] Mark the short videos that need to be analyzed as analysis videos, monitor the user's historical viewing records, and if the historical viewing records contain analysis videos, mark the user as an analysis user;
[0057] It should be noted that the user who is being monitored is the user who authorizes the short video platform to obtain his or her viewing history so as to make personalized recommendations; monitoring the user's historical viewing history means querying the user's historical viewing history with the title of the analyzed video as the index, and the query results are either empty or analyzed video, and the user's other viewing history cannot be queried; in specific implementation, the user's historical viewing history can be monitored on a daily basis, which can reduce the resources required for monitoring;
[0058] Obtaining a first number of analysis users and viewing information of the analysis users for the analysis video, where the viewing information includes viewing time and interaction information;
[0059] In specific implementation, the first number is set to 1000. The larger the first number is set, that is, the more samples there are, the more accurate the analysis results are. However, too many samples will increase the amount of calculation of the system during analysis. Considering both the calculation efficiency and the accuracy, the preferred value can be set to 1000. The specific value can still be adjusted during implementation.
[0060] Get the user profile of each analyzed user, including the user's age and gender;
[0061] Obtain the target profile and expected delivery information of the analyzed video. The target profile includes the target age range and target gender; the expected delivery information includes the expected viewing rate and the expected delivery value.
[0062] The user processing module is also configured with a user classification strategy, which includes:
[0063] Analyze the user portrait to determine the relationship between the user age and the target age range. If the target age range contains the user age, perform gender determination. If the target age range does not contain the user age, mark the analyzed user corresponding to the user portrait as a common user.
[0064] The gender determination process includes: determining whether the gender of the user in the user portrait is the same as the target gender. If they are the same, the analysis user corresponding to the user portrait is marked as the target user; if they are not the same, the analysis user corresponding to the user portrait is marked as a common user;
[0065] It should be noted that, taking the short video content of introducing young men's clothes as an example, the target age range of the short video may be [15, 30], and the target gender is male; similarly, if the video content is introducing young women's clothes, the target age range may be [15, 30], and the target gender is female;
[0066] In particular, for some short videos introducing products or other things, such as videos introducing children's tutoring materials, the target age range may be [20, 50], and there is no clear target gender. In this case, all genders are the target gender;
[0067] In particular, for videos without a target age range and target gender, the target age range is set to the minimum age to the maximum age that can be set by the video platform, and all genders are set to the target gender, thereby improving the applicability of the system;
[0068] It should be noted that the target age ranges mentioned above are all possible situations, and the specific values must be based on the ranges set by the video publisher during implementation.
[0069] The effective calculation module is used to calculate the viewing rate based on the viewing information, calculate different user types based on the viewing rate and the viewing information, and output the delivery reference value for each user type;
[0070] The effective calculation module configures the effective calculation strategy, which includes:
[0071] Get the video duration of the analyzed video, calculate the ratio of each target user's viewing duration to the video duration, and mark it as the target viewing rate;
[0072] Calculate the ratio of each common user's viewing time to the video duration, and mark it as the common viewing rate;
[0073] Performing a first effect calculation process on the target user, and outputting a delivery reference value for the target user based on the processing result of the first effect calculation; the first effect calculation includes:
[0074] Obtain the interaction information of each target user on the analyzed video, the interaction information includes feedback information and comment information; the feedback information includes positive feedback information, negative feedback information and empty feedback information, and the comment information includes valid comments and empty comments;
[0075] Assign 1 to positive feedback information, -1 to negative feedback information, and 0 to null feedback information;
[0076] It should be noted that positive feedback information indicates that the user likes and recommends the video, and negative feedback information indicates that the user is not interested in or does not like the video. Based on this, we can exclude the situation where a very small number of users like the video but post negative comments. If the user likes and recommends the video, he will post a comment, and the comment content is very likely to be positive content, and vice versa. When the user does not give feedback on the video, if the user posts a comment, it is necessary to use a large language model to analyze the user's comment, and then assign a value to the comment information based on the analysis results, which requires too many resources for the analysis. Therefore, when the user feedback information is empty but the comment information is not empty, the comment information is not analyzed, that is, the value is assigned to 0. In summary, in the subsequent steps, the valid comments can be assigned the same value as the feedback information.
[0077] Assign 0 to empty comments and assign the same value as the feedback information in the interaction information to valid comments;
[0078] Calculate the assignment of the interactive information, the target viewing rate, and the expected viewing rate using the first assignment calculation formula, and output a delivery reference value for the target user;
[0079] The first assignment calculation formula is configured as: BU is the target user's delivery reference value, F is the value assigned to the feedback information, P is the value assigned to the comment information, BK is the target viewing rate, WK is the expected viewing rate, is the rounding symbol;
[0080] It should be noted that the expected viewing rate is the proportion of users who are expected to watch the video to the end, as set by the video publisher. For example, if the length of a short video is 3 minutes, the expected viewing rate is 50%, which means that the video publisher hopes that users will watch the short video for at least 1.5 minutes so that they can fully understand the content of the short video. If the user submits feedback information before watching 1.5 minutes, it can be judged that the user does not fully understand the content of the video. Compared with the information submitted by users with a viewing rate greater than 50%, the information submitted by users with a viewing rate less than 50% is less meaningful for reference.
[0081] In specific implementation, if the target viewing rate BK is less than the expected viewing rate WK, then BK-WK is a negative number, which is rounded up to 0. The first assignment formula can filter the information submitted by the target user whose target viewing rate is less than the expected viewing rate. When BK is greater than WK, BK-WK is rounded up to 1, and the delivery reference value is equal to the sum of the feedback information and the comment information assignment plus 1. The significance of adding 1 is to indicate that a valid playback volume is obtained.
[0082] Performing a second effect calculation process on ordinary users, and outputting a delivery reference value for ordinary users based on the processing result of the second effect calculation; the second effect calculation includes:
[0083] Get the interactive information of each ordinary user on the analyzed video, assign 1 to positive feedback information, assign 0 to negative feedback information and empty feedback information; assign 0 to comment information;
[0084] It should be noted that for ordinary users, due to the video platform recommendation algorithm, the analyzed video may be pushed to ordinary users. Regardless of whether the ordinary users like or dislike the analyzed video, the negative feedback information and comment information submitted do not have sufficient analytical significance. Therefore, other feedback information except positive feedback information and all comment information are assigned a value of 0;
[0085] Calculate the assignment of the interactive information, the target viewing rate, and the expected viewing rate using the second assignment calculation formula, and output a delivery reference value for ordinary users;
[0086] The second assignment calculation formula is configured as: Where TU is the reference value for ordinary users, k is a constant, and TK is the ordinary viewing rate;
[0087] In the specific implementation, k is set to 0.5. The purpose of k is to reduce the impact of ordinary users' interactive information on the video delivery effect analysis, and the reduction effect is 2 times.
[0088] The delivery analysis module is used to calculate the delivery reference value and output the delivery effective value; calculate and analyze the delivery effective value based on the expected delivery information, and output the delivery reference information based on the calculation and analysis results;
[0089] The delivery analysis module is configured with delivery analysis strategies, which include:
[0090] Calculate the sum of all users' delivery reference values and mark it as delivery effective value;
[0091] Compare the effective value of the delivery with the expected value of the delivery. If the effective value of the delivery is greater than or equal to the expected value of the delivery, output the delivery high efficiency information; if the effective value of the delivery is less than the expected value of the delivery, perform video analysis processing, and output the delivery reference information based on the video analysis processing;
[0092] It should be noted that the delivery expectation value is the value obtained through data visualization of the effect that the video publisher hopes to achieve; for example, if the delivery expectation value is set to 30,000, it can provide effective playback volume for 30,000 target users or provide effective playback volume positive feedback information and positive and effective comments for 10,000 target users; the setting of the delivery expectation value needs to be comprehensively formulated based on the delivery cost, video quality, historical playback data, fan base, etc. of the video publisher;
[0093] Video analytics processing includes:
[0094] Marking a target viewing rate that is less than the expected viewing rate as an invalid viewing rate, and counting the invalid viewing rate to obtain an invalid number;
[0095] Calculate the ratio of the invalid number to the total number of target views, and mark it as invalid ratio;
[0096] When the invalid ratio is greater than or equal to the ratio threshold, the video problem information is output;
[0097] When the invalid ratio is less than the ratio threshold, output the delivery problem information;
[0098] In specific implementation, the ratio threshold is set to 30%. When the invalid ratio is greater than 30%, it means that more than 30% of the target users watch the analysis video briefly and then quit watching. It is inferred that the video quality may be insufficient, resulting in insufficient appeal of the video content to users, and video problem information is output; and when the invalid ratio is less than the ratio threshold, it means that most of the target users who watch the analysis video provide effective playback volume, that is, 1 expected value of delivery; at this time, the reason why the effective value of delivery is less than the expected value of delivery may be insufficient cost of publicity and promotion, resulting in not enough target users being pushed the analysis video, and delivery problem information is output at this time.
[0099] Example 2, second aspect, please refer to Figure 2 As shown, the present invention also provides a short video delivery effect analysis method based on big data, including:
[0100] Step S1, obtaining user portraits and viewing information of a first number of analyzed users, and obtaining target portraits and expected delivery information of analyzed videos; Step S1 also includes the following sub-steps:
[0101] Step S101, marking the short video that needs to be analyzed as an analysis video, monitoring the user's historical viewing record, and marking the user as an analysis user if the historical viewing record contains the analysis video;
[0102] Step S102, obtaining a first number of analysis users and viewing information of the analysis users for the analysis video, where the viewing information includes viewing time and interaction information;
[0103] Step S103, obtaining a user portrait of each analyzed user, the user portrait including the user's age and gender;
[0104] Step S104, obtaining the target portrait and expected delivery information of the analyzed video, wherein the target portrait includes the target age range and the target gender; the expected delivery information includes the expected viewing rate and the expected delivery value.
[0105] Step S2, analyzing the user portrait based on the target portrait, dividing the analyzed users based on the analysis results, and outputting different user types; Step S2 also includes the following sub-steps:
[0106] Step S201, analyzing the user portrait to determine the relationship between the user age and the target age range. If the target age range contains the user age, gender determination is performed; if the target age range does not contain the user age, the analyzed user corresponding to the user portrait is marked as a common user.
[0107] Step S202, gender judgment processing includes: judging whether the user gender in the user portrait is the same as the target gender, if the same, marking the analysis user corresponding to the user portrait as the target user; if not the same, marking the analysis user corresponding to the user portrait as a common user.
[0108] Step S3, calculating the viewing rate based on the viewing information, calculating different user types based on the viewing rate and the viewing information, and outputting a delivery reference value for each user type; Step S3 also includes the following sub-steps:
[0109] Step S301, obtaining the video duration of the analyzed video, calculating the ratio of the viewing duration of each target user to the video duration, and marking it as the target viewing rate;
[0110] Step S302, calculating the ratio of each common user's viewing time to the video duration, and marking it as the common viewing rate;
[0111] Step S303, performing a first effect calculation process on the target user, and outputting a delivery reference value for the target user based on the processing result of the first effect calculation;
[0112] Step S303 also includes the following sub-steps:
[0113] Step S3031, obtaining the interaction information of each target user on the analyzed video, the interaction information including feedback information and comment information; the feedback information includes positive feedback information, negative feedback information and empty feedback information, and the comment information includes valid comments and empty comments;
[0114] Step S3032, assigning a value of 1 to positive feedback information, -1 to negative feedback information, and 0 to null feedback information;
[0115] Step S3033, assigning a value of 0 to an empty comment, and assigning a value identical to the feedback information in the interactive information to a valid comment;
[0116] Step S3034, using the first value calculation formula to calculate the value of the interaction information, the target viewing rate, and the expected viewing rate, and output a delivery reference value for the target user;
[0117] The first assignment calculation formula is configured as: BU is the target user's delivery reference value, F is the value assigned to the feedback information, P is the value assigned to the comment information, BK is the target viewing rate, WK is the expected viewing rate, is the rounding symbol;
[0118] Step S304, performing a second effect calculation process on ordinary users, and outputting a delivery reference value for ordinary users based on the processing result of the second effect calculation;
[0119] Step S304 also includes the following sub-steps:
[0120] Step S3041, obtaining the interaction information of each common user on the analyzed video, assigning 1 to positive feedback information, assigning 0 to negative feedback information and empty feedback information; and assigning 0 to comment information;
[0121] Step S3042, using the second value calculation formula to calculate the value of the interaction information, the target viewing rate, and the expected viewing rate, and outputting a delivery reference value for ordinary users;
[0122] The second assignment calculation formula is configured as: TU is the reference value for ordinary users, k is a constant, and TK is the normal viewing rate.
[0123] Step S4, calculating the delivery reference value and outputting the delivery effective value; calculating and analyzing the delivery effective value based on the expected delivery information, and outputting the delivery reference information based on the calculation and analysis results; Step S4 also includes the following sub-steps:
[0124] Step S401, calculating the sum of the delivery reference values of all users and marking it as the delivery effective value;
[0125] Step S402, comparing the effective value of the delivery with the expected value of the delivery, if the effective value of the delivery is greater than or equal to the expected value of the delivery, outputting delivery high efficiency information;
[0126] Step S403: if the effective value of delivery is less than the expected value of delivery, perform video analysis processing, and output delivery reference information based on the video analysis processing;
[0127] Step S403 also includes the following sub-steps:
[0128] Step S4031, marking the target viewing rate that is less than the expected viewing rate as an invalid viewing rate, and counting the invalid viewing rates to obtain an invalid number;
[0129] Step S4032, calculating the ratio of the invalid number to the total number of target views, and marking it as invalid ratio;
[0130] Step S4033, when the invalid ratio is greater than or equal to the ratio threshold, output video problem information; when the invalid ratio is less than the ratio threshold, output delivery problem information.
[0131] Embodiment 3, the present invention also provides an electronic device, which may include: a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus. The memory stores computer-readable instructions, and the processor can call the instructions in the memory. When the computer-readable instructions are executed by the processor, the steps in the short video delivery effect analysis method based on big data are executed to achieve the following functions: obtaining the user portraits and viewing information of the first number of analyzed users, obtaining the target portraits and expected delivery information of the analyzed video; analyzing the user portraits based on the target portraits, dividing the analyzed users based on the analysis results, and outputting different user types; calculating the viewing rate based on the viewing information, calculating different user types based on the viewing rate and the viewing information, and outputting the delivery reference value of each user type; calculating the delivery reference value, and outputting the delivery effective value; calculating and analyzing the delivery effective value based on the expected delivery information, and outputting the delivery reference information based on the calculation and analysis results.
[0132] In addition, the logic instructions in the above-mentioned memory can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk.
[0133] Embodiment 4. The present application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the short video delivery effect analysis method based on big data provided by the above methods, the method including: obtaining user portraits and viewing information of a first number of analyzed users, obtaining a target portrait and expected delivery information of the analyzed video; analyzing the user portrait based on the target portrait, dividing the analyzed users based on the analysis results, and outputting different user types; calculating the viewing rate based on the viewing information, calculating different user types based on the viewing rate and the viewing information, and outputting a delivery reference value for each user type; calculating the delivery reference value, and outputting a delivery effective value; calculating and analyzing the delivery effective value based on the expected delivery information, and outputting the delivery reference information based on the calculation and analysis results.
[0134] Embodiment 5. The present application also provides a computer-readable storage medium. The present application provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the short video delivery effect analysis method based on big data are executed to achieve the following functions: obtain user portraits and viewing information of a first number of analyzed users, obtain a target portrait and expected delivery information of the analyzed video; analyze the user portrait based on the target portrait, divide the analyzed users based on the analysis results, and output different user types; calculate the viewing rate based on the viewing information, calculate different user types based on the viewing rate and the viewing information, and output a delivery reference value for each user type; calculate the delivery reference value and output the delivery effective value; calculate and analyze the delivery effective value based on the expected delivery information, and output the delivery reference information based on the calculation and analysis results.
[0135] Through the description of the above implementation methods, the embodiments of the present invention can be provided as methods, systems or computer program products. Based on such an understanding, the above technical solutions can be essentially or partly contributed to the prior art in the form of software products, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and include several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.
[0136] In the embodiments provided in the present application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of systems, modules and units can be electrical, mechanical or other forms.
[0137] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims
1. A short video delivery effect analysis method based on big data, characterized in that: include: Obtain user portraits and viewing information of a first number of analyzed users, and obtain target portraits and expected delivery information of analyzed videos; Analyze user portraits based on target portraits, divide analyzed users into different types based on the analysis results; Calculate the viewing rate based on the viewing information, calculate different user types based on the viewing rate and the viewing information, and output a delivery reference value for each user type; Calculate the delivery reference value and output the delivery effective value; The effective value of delivery is calculated and analyzed based on the expected delivery information, and delivery reference information is output based on the calculation and analysis results.
2. The short video delivery effect analysis method based on big data according to claim 1 is characterized in that: Obtaining the user portraits and viewing information of the first number of analyzed users, and obtaining the target portrait and expected delivery information of the analyzed videos includes: Mark the short videos that need to be analyzed as analysis videos, monitor the user's historical viewing records, and if the historical viewing records contain analysis videos, mark the user as an analysis user; Acquire a first number of analysis users and viewing information of the analysis users for the analysis video, wherein the viewing information includes viewing time and interaction information; Obtain a user profile of each analyzed user, wherein the user profile includes the user's age and the user's gender; The target portrait and expected delivery information of the analyzed video are obtained, wherein the target portrait includes a target age range and a target gender; the expected delivery information includes an expected viewing rate and an expected delivery value.
3. The short video delivery effect analysis method based on big data according to claim 2 is characterized in that: The user types include common users and target users. The user profile is analyzed based on the target profile, and the analyzed users are divided based on the analysis results. Different user types are output, including: Analyze the user portrait to determine the relationship between the user age and the target age range. If the target age range contains the user age, perform gender determination. If the target age range does not contain the user age, mark the analyzed user corresponding to the user portrait as a common user. The gender judgment process includes: judging whether the user gender in the user portrait is the same as the target gender. If they are the same, marking the analysis user corresponding to the user portrait as the target user; if they are not the same, marking the analysis user corresponding to the user portrait as an ordinary user.
4. The short video delivery effect analysis method based on big data according to claim 3 is characterized in that: The viewing rate includes a target viewing rate and a normal viewing rate. The viewing rate is calculated based on the viewing information. Different user types are calculated based on the viewing rate and the viewing information. The output of the delivery reference value for each user type includes: Get the video duration of the analyzed video, calculate the ratio of each target user's viewing duration to the video duration, and mark it as the target viewing rate; Calculate the ratio of each common user's viewing time to the video duration, and mark it as the common viewing rate; Performing a first effect calculation process on the target user, and outputting a delivery reference value for the target user based on the processing result of the first effect calculation; A second effect calculation process is performed on ordinary users, and a delivery reference value for ordinary users is output based on the processing result of the second effect calculation.
5. The short video delivery effect analysis method based on big data according to claim 4 is characterized in that: The first effect calculation includes: Obtaining interactive information of each target user on the analyzed video, wherein the interactive information includes feedback information and comment information; the feedback information includes positive feedback information, negative feedback information and empty feedback information, and the comment information includes valid comments and empty comments; Assign 1 to positive feedback information, -1 to negative feedback information, and 0 to null feedback information; Assign 0 to empty comments and assign the same value as the feedback information in the interaction information to valid comments; Calculate the assignment of the interactive information, the target viewing rate, and the expected viewing rate using the first assignment calculation formula, and output a delivery reference value for the target user; The first assignment calculation formula is configured as: BU is the target user's delivery reference value, F is the value assigned to the feedback information, P is the value assigned to the comment information, BK is the target viewing rate, WK is the expected viewing rate, The symbol for rounding up.
6. The short video delivery effect analysis method based on big data according to claim 5 is characterized in that: The second effect calculation includes: Get the interactive information of each ordinary user on the analyzed video, assign 1 to positive feedback information, assign 0 to negative feedback information and empty feedback information; assign 0 to comment information; Calculate the assignment of the interactive information, the target viewing rate, and the expected viewing rate using the second assignment calculation formula, and output a delivery reference value for ordinary users; The second assignment calculation formula is configured as: TU is the reference value for ordinary users, k is a constant, and TK is the normal viewing rate.
7. The short video delivery effect analysis method based on big data according to claim 6 is characterized in that: The delivery reference information includes delivery high efficiency information, calculates the delivery reference value, and outputs the delivery effective value; The effective value of delivery is calculated and analyzed based on the expected delivery information, and the delivery reference information is output based on the calculation and analysis results, including: Calculate the sum of all users' delivery reference values and mark it as delivery effective value; Compare the effective value of the delivery with the expected value of the delivery. If the effective value of the delivery is greater than or equal to the expected value of the delivery, output the delivery high efficiency information; if the effective value of the delivery is less than the expected value of the delivery, perform video analysis processing, and output the delivery reference information based on the video analysis processing.
8. The short video delivery effect analysis method based on big data according to claim 7 is characterized in that: The delivery reference information also includes video problem information and delivery problem information, and the video analysis process includes: Marking a target viewing rate that is less than the expected viewing rate as an invalid viewing rate, and counting the invalid viewing rate to obtain an invalid number; Calculate the ratio of the invalid number to the total number of target views, and mark it as invalid ratio; When the invalid ratio is greater than or equal to the ratio threshold, the video problem information is output; When the invalid ratio is less than the ratio threshold, the delivery problem information is output.
9. A short video delivery effect analysis system based on big data, used to implement the short video delivery effect analysis method based on big data as described in any one of claims 1 to 8, characterized in that: It includes user processing module, effective calculation module and delivery analysis module; The user processing module is used to obtain user portraits and viewing information of a first number of analyzed users, and obtain target portraits and expected delivery information of analyzed videos; The user processing module is also used to analyze the user portrait based on the target portrait, divide the analyzed users into different types based on the analysis results, and output different types of users; The effective calculation module is used to calculate the viewing rate based on the viewing information, calculate different user types based on the viewing rate and the viewing information, and output a delivery reference value for each user type; The delivery analysis module is used to calculate the delivery reference value and output the delivery effective value; calculate and analyze the delivery effective value based on the expected delivery information, and output the delivery reference information based on the calculation and analysis results.
10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are executed.
Citation Information
Patent Citations
Video advertisement putting effect intelligent analysis and management system based on big data analysis
CN113327140A
