A method and system for accurate live streaming based on big data

Abstract

The present invention relates to the field of accurate traffic diversion, and in particular to a method and system for accurate traffic diversion of live broadcast based on big data, the system comprising: a time determination module, which analyzes the live broadcast time of social media and determines the live broadcast time; a crowd positioning module, which uses information collection technology to search for potential target users to locate preliminary target users; a video push module, which automatically edits the content to be broadcast into several short videos for push and analysis, selects the short video with the best effect for multi-channel push, and determines the final target user; a process arrangement module, which analyzes the viewing record and duration of the final target user, plans the live broadcast duration and arranges the process; a feedback adjustment module, which counts the viewing duration and records the interactive content of the final target user in the live broadcast room, so as to adjust the live broadcast process in the next live broadcast period. The accurate traffic diversion effect of live broadcast is achieved and the live broadcast process is optimized.

Description

Technical Field

[0001] The present invention relates to the field of live broadcast traffic diversion, and in particular to a live broadcast precise traffic diversion method and system based on big data. Background Art

[0002] With the advancement of technology, live streaming e-commerce will continue to explore business models that are deeply integrated with content, provide richer and more attractive live streaming content, and promote innovation and development in the industry.

[0003] The patent document with Chinese patent publication number CN117768665AA discloses a method and system for accurate live broadcast traffic diversion based on Internet big data analysis, including: a data collection module, a data processing module, a user portrait module, a matching algorithm module, a traffic diversion strategy module and a real-time interactive optimization module; wherein, the data collection module is used to collect browsing history, search history and social interaction data about users from major social media platforms, search engines and related websites; the data processing module receives data transmitted from the data collection module and performs preprocessing including denoising, standardization and inductive classification; the user portrait module establishes a personalized portrait of each user based on the preprocessed data; the matching algorithm module uses the user portrait data to recommend live broadcast content that matches the interests of each user; the traffic diversion strategy module recommends live broadcast content that matches the interests of each user based on the matching results; the real-time interactive optimization module adjusts the matching algorithm in real time according to the user's interactive behavior in the live broadcast room, so as to more accurately recommend the next live broadcast content to the user.

[0004] In the prior art, live broadcast content matching the interests of each user is recommended to more accurately recommend the next live broadcast content to the user, but the live broadcast process is not optimized, resulting in inaccurate live broadcast traffic. Summary of the invention

[0005] To this end, the present invention provides a method and system for accurate live broadcast traffic diversion based on big data, which can solve the problem of inaccurate live broadcast traffic diversion.

[0006] To achieve the above purpose, the present invention provides a live broadcast accurate traffic diversion system based on big data, comprising:

[0007] The time determination module uses big data to analyze the live broadcast time of major social media platforms and determines the live broadcast time based on the analysis results;

[0008] A crowd positioning module, connected to the time determination module, searches for the ID location of potential target users based on the determined live broadcast time using information collection technology, and locates preliminary target users according to the ID;

[0009] A video push module is connected to the crowd positioning module, automatically edits the content to be broadcast live into several short videos, pushes them to the preliminary target users and analyzes them, selects the short video with the best effect for multi-channel push according to the analysis results, and determines the final target user according to the multi-channel push results;

[0010] A process arrangement module is connected to the video push module, and analyzes the viewing history and duration of the final target user, obtains the most active viewing period of the historical live broadcast and the retention of the final target user according to the analysis result, and plans the live broadcast duration, and determines the conversion node by analyzing the purchase behavior of the final target user in the historical data, and determines the live broadcast process based on the conversion node and the most active viewing period;

[0011] The feedback adjustment module is connected to the process scheduling module to count the live broadcast time and record the interactive content of the final target users who enter the live broadcast room during the live broadcast time, so as to adjust the live broadcast process in the next live broadcast time period.

[0012] Furthermore, the time determination module includes:

[0013] A statistical unit, used to collect statistics of large-scale live broadcast information of the multiple platforms;

[0014] A time determination unit is connected to the statistical unit and is used to determine the live broadcast time according to the statistical result.

[0015] Furthermore, the crowd positioning module includes:

[0016] Social media analysis unit to analyze and monitor discussions and trends related to live content on social media platforms and identify potential target users;

[0017] The crowd tracking unit is connected to the social media analysis unit, searches for the ID location of potential target users through information collection technology, and locates the preliminary target users according to the ID.

[0018] Furthermore, the video push module includes:

[0019] A video editing unit automatically edits the content to be broadcast live into several short videos;

[0020] A short video push unit, connected to the video editing unit, for publishing the short video on major social media platforms and pushing it to the initial target users;

[0021] A click-through rate tracking unit is connected to the short video push unit, and uses an analysis tool to track key indicators of click-through rate and viewing time of each short video, and selects the short video with the best effect for multi-channel push based on the key indicators;

[0022] The target user adjustment unit is connected to the click rate tracking unit, collects and analyzes the data pushed through multiple channels, and determines the final target user based on the data analysis result.

[0023] Furthermore, the process arrangement module includes:

[0024] A data collection unit, which aggregates the historical viewing data of the final target user;

[0025] a behavior analysis unit connected to the data collection unit, analyzing the behavior pattern of the final target user according to the historical viewing data, and identifying the most active viewing time period of the final target user according to the analysis result;

[0026] A retention rate analysis unit, connected to the behavior analysis unit, evaluates the final target user retention rate under different live broadcast durations in the historical data, and finds out the loss point of the final target user;

[0027] A live broadcast duration planning unit, connected to the behavior analysis unit and the retention rate analysis unit, plans the live broadcast duration based on the most active viewing period of the final target user and the retention rate;

[0028] A conversion effect analysis unit, connected to the live broadcast duration planning unit, analyzes the conversion effect of the live broadcast in the historical data and identifies which links promote the purchase behavior of the final target user;

[0029] The format design unit is connected with the behavior analysis unit and the conversion effect analysis unit, and designs the live broadcast process based on the most active viewing period and the purchasing behavior of the final target user.

[0030] Furthermore, the behavior analysis unit includes:

[0031] A feature extraction subunit, used to extract features of viewing duration and viewing frequency from the historical viewing data;

[0032] The active time period analysis subunit is connected to the feature extraction subunit and is used to count the viewing frequency and viewing duration in different time periods, and obtain the most active viewing time period according to the statistical results.

[0033] Furthermore, the retention rate analysis unit includes:

[0034] The data cleaning subunit is used to clean the collected data and remove invalid or abnormal data;

[0035] A retention rate calculation subunit is connected to the data cleaning subunit, and divides the different live broadcast durations in the cleaned data into several time periods evenly, counts the final target users retained in each time period, and calculates the retention rate of the target users in each time period;

[0036] A retention curve drawing subunit is connected to the retention rate calculation subunit, and draws a retention curve according to the final target users retained in each time period to show the retention status of the final target users under different live broadcast durations;

[0037] The churn point analysis subunit is connected to the retention curve drawing subunit, and identifies the peak value of the final target user churn through the drawn retention curve, and the point corresponding to the peak value is the churn point.

[0038] Furthermore, the conversion effect analysis unit includes:

[0039] Define the conversion indicator sub-unit, and make it clear that the indicator of conversion effect is the user's purchase behavior;

[0040] A key link identification subunit is connected to the definition conversion index subunit, and analyzes each link in the historical live broadcast process to identify which links have a significant impact on the purchase behavior;

[0041] The correlation analysis subunit is connected to the key link identification subunit, and uses statistical methods to analyze the influencing factors of the purchase link, and obtains the reasons for the conversion effect based on the analysis results.

[0042] Furthermore, the feedback adjustment module includes:

[0043] Identify and track the unit, using live broadcast platform analysis tools or third-party services to identify and track the final target users who enter the live broadcast room;

[0044] A viewing time statistics unit, connected to the identification and tracking unit, for recording the viewing time of each final target user, including the time points of entering and leaving the live broadcast room;

[0045] An interactive content recording unit, connected to the viewing time statistics unit, for collecting and recording the interactive behavior of the final target user in the live broadcast room;

[0046] The strategy iteration unit is connected to the interactive content recording unit to aggregate the viewing time and interactive content data of all final target users, perform optimization analysis for the live broadcast content, and continuously iterate the live broadcast strategy based on the analysis results.

[0047] On the other hand, the present invention also provides a method for accurate live streaming based on big data, the method comprising:

[0048] Use big data to analyze the live broadcast time of major social media platforms and determine the live broadcast time based on the analysis results;

[0049] Based on the determined live broadcast time, using information collection technology to search for the ID location of potential target users, and locating preliminary target users according to the ID;

[0050] Shoot a number of short videos for the live broadcast content and push them to the preliminary target users and analyze them. According to the analysis results, select the short video with the best effect for multi-channel push, and determine the final target user according to the multi-channel push results;

[0051] Analyze the viewing history and duration of the final target user, obtain the most active viewing period of the historical live broadcast and the retention of the final target user according to the analysis results, plan the live broadcast duration, determine the conversion node by analyzing the purchase behavior of the final target user in the historical data, and determine the live broadcast process based on the conversion node and the most active viewing period;

[0052] The live broadcast duration and interactive content of the final target users who enter the live broadcast room during the live broadcast time are counted to adjust the live broadcast process in the next live broadcast time period.

[0053] Compared with the prior art, the beneficial effects of the present invention are that the live broadcast time is determined by a time determination module to avoid overlapping with large live broadcasts, thereby improving the dissemination rate of live broadcasts; the initial target users are located according to the crowd positioning module to attract more potential audiences in a targeted manner; a number of short videos are shot for the live broadcast content according to the video push module for push; the short video with the highest click-through rate is selected according to the push results for multi-channel push to determine the final target user, which helps to improve the exposure rate of the live broadcast content and increase the number of new viewers; the viewing history live broadcast records and duration of the final target user are analyzed according to the process arrangement module; the live broadcast duration and the form of the live broadcast process are planned according to the analysis results to avoid being too long or too short, ensuring that the content is compact and attracts the audience; the viewing time and interactive content of users who enter the live broadcast room during the live broadcast period are counted to adjust the live broadcast process in the next live broadcast time period, optimize the live broadcast process, reduce invalid links, and improve the overall efficiency of the live broadcast.

[0054] In particular, by analyzing the large-scale live broadcasts on various social media, the appropriate live broadcast time can be determined to avoid overlap with large-scale live broadcasts, increase the visibility of the live broadcast, and maximize the communication benefits.

[0055] In particular, potential target users are identified through the social media analysis unit, and preliminary target users are located based on the crowd tracking unit, thereby achieving preliminary positioning and providing a basis for precise traffic diversion.

[0056] In particular, by selecting the best short video from several short videos for multi-channel push, it is beneficial to test different short video content, promote innovation in content creation and marketing strategies, expand the initial target users, and help increase the exposure of live broadcast content.

[0057] In particular, the historical viewing data of the final target user is summarized through the data collection unit, and the most active viewing time period is identified based on the analysis of the viewing habits of the final target user, so as to convey key information or conduct promotional activities, and plan the duration and process of the live broadcast more reasonably. By evaluating the retention rate of the final target user at different live broadcast durations in the historical data, the loss point of the final target user is found, and a more personalized viewing experience is provided. According to the retention rate data, the content arrangement is adjusted to ensure that the best content is provided during the most active time period of the audience. According to the analysis of the conversion effect of the live broadcast in the historical data, it is identified which links promote the purchasing behavior of the final target user, and these links can be strengthened in a targeted manner to improve the overall purchase rate, providing a reference for the design of the live broadcast format. Finally, the live broadcast process is designed and the live broadcast script is optimized.

[0058] In particular, the feature extraction subunit extracts the features of viewing time and viewing frequency from the historical viewing data to obtain the most active viewing time period, which provides a reference for the live broadcast process arrangement and helps to arrange the live broadcast or push content at the time when users are most likely to participate, thereby increasing the viewing rate and interaction rate.

[0059] In particular, by calculating the retention rate of the final target users in each time period and drawing the retention curve, finding the loss points, and analyzing the live broadcast situation intuitively through data, it is helpful to accurately find the key time period when the final target users are lost during the live broadcast, and then take corresponding measures to improve user stickiness and loyalty.

[0060] In particular, by analyzing the user's purchasing behavior and the different factors in the purchasing process of the live broadcast room, we can find out the key factors that promote conversion and provide optimization directions for future live broadcast activities.

[0061] In particular, by counting the viewing time and recording the interactive content of the final target users who enter the live broadcast room during the live broadcast period, we can adjust the format of future live broadcasts, continuously iterate the live broadcast strategy, and ensure that the quality of live broadcast content and format continues to improve by optimizing the live broadcast format and content. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 A schematic diagram of the structure of a live broadcast precision traffic diversion system based on big data provided by an embodiment of the present invention;

[0063] Figure 2 A schematic diagram of the structure of a time determination module of a live broadcast precision traffic diversion system based on big data provided by an embodiment of the present invention;

[0064] Figure 3 A schematic diagram of the structure of a crowd positioning module of a live broadcast precision traffic diversion system based on big data provided by an embodiment of the present invention;

[0065] Figure 4 A schematic diagram of the structure of a video push module of a live broadcast precision traffic diversion system based on big data provided by an embodiment of the present invention;

[0066] Figure 5 A schematic diagram of the structure of a process arrangement module of a live broadcast precision traffic diversion system based on big data provided by an embodiment of the present invention;

[0067] Figure 6 A schematic diagram of the structure of the feedback adjustment module of the live broadcast precision traffic diversion system based on big data provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0068] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0069] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0070] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be understood as a limitation on the present invention.

[0071] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0072] See also Figure 1 As shown, an embodiment of the present invention provides a live broadcast accurate flow diversion system based on big data, the system comprising:

[0073] The time determination module 10 uses big data to analyze the live broadcast time of major social media platforms and determines the live broadcast time according to the analysis results;

[0074] The crowd positioning module 20 is connected to the time determination module 10, and uses information collection technology to search for the ID positioning of potential target users based on the determined live broadcast time, and locates the preliminary target users according to the ID;

[0075] The video push module 30 is connected to the crowd positioning module 20, shoots a number of short videos for the live broadcast content, pushes them to the preliminary target users and analyzes them, selects the short video with the best effect for multi-channel push according to the analysis results, and determines the final target user according to the multi-channel push results;

[0076] The process arrangement module 40 is connected to the video push module 30, analyzes the viewing history and duration of the final target user, obtains the most active viewing period of the historical live broadcast and the retention of the final target user according to the analysis result, plans the live broadcast duration, determines the conversion node by analyzing the purchase behavior of the final target user in the historical data, and determines the live broadcast process based on the conversion node and the most active viewing period;

[0077] The feedback adjustment module 50 is connected to the process scheduling module 40, and counts the live broadcast time and records the interactive content of the final target users who enter the live broadcast room during the live broadcast time, so as to adjust the live broadcast process in the next live broadcast time period.

[0078] Specifically, the embodiment of the present invention determines the live broadcast time through a time determination module to avoid overlapping with large-scale live broadcasts, thereby improving the dissemination rate of live broadcasts; locates preliminary target users according to the crowd positioning module to attract more potential audiences in a targeted manner; shoots several short videos for the live broadcast content according to the video push module for push; selects the short video with the highest click-through rate according to the push results for multi-channel push to determine the final target user, which helps to improve the exposure rate of the live broadcast content and increase the number of new viewers; analyzes the viewing history live broadcast records and duration of the final target user according to the process scheduling module, plans the live broadcast duration and the form of the live broadcast process according to the analysis results to avoid being too long or too short, ensures that the content is compact and attracts the audience, and counts the viewing time and interactive content of users who enter the live broadcast room during the live broadcast period to adjust the live broadcast process in the next live broadcast time period, optimize the live broadcast process, reduce invalid links, and improve the overall efficiency of the live broadcast.

[0079] See also Figure 2 As shown, the time determination module 10 includes:

[0080] A statistics unit 11, used for collecting statistics of large-scale live broadcast information of the multi-platforms;

[0081] The time determination unit 12 is connected to the statistical unit 11 and is used to determine the live broadcast time according to the statistical result.

[0082] Specifically, this embodiment determines the appropriate live broadcast time by analyzing the large-scale live broadcasts on various social media, thereby avoiding overlap with large-scale live broadcasts, increasing the visibility of the live broadcasts, and maximizing the communication benefits.

[0083] See also Figure 3 As shown, the crowd positioning module 20 includes:

[0084] Social media analysis unit 21, analyzing and monitoring discussions and trends related to live content on social media platforms to identify potential target users;

[0085] The crowd tracking unit 22 is connected to the social media analysis unit 21, searches for the ID location of potential target users through information collection technology, and locates the preliminary target users according to the ID.

[0086] Specifically, this embodiment identifies potential target users through a social media analysis unit, and locates preliminary target users according to a crowd tracking unit, thereby achieving preliminary positioning and providing a basis for accurate traffic diversion.

[0087] See also Figure 4 As shown, the video push module 30 includes:

[0088] A video editing unit 31 automatically edits the content to be broadcast live into several short videos;

[0089] A short video push unit 32, connected to the video editing unit 31, is used to publish the short video on major social media platforms and push it to the initial target users;

[0090] The click rate tracking unit 33 is connected to the short video pushing unit 32, and uses an analysis tool to track the key indicators of the click rate and viewing time of each short video, and selects the short video with the best effect for multi-channel pushing according to the key indicators;

[0091] The target user adjustment unit 34 is connected to the click rate tracking unit 33, collects and analyzes the data pushed through multiple channels, and determines the final target user according to the data analysis result.

[0092] Specifically, this embodiment selects the best short video from several short videos for multi-channel push, which is conducive to testing different short video contents, promoting the innovation of content creation and marketing strategies, so as to expand the initial target users and help improve the exposure rate of live broadcast content.

[0093] See also Figure 5As shown, the process arrangement module 40 includes:

[0094] A data collection unit 41, which collects historical viewing data of the final target user;

[0095] A behavior analysis unit 42, connected to the data collection unit 41, analyzes the behavior pattern of the final target user according to the historical viewing data, and identifies the most active viewing time period of the final target user according to the analysis result;

[0096] A retention rate analysis unit 43, connected to the behavior analysis unit 42, evaluates the final target user retention rate under different live broadcast durations in the historical data, and finds out the loss point of the final target user;

[0097] A live broadcast duration planning unit 44, connected to the behavior analysis unit 42 and the retention rate analysis unit 43, plans the live broadcast duration based on the most active viewing period of the final target user and the retention rate;

[0098] The conversion effect analysis unit 45 is connected to the live broadcast duration planning unit 44, and analyzes the conversion effect of the live broadcast in the historical data to identify which links promote the purchase behavior of the final target user;

[0099] The format design unit 46 is connected to the behavior analysis unit 42 and the conversion effect analysis unit 45, and designs the live broadcast process based on the most active viewing time period and the purchasing behavior of the final target user.

[0100] Specifically, the live broadcast process described in this embodiment includes: opening, interactive session, and product display.

[0101] Specifically, this embodiment summarizes the historical viewing data of the final target user through a data collection unit and identifies the most active viewing time period based on the analysis of the viewing habits of the final target user, conducts key information communication or promotional activities, and more reasonably plans the duration and process of the live broadcast. By evaluating the retention rate of the final target user at different live broadcast durations in the historical data, the loss points of the final target user are found, and a more personalized viewing experience is provided. According to the retention rate data, the content arrangement is adjusted to ensure that the best content is provided during the most active time period of the audience. According to the analysis of the conversion effect of the live broadcast in the historical data, it is identified which links promote the purchasing behavior of the final target user, and these links can be strengthened in a targeted manner to improve the overall purchase rate, providing a reference for the design of the live broadcast format. Finally, the live broadcast process is designed and the live broadcast script is optimized.

[0102] Specifically, the behavior analysis unit includes:

[0103] A feature extraction subunit, used to extract features of viewing duration and viewing frequency from the historical viewing data;

[0104] The active time period analysis subunit is connected to the feature extraction subunit and is used to count the viewing frequency and viewing duration in different time periods, and obtain the most active viewing time period according to the statistical results.

[0105] Specifically, this embodiment uses a feature extraction subunit to extract features of viewing duration and viewing frequency from historical viewing data to obtain the most active viewing time period, which provides a reference for arranging the live broadcast process, helps to arrange live broadcasts or push content at times when users are most likely to participate, thereby increasing viewing rates and interaction rates.

[0106] Specifically, the retention rate analysis unit includes:

[0107] The data cleaning subunit is used to clean the collected data and remove invalid or abnormal data;

[0108] A retention rate calculation subunit is connected to the data cleaning subunit, and divides the different live broadcast durations in the cleaned data into several time periods evenly, counts the final target users retained in each time period, and calculates the retention rate of the target users in each time period;

[0109] A retention curve drawing subunit is connected to the retention rate calculation subunit, and draws a retention curve according to the final target users retained in each time period to show the retention status of the final target users under different live broadcast durations;

[0110] The churn point analysis subunit is connected to the retention curve drawing subunit, and identifies the peak value of the final target user churn through the drawn retention curve, and the point corresponding to the peak value is the churn point.

[0111] Specifically, the present embodiment Where N i is the final target user retained in the i-th period, and N0 is the initial final target user.

[0112] Specifically, this embodiment calculates the retention rate of the final target users in each time period and draws a retention curve to find the loss points and analyze the live broadcast situation in a digital and intuitive manner, which is conducive to accurately finding the key time period for the loss of the final target users in the live broadcast, and then taking corresponding measures to improve user stickiness and loyalty.

[0113] Specifically, the conversion effect analysis unit includes:

[0114] Define the conversion indicator sub-unit, and make it clear that the indicator of conversion effect is the user's purchase behavior;

[0115] A key link identification subunit is connected to the definition conversion index subunit, and analyzes each link in the historical live broadcast process to identify which links have a significant impact on the purchase behavior;

[0116] The correlation analysis subunit is connected to the key link identification subunit, and uses statistical methods to analyze the influencing factors of the purchase link, and obtains the reasons for the conversion effect based on the analysis results.

[0117] Specifically, the influencing factors described in this embodiment include live broadcast duration, number of interactions and promotion intensity.

[0118] Specifically, this embodiment analyzes the user's purchasing behavior and different factors in the purchasing process of the live broadcast room to find out the key factors that promote conversion and provide optimization directions for future live broadcast activities.

[0119] See also Figure 6 As shown, the feedback adjustment module 50 includes:

[0120] An identification and tracking unit 51 uses a live broadcast platform analysis tool or a third-party service to identify and track the final target user who enters the live broadcast room;

[0121] A viewing time statistics unit 52, connected to the identification and tracking unit 51, is used to record the viewing time of each final target user, including the time points of entering and leaving the live broadcast room;

[0122] The interactive content recording unit 53 is connected to the viewing time counting unit 52 to collect and record the interactive behavior of the final target user in the live broadcast room;

[0123] The strategy iteration unit 54 is connected to the interactive content recording unit 53 to summarize the viewing time and interactive content data of all final target users, perform optimization analysis for the live broadcast content, and continuously iterate the live broadcast strategy according to the analysis results.

[0124] Specifically, the interactive behaviors described in this embodiment include: commenting, liking, sharing, and gift giving.

[0125] Specifically, this embodiment counts the viewing time and records the interactive content of the final target users entering the live broadcast room to adjust the future live broadcast format, continuously iterate the live broadcast strategy, and optimize the live broadcast format and content to ensure that the quality of the live broadcast content and format continues to improve.

[0126] The method in the embodiment of the present invention is applied in the above-mentioned live broadcast precise traffic diversion system based on big data, and can achieve the same technical effect, which will not be repeated here.

[0127] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0128] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A live broadcast precision traffic diversion system based on big data, characterized in that: include: The time determination module uses big data to analyze the live broadcast time of major social media platforms and determines the live broadcast time based on the analysis results; A crowd positioning module, connected to the time determination module, searches for the ID of potential target users based on the determined live broadcast time using information collection technology, and locates the preliminary target users according to the ID; A video push module is connected to the crowd positioning module, automatically edits the content to be broadcast live into several short videos, pushes them to the preliminary target users and analyzes them, selects the short video with the best effect for multi-channel push according to the analysis results, and determines the final target user according to the multi-channel push results; A process arrangement module is connected to the video push module, and analyzes the viewing history and duration of the final target user, obtains the most active viewing period of the historical live broadcast and the retention of the final target user according to the analysis result, and plans the live broadcast duration, and determines the conversion node by analyzing the purchase behavior of the final target user in the historical data, and determines the live broadcast process based on the conversion node and the most active viewing period; The feedback adjustment module is connected to the process scheduling module, and counts the live broadcast time watched by the final target users who enter the live broadcast room during the live broadcast time and records the interactive content, so as to adjust the live broadcast process in the next live broadcast time period.

2. The live broadcast accurate traffic diversion system based on big data according to claim 1 is characterized in that: The time determination module comprises: Statistics unit, used to collect statistics of large-scale live broadcast information on multiple platforms; The time determination unit is connected to the statistical unit and is used to determine the live broadcast time according to the statistical result.

3. The live broadcast accurate traffic diversion system based on big data according to claim 2 is characterized in that: The crowd positioning module includes: Social media analysis unit to analyze and monitor discussions and trends related to live content on social media platforms and identify potential target users; The crowd tracking unit is connected to the social media analysis unit, searches for the ID of the potential target user through information collection technology, and locates the preliminary target user according to the ID.

4. The live broadcast accurate traffic diversion system based on big data according to claim 3 is characterized in that: The video push module includes: A video editing unit automatically edits the content to be broadcast live into several short videos; A short video push unit, connected to the video editing unit, for publishing the short video on major social media platforms and pushing it to the initial target users; A click-through rate tracking unit is connected to the short video push unit, and uses an analysis tool to track key indicators of click-through rate and viewing time of each short video, and selects the short video with the best effect for multi-channel push based on the key indicators; The target user adjustment unit is connected to the click rate tracking unit, collects and analyzes the data pushed through multiple channels, and determines the final target user based on the data analysis result.

5. The live broadcast accurate traffic diversion system based on big data according to claim 4 is characterized in that: The process arrangement module includes: A data collection unit, which aggregates the historical viewing data of the final target user; a behavior analysis unit connected to the data collection unit, analyzing the behavior pattern of the final target user according to the historical viewing data, and identifying the most active viewing time period of the final target user according to the analysis result; A retention rate analysis unit, connected to the behavior analysis unit, evaluates the final target user retention rate under different live broadcast durations in the historical data, and finds out the loss point of the final target user; A live broadcast duration planning unit, connected to the behavior analysis unit and the retention rate analysis unit, plans the live broadcast duration based on the most active viewing period of the final target user and the retention rate; A conversion effect analysis unit, connected to the live broadcast duration planning unit, analyzes the conversion effect of the live broadcast in the historical data and identifies which links promote the purchase behavior of the final target user; The format design unit is connected with the behavior analysis unit and the conversion effect analysis unit, and designs the live broadcast process based on the most active viewing period and the purchasing behavior of the final target user.

6. The live broadcast accurate traffic diversion system based on big data according to claim 5 is characterized in that: The behavior analysis unit comprises: A feature extraction subunit, used to extract features of viewing duration and viewing frequency from the historical viewing data; The active time period analysis subunit is connected to the feature extraction subunit and is used to count the viewing frequency and viewing duration in different time periods, and obtain the most active viewing time period according to the statistical results.

7. The live broadcast accurate traffic diversion system based on big data according to claim 6 is characterized in that: The retention rate analysis unit includes: The data cleaning subunit is used to clean the collected data and remove invalid or abnormal data; A retention rate calculation subunit is connected to the data cleaning subunit, and evenly divides the different live broadcast durations in the cleaned data into several time periods, counts the final target users retained in each time period, and calculates the retention rate of the target users in each time period; A retention curve drawing subunit is connected to the retention rate calculation subunit, and draws a retention curve according to the final target users retained in each time period to show the retention status of the final target users under different live broadcast durations; The churn point analysis subunit is connected to the retention curve drawing subunit, and identifies the peak value of the final target user churn through the drawn retention curve, and the point corresponding to the peak value is the churn point.

8. The live broadcast accurate traffic diversion system based on big data according to claim 7 is characterized in that: The conversion effect analysis unit comprises: Define the conversion indicator sub-unit, and make it clear that the indicator of conversion effect is the user's purchase behavior; A key link identification subunit is connected to the definition conversion indicator subunit to analyze various links in the historical live broadcast process and identify which links have a significant impact on the purchase behavior; The correlation analysis subunit is connected to the key link identification subunit, and uses statistical methods to analyze the influencing factors of the purchase link, and obtains the reasons for the conversion effect based on the analysis results.

9. The live broadcast accurate traffic diversion system based on big data according to claim 8 is characterized in that: The feedback adjustment module comprises: Identify and track the unit, using live broadcast platform analysis tools or third-party services to identify and track the final target users who enter the live broadcast room; A viewing time statistics unit, connected to the identification and tracking unit, for recording the viewing time of each final target user, including the time points of entering and leaving the live broadcast room; An interactive content recording unit, connected to the viewing time statistics unit, for collecting and recording the interactive behavior of the final target user in the live broadcast room; The strategy iteration unit is connected to the interactive content recording unit to aggregate the viewing time and interactive content data of all final target users, perform optimization analysis for the live broadcast content, and continuously iterate the live broadcast strategy based on the analysis results.

10. A method for accurate live streaming based on big data, characterized in that: include: Use big data to analyze the live broadcast time of major social media platforms and determine the live broadcast time based on the analysis results; Based on the determined live broadcast time, use information collection technology to search for the ID of potential target users, and locate the preliminary target users according to the ID; Automatically edit the content to be broadcast into several short videos, push them to the preliminary target users and analyze them, select the short videos with the best effects for multi-channel push according to the analysis results, and determine the final target users according to the multi-channel push results; Analyze the viewing history and duration of the final target user, obtain the most active viewing period of the historical live broadcast and the retention of the final target user according to the analysis results, plan the live broadcast duration, determine the conversion node by analyzing the purchase behavior of the final target user in the historical data, and determine the live broadcast process based on the conversion node and the most active viewing period; Statistics are collected on the live broadcast duration watched by the final target users who enter the live broadcast room during the live broadcast time, and the interactive content is recorded to adjust the live broadcast process in the next live broadcast time period.

Images