Big data-based live broadcast e-commerce content management method and system

Through big data analysis and adaptive strategy configuration, the problems of insufficient flexibility and neglect of user participation in traditional e-commerce live broadcast promotion methods are solved, the interactive participation and promotion effect of the live broadcast room are improved, and more accurate user behavior understanding and Strategy optimization.

CN119090563AActive Publication Date: 2024-12-06ZHEJIANG ZHONGMIAO INFORMATION TECHNOLOGY CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202311385287.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-10-24
Publication Date
2024-12-06
Estimated Expiration
2043-10-24

AI Technical Summary

Technical Problem

Traditional e-commerce live broadcast promotion methods lack flexibility and cannot adjust strategies based on different live broadcast promotion nodes and real-time interactive data, resulting in poor promotion results, and neglecting user interaction participation, affecting the influence of the live broadcast room.

Method used

Adopt the live e-commerce content management method based on big data, and adaptively adjust content response parameters by obtaining live interactive big data, determining the participation in live interaction, and configuring push strategies based on participation and impact, selecting appropriate promotion nodes to improve User engagement and live broadcast room influence.

Benefits of technology

It has improved the interactive participation and promotion effect of the live broadcast room, achieved more accurate user behavior understanding and strategy optimization, and enhanced the overall influence of e-commerce live broadcasts.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119090563A_ABST
    Figure CN119090563A_ABST
Patent Text Reader

Abstract

The embodiment of the invention provides a live broadcast e-commerce content management method and system based on big data, and the method can effectively improve the live broadcast interaction participation degree through obtaining the live broadcast guide information and the content response parameter data of an e-commerce live broadcast room under each live broadcast promotion node, and carrying out the adaptive adjustment. Therefore, the promotion effect of the e-commerce live broadcast room is optimized. Based on the live broadcast interaction participation degree, the target live broadcast promotion node meeting the set live broadcast participation requirement can be selected, and the matched live broadcast promotion node is further determined, so that the participation degree of the user is improved. According to the target live broadcast promotion node and the live broadcast interaction participation degree corresponding to the matched live broadcast promotion node, the influence degree of the e-commerce live broadcast room can be more accurately determined. Based on the direct broadcasting room influence degree and the direct broadcasting interaction participation degree, push strategy configuration can be performed on the direct broadcasting guide information under each direct broadcasting promotion node, and the effect of the push strategy can be effectively improved, so that an e-commerce platform can be helped to better understand user behaviors.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of big data technology, and more specifically, to a live e-commerce content management method and system based on big data. Background Art

[0002] In recent years, e-commerce live streaming has become a popular sales model, which displays products to users through live streaming, attracting users to participate and purchase products. However, when conducting e-commerce live streaming, how to effectively promote the live streaming room, attract more users to participate, and increase the influence and sales of the live streaming room has become an important issue.

[0003] Traditional e-commerce live broadcast promotion methods are usually fixed, that is, one or several promotion methods are preset and applied to all live broadcast promotion nodes. This method ignores the fact that different live broadcast promotion nodes may require different promotion strategies, which may lead to poor promotion results. In addition, traditional promotion methods cannot adjust promotion strategies according to real-time interactive data and lack flexibility.

[0004] At the same time, traditional promotion methods often ignore the interactive participation of users. The interactive participation of the live broadcast room is an important factor affecting the influence of the live broadcast room. If users are not interested in the live broadcast content and the interactive participation is low, the influence of the live broadcast room will also be reduced. Therefore, a new promotion strategy configuration method is urgently needed, which can adjust the promotion strategy according to different live broadcast promotion nodes and real-time interactive data, improve the interactive participation and influence of the live broadcast room, and thus optimize the promotion effect of e-commerce live broadcast. Summary of the invention

[0005] In view of this, the purpose of this application is to provide a live e-commerce content management method and system based on big data.

[0006] According to the first aspect of the present application, a live e-commerce content management method based on big data is provided, which is applied to a live e-commerce service platform, and the method includes:

[0007] Acquire the live broadcast guidance information of the target e-commerce live broadcast room at each live broadcast promotion node and the content response parameter data of each live broadcast promotion node; the content response parameter data is generated by adaptively adjusting the live broadcast interaction big data of the live broadcast guidance information of the e-commerce live broadcast room at each live broadcast promotion node;

[0008] Determine the live broadcast interaction participation degree corresponding to each of the live broadcast promotion nodes based on the content response parameter data;

[0009] In each of the live broadcast promotion nodes, a target live broadcast promotion node corresponding to the live broadcast interactive participation degree that meets the set live broadcast participation requirement is obtained, and a matching live broadcast promotion node is determined based on the live broadcast guidance information under the target live broadcast promotion node;

[0010] Determine the live broadcast room influence corresponding to the target e-commerce live broadcast room based on the live broadcast interaction participation degree corresponding to the target live broadcast promotion node and the live broadcast interaction participation degree corresponding to the matched live broadcast promotion node;

[0011] Based on the influence of the live broadcast room and the interactive participation in the live broadcast, a push strategy is configured for the live broadcast guidance information under each live broadcast promotion node.

[0012] In a possible implementation of the first aspect, before obtaining the live broadcast guidance information of the target e-commerce live broadcast room under each live broadcast promotion node and the content response parameter data of each live broadcast promotion node, the method further includes:

[0013] Obtaining live interactive big data of the live broadcast guidance information of the e-commerce live broadcast room under each of the live broadcast promotion nodes; the live interactive big data includes live dialogue event data, live dialogue trigger domains corresponding to the live dialogue event data, and dialogue response results corresponding to the live dialogue event data;

[0014] Analyze the live conversation event data, the live conversation triggering domain, and the conversation response result of each live promotion node to generate the conversation conversion degree, actual conversation participation degree, and average conversation frequency of each live promotion node;

[0015] The conversation conversion degree, the actual conversation participation degree and the average conversation frequency of each of the live broadcast promotion nodes are weighted and integrated to generate content response parameter data of each of the live broadcast promotion nodes.

[0016] In a possible implementation of the first aspect, before obtaining the live broadcast interaction big data of the live broadcast guidance information of the e-commerce live broadcast room under each of the live broadcast promotion nodes, the method further includes:

[0017] Obtaining basic live broadcast dialogue event data of the live broadcast guidance information of the e-commerce live broadcast room under each of the live broadcast promotion nodes;

[0018] Noise dialogue event data is removed from the basic live dialogue event data to generate live dialogue event data under each live broadcast promotion node;

[0019] The live conversation content, conversation order data and e-commerce live broadcast room in the live conversation event data are analyzed to generate a conversation response result.

[0020] In a possible implementation of the first aspect, before removing noise dialogue event data from the basic live dialogue event data, the method further includes:

[0021] Obtaining the dialogue attributes corresponding to each of the basic live dialogue event data;

[0022] Performing violation analysis on the basic live conversation event data to generate first violation analysis data;

[0023] Performing violation analysis on the dialog attributes to generate second violation analysis data;

[0024] In the basic live conversation event data, the basic live conversation event data corresponding to at least one of the first violation analysis data and the second violation analysis data that is an unqualified result is determined as noise conversation event data.

[0025] In a possible implementation of the first aspect, weighted fusion of the conversation conversion degree, the actual conversation participation degree, and the average conversation frequency of each of the live broadcast promotion nodes is performed to generate content response parameter data of each of the live broadcast promotion nodes, including:

[0026] The conversation conversion degree, the actual conversation participation degree and the average conversation frequency of each of the live broadcast promotion nodes are respectively converted into regularization to generate a regularized conversation conversion degree, a regularized actual conversation participation degree and a regularized average conversation frequency of each of the live broadcast promotion nodes;

[0027] Obtaining a first influence coefficient corresponding to the conversation conversion degree, a second influence coefficient corresponding to the actual conversation participation degree, and a third influence coefficient corresponding to the average conversation frequency;

[0028] Based on the first influence coefficient, the second influence coefficient and the third influence coefficient, the regularized conversation conversion degree, the regularized actual conversation participation degree and the regularized conversation average frequency are integrated to generate content response parameter data of each of the live broadcast promotion nodes.

[0029] In a possible implementation manner of the first aspect, the content response parameter includes time domain content response parameters in multiple live content time domains;

[0030] The determining the live broadcast interaction participation degree corresponding to each live broadcast promotion node based on the content response parameter data includes:

[0031] Determining the time domain live broadcast interactive participation degree of each of the live broadcast promotion nodes in the multiple live broadcast content time domains according to the time domain content response parameters in the multiple live broadcast content time domains;

[0032] Determine the difference between the time domain live broadcast interactive participation degrees under the multiple live broadcast content time domains;

[0033] Determining, according to the difference, the live content time domain impact coefficients corresponding to the multiple live content time domains;

[0034] According to the time domain influence coefficient of the live broadcast content, the time domain live broadcast interaction participation degrees under the multiple live broadcast content time domains are integrated to generate the live broadcast interaction participation degree corresponding to each of the live broadcast promotion nodes.

[0035] In a possible implementation of the first aspect, the number of the matched live broadcast promotion nodes is multiple;

[0036] The determining the live broadcast room influence corresponding to the target e-commerce live broadcast room based on the live broadcast interaction participation degree corresponding to the target live broadcast promotion node and the live broadcast interaction participation degree corresponding to the matched live broadcast promotion node includes:

[0037] Determining a participation influence weight according to the live broadcast interactive participation of the plurality of matched live broadcast promotion nodes;

[0038] The live broadcast interactive participation degree corresponding to the target live broadcast promotion node is updated according to the participation influence weight, and the live broadcast room influence degree corresponding to the target e-commerce live broadcast room is generated.

[0039] In a possible implementation of the first aspect, the push strategy configuration for the live broadcast guidance information under each live broadcast promotion node based on the live broadcast room influence and the live broadcast interactive participation includes:

[0040] Determining a candidate live broadcast promotion node among the live broadcast promotion nodes according to the live broadcast interaction participation degree;

[0041] Determine the live broadcast guidance information of the candidate live broadcast promotion node in the live broadcast guidance information as the candidate live broadcast guidance information;

[0042] When the influence of the live broadcast room meets the first set live broadcast participation requirement, directly push strategy configuration for the candidate live broadcast guide information;

[0043] When the influence of the live broadcast room meets the second set live broadcast participation requirement, generating recommended guidance data for the candidate live broadcast promotion node that matches the second set live broadcast participation requirement, and pushing the candidate live broadcast guidance information and the recommended guidance data;

[0044] When the influence of the live broadcast room meets the third set live broadcast participation requirement and the guidance environment of the live broadcast guidance information meets the environmental conditions, a push strategy configuration is performed on the candidate live broadcast guidance information under the guidance environment.

[0045] In a possible implementation of the first aspect, the method further includes:

[0046] Obtain target live broadcast interaction big data of the candidate live broadcast guide information;

[0047] According to the candidate live broadcast guiding information, adjusting the content response parameter data of the candidate live broadcast promotion node to generate adjusted content response parameter data of the candidate live broadcast promotion node;

[0048] The adjusted live broadcast interaction participation degree of the candidate live broadcast promotion node is determined based on the adjusted content response parameter data, and the adjusted live broadcast interaction participation degree is used to guide the push strategy configuration of the live broadcast guidance information under the candidate live broadcast promotion node.

[0049] According to the second aspect of the present application, a live broadcast e-commerce service platform is provided, which includes a processor and a readable storage medium, and the readable storage medium stores a program, which, when executed by the processor, implements the aforementioned live broadcast e-commerce content management method based on big data.

[0050] According to the third aspect of the present application, a computer-readable storage medium is provided, in which computer-executable instructions are stored. When the computer-executable instructions are monitored to be executed, the aforementioned live e-commerce content management method based on big data is implemented.

[0051] According to any of the above aspects, in the present application, by obtaining the live broadcast guidance information and content response parameter data of the e-commerce live broadcast room under each live broadcast promotion node, and making adaptive adjustments, the live broadcast interactive participation can be effectively improved, thereby optimizing the promotion effect of the e-commerce live broadcast room. Based on the live broadcast interactive participation, the target live broadcast promotion node that meets the set live broadcast participation requirements can be selected, and the matching live broadcast promotion node can be further determined to increase user participation. According to the live broadcast interactive participation corresponding to the target live broadcast promotion node and the matching live broadcast promotion node, the influence of the e-commerce live broadcast room can be more accurately determined. Based on the influence of the live broadcast room and the live broadcast interactive participation, the push strategy configuration can be performed for the live broadcast guidance information under each live broadcast promotion node. This method is both flexible and practical, and can effectively improve the effect of the push strategy, thereby helping the e-commerce platform to better understand user behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0053] Figure 1 A flowchart of a live e-commerce content management method based on big data provided in an embodiment of the present application;

[0054] Figure 2 A schematic diagram of the component structure of a live e-commerce service platform provided in an embodiment of the present application for implementing the above-mentioned live e-commerce content management method based on big data is shown. DETAILED DESCRIPTION

[0055] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present application.

[0056] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0057] Figure 1 The flowchart of the live broadcast e-commerce content management method based on big data provided by the embodiment of the present application is shown. It should be understood that in other embodiments, the order of some steps in the live broadcast e-commerce content management method based on big data of this embodiment can be interchanged according to the need for annotation, or some steps can be omitted or deleted. The detailed steps of the live broadcast e-commerce content management method based on big data are introduced as follows.

[0058] Step S110, obtaining the live broadcast guidance information of the target e-commerce live broadcast room under each live broadcast promotion node and the content response parameter data of each live broadcast promotion node. The content response parameter data is generated by adaptively adjusting the live broadcast interaction big data of the live broadcast guidance information of the e-commerce live broadcast room under each live broadcast promotion node.

[0059] For example, the goal of the embodiment of the present application is to optimize a live broadcast room that is about to launch a new product. First, it is necessary to collect the live broadcast guidance information of this live broadcast room at various promotion nodes (such as social media platforms, advertising positions, email notifications, etc.), such as click-through rate, viewing time, etc. At the same time, it is also necessary to obtain the content response parameter data of each promotion node, which may include user behavior patterns, user feedback, etc.

[0060] Step S120, determining the live broadcast interaction participation degree corresponding to each of the live broadcast promotion nodes based on the content response parameter data.

[0061] For example, the collected data can be used to calculate the live interactive participation rate of each promotion node, that is, to evaluate the user's activity level at each promotion node. For example, if a social media platform has a high click-through rate and a long viewing time, then the interactive participation rate of this promotion node is relatively high.

[0062] Step S130, in each of the live broadcast promotion nodes, obtain the target live broadcast promotion node corresponding to the live broadcast interaction participation degree that meets the set live broadcast participation requirements, and determine the matching live broadcast promotion node based on the live broadcast guidance information under the target live broadcast promotion node.

[0063] For example, the goal of this step is to find promotion nodes that have high interactive participation and meet the set requirements (such as viewing time, click-through rate, etc.). For example, if a specific ad slot is found to have a high click-through rate and viewing time, then this ad slot may be an ideal target promotion node. Then, based on the live broadcast guidance information under this target promotion node, such as live broadcast content, time, etc., other promotion nodes with similar characteristics can be found.

[0064] Step S140, based on the live broadcast interaction participation corresponding to the target live broadcast promotion node and the live broadcast interaction participation corresponding to the matched live broadcast promotion node, determine the live broadcast room influence corresponding to the target e-commerce live broadcast room.

[0065] For example, in this step, the influence of the entire live broadcast room will be evaluated based on the interactive participation of the target promotion node and the matching promotion node. If the interactive participation of these nodes is high, the overall influence of the live broadcast room may be greater.

[0066] Step S150, based on the influence of the live broadcast room and the live broadcast interactive participation, push strategy configuration is performed on the live broadcast guidance information under each live broadcast promotion node.

[0067] For example, the push strategy can be adjusted based on the influence of the live broadcast room and the interactive participation of each promotion node. For example, if the interactive participation of a promotion node is very high, the frequency of delivery at this node may be increased; conversely, if the participation of a node is low, the promotion content may be changed or the number of pushes at this node may be reduced.

[0068] Based on the above steps, by obtaining the live broadcast guidance information and content response parameter data of the e-commerce live broadcast room under each live broadcast promotion node, and making adaptive adjustments, the live broadcast interactive participation can be effectively improved, thereby optimizing the promotion effect of the e-commerce live broadcast room. Based on the live broadcast interactive participation, the target live broadcast promotion node that meets the set live broadcast participation requirements can be selected, and the matching live broadcast promotion node can be further determined to increase user participation. According to the live broadcast interactive participation corresponding to the target live broadcast promotion node and the matching live broadcast promotion node, the influence of the e-commerce live broadcast room can be more accurately determined. Based on the influence of the live broadcast room and the live broadcast interactive participation, the push strategy configuration can be performed for the live broadcast guidance information under each live broadcast promotion node. This method is both flexible and practical, and can effectively improve the effect of the push strategy, thereby helping the e-commerce platform to better understand user behavior.

[0069] In a possible implementation, before step S110, the method further includes:

[0070] Step S101, obtaining live interaction big data of live broadcast guidance information of the e-commerce live broadcast room under each live broadcast promotion node. The live interaction big data includes live dialogue event data, live dialogue trigger domain corresponding to the live dialogue event data, and dialogue response result corresponding to the live dialogue event data.

[0071] Step S1021, analyzing the live conversation event data, the live conversation triggering domain and the conversation response result of each live promotion node, and generating the conversation conversion degree, actual conversation participation degree and average conversation frequency of each live promotion node.

[0072] Step S103, performing weighted fusion on the conversation conversion degree, the actual conversation participation degree and the average conversation frequency of each of the live broadcast promotion nodes, to generate content response parameter data of each of the live broadcast promotion nodes.

[0073] For example, in the case of optimizing the live broadcast room for the launch of a new product, it is necessary to first collect live interactive big data at each promotion node. This data includes, but is not limited to, live dialogue event data (e.g., user comments, likes, etc.), dialogue trigger domains (e.g., the time and context of these behaviors), and dialogue response results (e.g., the host or other users’ responses to user comments).

[0074] Then, we need to analyze these data to evaluate the performance of each promotion node. Specifically, we can calculate the conversation conversion rate (for example, the proportion of users participating in comments), the actual conversation participation rate (for example, the number of users who actually participated in the conversation), and the average conversation frequency (for example, the average number of comments posted by each user).

[0075] Finally, the above indicators can be weighted and fused to generate content response parameter data for each promotion node. For example, if it is believed that the conversation conversion rate has a greater impact on the promotion effect, then this indicator may be given a higher weight.

[0076] Through such analysis and calculation, we can obtain the content response parameter data of each promotion node, so as to make further optimization decisions.

[0077] In a possible implementation, before step S101, the method further includes:

[0078] Step A110, obtaining the basic live broadcast dialogue event data of the live broadcast guidance information of the e-commerce live broadcast room under each live broadcast promotion node.

[0079] Step A120, removing noise dialogue event data from the basic live dialogue event data, and generating live dialogue event data under each live broadcast promotion node.

[0080] Step A130, analyze the live conversation content, conversation order data and e-commerce live broadcast room in the live conversation event data to generate a conversation response result.

[0081] For example, in the example of continuing to optimize the live broadcast room for new product launches, first of all, it is necessary to collect all the basic live broadcast dialogue event data. This data includes all the comments, questions, likes and other interactive information sent by users at each promotion node.

[0082] Next, the data needs to be cleaned up to remove irrelevant or misleading information (noise conversation event data), such as advertising information, malicious attack speech, etc. After this step of processing, more accurate and effective live conversation event data will be obtained.

[0083] Finally, the conversation response results can be generated by analyzing what users talk about in the conversation, whether they place an order in the conversation, and their other activities in the e-commerce live broadcast room. For example, if many users are discussing a certain product and place an order during this period, it can be inferred that this product has a greater influence in the live broadcast.

[0084] The above three steps can better understand user behavior and extract valuable information from it to optimize the e-commerce live broadcast room.

[0085] In a possible implementation, before step A120, the method further includes:

[0086] Step A111, obtaining the dialogue attributes corresponding to each of the basic live dialogue event data.

[0087] Step A112, performing violation analysis on the basic live dialogue event data to generate first violation analysis data.

[0088] Step A113, performing violation analysis on the dialog attributes to generate second violation analysis data.

[0089] Step A114, in the basic live conversation event data, the basic live conversation event data corresponding to at least one of the first violation analysis data and the second violation analysis data having an unqualified result is determined as noise conversation event data.

[0090] For example, in the example of continuously optimizing the live broadcast room for new product releases, first, it is necessary to obtain the conversation attributes corresponding to each basic live dialogue event data. The conversation attributes may include the user's identity information, speech time, nature of the speech content, etc.

[0091] Next, the basic live conversation event data can be analyzed for violations. For example, some comments may contain illegal content such as advertisements or malicious attacks. Through natural language processing technology and machine learning algorithms, these illegal comments can be identified and marked as the first violation analysis data.

[0092] Then, it is necessary to conduct violation analysis on the conversation attributes. For example, if it is found that some users frequently post similar comments in a short period of time, these behaviors may be screen-swiping behaviors, and these conversation attributes can be marked as the second violation analysis data.

[0093] Finally, the corresponding basic live conversation event data for which at least one of the first violation analysis data and the second violation analysis data is an unqualified result can be determined as noise conversation event data. These data determined to be noise will be eliminated in subsequent analysis.

[0094] The above steps can help filter out valuable information from massive conversation event data, improving the accuracy and efficiency of data analysis.

[0095] In a possible implementation, step S103 may include:

[0096] Step S1031, respectively performing regular conversion on the conversation conversion degree, the actual conversation participation degree and the average conversation frequency of each of the live broadcast promotion nodes to generate the regularized conversation conversion degree, the regularized actual conversation participation degree and the regularized average conversation frequency of each of the live broadcast promotion nodes.

[0097] Step S1032, obtaining a first influence coefficient corresponding to the conversation conversion degree, a second influence coefficient corresponding to the actual conversation participation degree, and a third influence coefficient corresponding to the average conversation frequency.

[0098] Step S1033, based on the first influence coefficient, the second influence coefficient and the third influence coefficient, the regularized conversation conversion degree, the regularized actual conversation participation degree and the regularized conversation average frequency are integrated to generate content response parameter data of each of the live broadcast promotion nodes.

[0099] For example, in the optimization of the live broadcast room for new product launches, first of all, in order to allow different promotion nodes to be compared on the same scale, each indicator needs to be converted into a regularized form. For example, the conversation conversion rate, actual conversation participation rate, and average conversation frequency are all mapped to between 0 and 1.

[0100] Then, you need to determine the influence of each indicator on the final result. For example, you may find that the conversation conversion rate has a greater influence on the live broadcast room, so its influence coefficient will be set higher.

[0101] Finally, the regularized conversation conversion degree, actual conversation participation degree, and average conversation frequency can be integrated according to each influence coefficient. For example, the content response parameter data of a promotion node may be 0.6*regularized conversation conversion degree + 0.3*regularized actual conversation participation degree + 0.1*regularized average conversation frequency.

[0102] In a possible implementation manner, the content response parameters include time domain content response parameters in multiple live content time domains.

[0103] Step S120 may include:

[0104] Step S121, determining the time domain live broadcast interaction participation of each of the live broadcast promotion nodes in the multiple live broadcast content time domains according to the time domain content response parameters in the multiple live broadcast content time domains.

[0105] Step S122, determining the difference between the time domain live broadcast interaction participation degrees under the multiple live broadcast content time domains.

[0106] Step S123: determining the live content time domain impact coefficients corresponding to the multiple live content time domains according to the difference.

[0107] Step S124, based on the time domain impact coefficient of the live content, the time domain live interaction participation degrees under the multiple live content time domains are integrated to generate the live interaction participation degree corresponding to each of the live promotion nodes.

[0108] For example, in the optimization of the live broadcast room for new product releases, first of all, it is necessary to know that the content response parameter data may include time domain content response parameters under multiple live content time domains. For example, the behavior data of users watching live broadcasts in the morning, noon, and evening may be collected.

[0109] Next, you need to analyze this data to determine the live broadcast interaction and engagement of each promotion node at different time periods. For example, a social media platform may have higher interaction and engagement in the evening.

[0110] Then, we need to calculate the difference between the interactive participation in different time periods. For example, if the interactive participation of a promotion node in the morning is much lower than that in the evening, then the difference will be large.

[0111] Next, we need to determine the impact coefficient of each time period based on these differences. For example, if the difference in interactive participation in a time period is large, then the impact coefficient of this time period may be high.

[0112] Finally, the interaction participation of different time periods can be integrated according to the influence coefficient of each time period. For example, the interaction participation of a promotion node may be 0.4* interaction participation in the morning + 0.3* interaction participation in the afternoon + 0.3* interaction participation in the evening.

[0113] The above steps can more accurately understand the user behavior in different time periods, thereby optimizing the e-commerce live broadcast room.

[0114] In a possible implementation, there are multiple matching live broadcast promotion nodes.

[0115] Step S140 may include:

[0116] Step S141, determining the participation influence weight according to the live broadcast interaction participation of the plurality of matched live broadcast promotion nodes.

[0117] Step S142, updating the live broadcast interaction participation degree corresponding to the target live broadcast promotion node according to the participation influence weight, and generating the live broadcast room influence degree corresponding to the target e-commerce live broadcast room.

[0118] For example, if there are multiple matching live broadcast promotion nodes, in the optimization of the live broadcast room for new product releases, first, multiple promotion nodes matching the target promotion node may be found. For example, it is found that in addition to the target promotion node (a certain social media platform), the user behaviors of several other social media platforms are very close to the target promotion node.

[0119] Next, the participation influence weights of these matched promotion nodes need to be determined based on their interactive participation. For example, if a matched promotion node has a high interactive participation, its participation influence weight may be set higher.

[0120] Finally, the interactive participation of the target promotion node can be updated according to the participation influence weight of each matching promotion node. For example, if a matching promotion node has a high participation influence weight, it may increase the interactive participation of the target promotion node, thereby improving the influence of the entire e-commerce live broadcast room.

[0121] The above steps can provide a more comprehensive understanding of user behavior and extract valuable information from it to optimize the e-commerce live broadcast room.

[0122] In a possible implementation, step S150 may include:

[0123] Step S151, determining candidate live broadcast promotion nodes among the live broadcast promotion nodes according to the live broadcast interaction participation degree.

[0124] Step S152: Determine that the live broadcast guidance information of the candidate live broadcast promotion node in the live broadcast guidance information is the candidate live broadcast guidance information.

[0125] Step S153: When the influence of the live broadcast room meets the first set live broadcast participation requirement, a push strategy configuration is directly performed on the candidate live broadcast guide information.

[0126] Step S154, when the influence of the live broadcast room meets the second set live broadcast participation requirement, generate recommended guidance data for the candidate live broadcast promotion node that matches the second set live broadcast participation requirement, and push the candidate live broadcast guidance information and the recommended guidance data.

[0127] Step S155, when the influence of the live broadcast room meets the third set live broadcast participation requirement and the guidance environment of the live broadcast guidance information meets the environmental conditions, the push strategy configuration is performed on the candidate live broadcast guidance information under the guidance environment.

[0128] For example, in the optimization of the live broadcast room for new product releases, first, it is necessary to select candidate live broadcast promotion nodes based on the live broadcast interactive participation of each promotion node. For example, if the interactive participation of a promotion node exceeds the set threshold, then it can be considered a candidate promotion node.

[0129] Next, it is necessary to select the live broadcast guidance information of the candidate promotion node from all the live broadcast guidance information as the candidate live broadcast guidance information.

[0130] If the influence of the live broadcast room meets the first set participation requirement (for example, the influence is higher than a certain threshold), the candidate live broadcast guide information can be directly pushed.

[0131] If the influence of the live broadcast room meets the second set participation requirement (for example, the influence is within a certain range), then it is necessary to generate some recommended guidance data that matches this requirement (for example, changing the live broadcast time, adding interactive links, etc.) and push these guidance data together with the candidate live broadcast guidance information.

[0132] If the influence of the live broadcast room meets the participation requirements of the third setting (for example, the influence is lower than a certain threshold), and the guidance environment also meets certain set conditions (for example, most users are active at night), then the candidate live broadcast guidance information can be pushed under the guidance environment that meets the conditions.

[0133] The above steps can be used to formulate more effective push strategies based on different conditions, thereby optimizing the e-commerce live broadcast room.

[0134] In a possible implementation, the method further includes:

[0135] Step B110, obtaining the target live broadcast interactive big data of the candidate live broadcast guide information.

[0136] Step B120: adjusting the content response parameter data of the candidate live broadcast promotion node according to the candidate live broadcast guidance information to generate adjusted content response parameter data of the candidate live broadcast promotion node.

[0137] Step B130, determining the adjusted live broadcast interaction participation degree of the candidate live broadcast promotion node based on the adjusted content response parameter data, and the adjusted live broadcast interaction participation degree is used to guide the push strategy configuration of the live broadcast guidance information under the candidate live broadcast promotion node.

[0138] For example, in the optimization of the live broadcast room for new product launches, first of all, it is necessary to collect and analyze the target live broadcast interaction big data of the candidate live broadcast guide information. For example, data such as the click rate and participation of users in the live broadcast after seeing specific guide information may be collected.

[0139] Next, the content response parameters of the candidate promotion nodes need to be adjusted based on these data and the candidate live broadcast guidance information. For example, if the data shows that users’ engagement increases after seeing a specific guidance information, then the weight of this type of guidance information may be increased.

[0140] Finally, the adjusted live interactive participation of the candidate promotion node can be determined based on the adjusted content response parameter data. This adjusted interactive participation can be used to guide the push strategy of the live guide information under the candidate live promotion node. For example, if the interactive participation of a node after the guide information is adjusted is improved, then this type of guide information may be pushed more.

[0141] The above steps can help us understand user behavior more accurately and optimize the push strategy of the e-commerce live broadcast room based on this understanding.

[0142] Further, Figure 2 FIG. 1 is a schematic diagram showing the hardware structure of a live broadcast e-commerce service platform 100 for implementing the method provided in the embodiment of the present application. Figure 2 As shown, the live broadcast e-commerce service platform 100 may include one or more processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 and a controller 108 for communication functions. It can be understood by those skilled in the art that Figure 2 The structure shown is for illustration only and does not limit the structure of the live e-commerce service platform 100. For example, the live e-commerce service platform 100 may also include Figure 2 More or fewer components as shown, or with Figure 2 Different configurations shown.

[0143] The memory 104 can be used to store software programs and modules of application software, such as program instructions corresponding to the above-mentioned method embodiments in the embodiments of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, realizing the above-mentioned method of live e-commerce content management based on big data. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the live e-commerce service platform 100 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0144] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the live e-commerce service platform 100. In one example, the transmission device 106 includes a network adapter, which can be connected to other network devices through a base station so that it can communicate with the Internet. In one example, the transmission device 106 can be a radio frequency module, which is used to communicate with the Internet wirelessly.

[0145] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of the embodiments of the present application are described. Other embodiments are within the scope of the attached claims. In some cases, the actions or steps recorded in the claims can be performed according to an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0146] Each embodiment in the embodiments of the present application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the above different embodiments, since they are basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0147] A person of ordinary skill in the art will appreciate that all or part of the steps for implementing the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the above program may be loaded into a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

Claims

1. A live e-commerce content management method based on big data, characterized in that: The method comprises: Acquire the live broadcast guidance information of the target e-commerce live broadcast room at each live broadcast promotion node and the content response parameter data of each live broadcast promotion node; the content response parameter data is generated by adaptively adjusting the live broadcast interaction big data of the live broadcast guidance information of the e-commerce live broadcast room at each live broadcast promotion node; Determine the live broadcast interaction participation degree corresponding to each of the live broadcast promotion nodes based on the content response parameter data; In each of the live broadcast promotion nodes, a target live broadcast promotion node corresponding to the live broadcast interactive participation degree that meets the set live broadcast participation requirement is obtained, and a matching live broadcast promotion node is determined based on the live broadcast guidance information under the target live broadcast promotion node; Determine the live broadcast room influence corresponding to the target e-commerce live broadcast room based on the live broadcast interaction participation degree corresponding to the target live broadcast promotion node and the live broadcast interaction participation degree corresponding to the matched live broadcast promotion node; Based on the influence of the live broadcast room and the interactive participation in the live broadcast, a push strategy is configured for the live broadcast guidance information under each live broadcast promotion node.

2. The live e-commerce content management method based on big data according to claim 1 is characterized in that: Before obtaining the live broadcast guidance information of the target e-commerce live broadcast room under each live broadcast promotion node and the content response parameter data of each live broadcast promotion node, the method further includes: Obtaining live interactive big data of the live broadcast guidance information of the e-commerce live broadcast room under each of the live broadcast promotion nodes; the live interactive big data includes live dialogue event data, live dialogue trigger domains corresponding to the live dialogue event data, and dialogue response results corresponding to the live dialogue event data; Analyze the live conversation event data, the live conversation triggering domain, and the conversation response result of each live promotion node to generate the conversation conversion degree, actual conversation participation degree, and average conversation frequency of each live promotion node; The conversation conversion degree, the actual conversation participation degree and the average conversation frequency of each of the live broadcast promotion nodes are weighted and integrated to generate content response parameter data of each of the live broadcast promotion nodes.

3. The live e-commerce content management method based on big data according to claim 2 is characterized in that: Before obtaining the live broadcast interaction big data of the live broadcast guidance information of the e-commerce live broadcast room under each of the live broadcast promotion nodes, the method further includes: Obtaining basic live broadcast dialogue event data of the live broadcast guidance information of the e-commerce live broadcast room under each of the live broadcast promotion nodes; Noise dialogue event data is removed from the basic live dialogue event data to generate live dialogue event data under each live broadcast promotion node; The live conversation content, conversation order data and e-commerce live broadcast room in the live conversation event data are analyzed to generate a conversation response result.

4. The live broadcast e-commerce content management method based on big data according to claim 3 is characterized in that: Before removing noise dialogue event data from the basic live dialogue event data, the method further includes: Obtaining the dialogue attributes corresponding to each of the basic live dialogue event data; Performing violation analysis on the basic live conversation event data to generate first violation analysis data; Performing violation analysis on the dialog attributes to generate second violation analysis data; In the basic live conversation event data, the basic live conversation event data corresponding to at least one of the first violation analysis data and the second violation analysis data that is an unqualified result is determined as noise conversation event data.

5. The live e-commerce content management method based on big data according to claim 2 is characterized in that: The weighted fusion of the conversation conversion degree, the actual conversation participation degree and the average conversation frequency of each live broadcast promotion node is performed to generate content response parameter data of each live broadcast promotion node, including: The conversation conversion degree, the actual conversation participation degree and the average conversation frequency of each of the live broadcast promotion nodes are respectively converted into regularization to generate a regularized conversation conversion degree, a regularized actual conversation participation degree and a regularized average conversation frequency of each of the live broadcast promotion nodes; Obtaining a first influence coefficient corresponding to the conversation conversion degree, a second influence coefficient corresponding to the actual conversation participation degree, and a third influence coefficient corresponding to the average conversation frequency; Based on the first influence coefficient, the second influence coefficient and the third influence coefficient, the regularized conversation conversion degree, the regularized actual conversation participation degree and the regularized conversation average frequency are integrated to generate content response parameter data of each of the live broadcast promotion nodes.

6. The live broadcast e-commerce content management method based on big data according to claim 1 is characterized in that: The content response parameters include time domain content response parameters under multiple live content time domains; The determining the live broadcast interaction participation degree corresponding to each live broadcast promotion node based on the content response parameter data includes: Determining the time domain live broadcast interactive participation degree of each of the live broadcast promotion nodes in the multiple live broadcast content time domains according to the time domain content response parameters in the multiple live broadcast content time domains; Determine the difference between the time domain live broadcast interactive participation degrees under the multiple live broadcast content time domains; Determining, according to the difference, the live content time domain impact coefficients corresponding to the multiple live content time domains; According to the time domain influence coefficient of the live broadcast content, the time domain live broadcast interaction participation degrees under the multiple live broadcast content time domains are integrated to generate the live broadcast interaction participation degree corresponding to each of the live broadcast promotion nodes.

7. The live e-commerce content management method based on big data according to claim 1 is characterized in that: The number of the matched live broadcast promotion nodes is multiple; The determining the live broadcast room influence corresponding to the target e-commerce live broadcast room based on the live broadcast interaction participation degree corresponding to the target live broadcast promotion node and the live broadcast interaction participation degree corresponding to the matched live broadcast promotion node includes: Determining a participation influence weight according to the live broadcast interactive participation of the plurality of matched live broadcast promotion nodes; The live broadcast interactive participation degree corresponding to the target live broadcast promotion node is updated according to the participation influence weight, and the live broadcast room influence degree corresponding to the target e-commerce live broadcast room is generated.

8. The live e-commerce content management method based on big data according to any one of claims 1 to 7, characterized in that: The push strategy configuration of the live broadcast guidance information under each live broadcast promotion node based on the live broadcast room influence and the live broadcast interactive participation includes: Determining a candidate live broadcast promotion node among the live broadcast promotion nodes according to the live broadcast interaction participation degree; Determine the live broadcast guidance information of the candidate live broadcast promotion node in the live broadcast guidance information as the candidate live broadcast guidance information; When the influence of the live broadcast room meets the first set live broadcast participation requirement, directly push strategy configuration for the candidate live broadcast guide information; When the influence of the live broadcast room meets the second set live broadcast participation requirement, generating recommended guidance data for the candidate live broadcast promotion node that matches the second set live broadcast participation requirement, and pushing the candidate live broadcast guidance information and the recommended guidance data; When the influence of the live broadcast room meets the third set live broadcast participation requirement and the guidance environment of the live broadcast guidance information meets the environmental conditions, a push strategy configuration is performed on the candidate live broadcast guidance information under the guidance environment.

9. The live e-commerce content management method based on big data according to claim 8 is characterized in that: The method further comprises: Obtain target live broadcast interaction big data of the candidate live broadcast guide information; According to the candidate live broadcast guiding information, adjusting the content response parameter data of the candidate live broadcast promotion node to generate adjusted content response parameter data of the candidate live broadcast promotion node; The adjusted live broadcast interaction participation degree of the candidate live broadcast promotion node is determined based on the adjusted content response parameter data, and the adjusted live broadcast interaction participation degree is used to guide the push strategy configuration of the live broadcast guidance information under the candidate live broadcast promotion node.

10. A live broadcast e-commerce service platform, characterized in that: The live e-commerce service platform includes a processor and a readable storage medium, wherein the readable storage medium stores a program, which, when executed by the processor, implements the live e-commerce content management method based on big data as described in any one of claims 1 to 9.

Citation Information

Patent Citations

  • A channel data analysis system and method

    CN109583722A

  • Advertisement putting method based on big data technology, electronic equipment and storage medium

    CN112488747A

  • Online advertisement pushing method combined with big data analysis and cloud server

    CN112819553A

  • Big data-based live broadcast interaction control method, apparatus and device, and storage medium

    CN115225921A

  • Livestreaming content generation method and device

    WO2023138508A1