E-commerce live broadcast flow data analysis method and system

Abstract

The invention is suitable for the technical field of flow data analysis, and particularly relates to an e-commerce live broadcast flow data analysis method and system, and the method comprises the steps: S100, collecting the interactive behaviors of a user in an e-commerce live broadcast process, and enabling the interactive behaviors to at least comprise watching, commenting, liking, browsing a commodity, and placing an order; configuring a corresponding relationship between the interactive behavior and a user ID; and S200, determining the sequence of the interactive behaviors, counting the number of participants of each interactive behavior, creating step items in one-to-one correspondence with the interactive behaviors, and based on the number of participants. According to the invention, by constructing the funnel ladder table, the ladder items can be integrated, the data analysis efficiency is improved, and by collecting the behavior data, the specific behaviors of the user in live broadcast can be analyzed, so that the user demand can be accurately known, the user experience is improved, the problem link can be accurately positioned by determining lost personnel, and the user experience is improved. And the content pertinence and the live broadcast conversion rate are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic data analysis, and in particular to a method and system for analyzing traffic data of an e-commerce live broadcast. Background Art

[0002] E-commerce live streaming traffic data analysis is a method to understand the performance of live streaming and user behavior by collecting, processing and analyzing various types of data during the e-commerce live streaming process. The main data analyzed include the number of viewers, interaction volume, conversion rate, order volume and stay time, etc. Among them, users need to watch the live streaming, click on the purchase link, view the details and purchase to place an order, but not all users will complete all the steps. Some users will only watch the live streaming, comment and interact, and will not click on the purchase link, while some users will not comment or view the details, but will directly purchase; these data have unexpected benefits for improving the conversion efficiency of live streaming.

[0003] Therefore, “how to intuitively display the user's data in each interactive behavior and the reasons for user loss” is the technical problem that the present invention needs to solve. Summary of the invention

[0004] The purpose of the present invention is to provide an e-commerce live broadcast traffic data analysis method and system to solve the problem of "how to intuitively display the data of users in various interactive behaviors and the reasons for user loss" raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: A method for analyzing e-commerce live broadcast traffic data, the method comprising: S100: Collecting interactive behaviors of users during e-commerce live broadcasting, wherein the interactive behaviors at least include: watching, commenting, liking, browsing products, and placing orders, and configuring a corresponding relationship between the interactive behaviors and user IDs; S200: determining the order between the interactive behaviors, counting the number of participants in each of the interactive behaviors, creating ladder items corresponding to the interactive behaviors one by one, and querying a preset comparison table based on the number of participants, wherein the comparison table includes: a number item and a length item, determining the length of the ladder item, arranging and integrating all the ladder items in the order, and generating a funnel ladder table, wherein the top ladder item in the funnel ladder table corresponds to the viewing in the interactive behaviors; S300: collecting behavior data of all users in the funnel ladder table, and mounting the behavior data to corresponding user IDs, aligning the user IDs in the funnel ladder table, and linking to obtain a behavior chain; S400: comparing two adjacent ladder items, determining lost personnel, counting the number of lost personnel, and based on the comparison table, segmenting lost items from the ladder items, traversing the key nodes of each lost personnel, wherein the key nodes are used to characterize the reasons affecting the conversion of lost personnel, and attaching the key nodes to the lost items; S500: Input the behavior chain and the churn item into a preset diagnostic analysis model, output an analysis result, integrate the funnel ladder table into the analysis result, and push it to a preset terminal.

[0006] Furthermore, the S100 includes: Expanding the interactive behavior, constructing additional items, and inserting the additional items into the ladder items, wherein the additional items at least include: sharing, gift giving, and participating in a lucky draw; The user's identity information is obtained, and a user portrait is established based on the interactive behavior. The user portrait is used to classify the user IDs in the funnel ladder table to obtain several levels, wherein the levels include at least: high-intent users, potential users, and low-intent users.

[0007] Further, the S200 includes: Embed association rule mining mechanism between ladder items to traverse the mutual relationships; Receive the real-time data of the interactive behavior uploaded by the user, use the mutual relationship to predict the real-time data, and integrate the prediction result into the analysis result.

[0008] Further, the S200 includes: Numbering the sessions of the e-commerce live broadcasts, and inserting labels generated by the numbers into the funnel ladder table; The changing trends of the step items in different sessions are compared, and the changing trends are corrected using preset adjustment rules.

[0009] Further, the S300 includes: Integrate all behavior chains and construct historical data sets; Collect visitor data during the e-commerce live broadcast process, mark long-term visitors, where the number of visits of the long-term visitors is greater than 1, traverse the historical data set, and determine the promotion strategy.

[0010] Furthermore, the S400 includes: Define several time periods and split the e-commerce live broadcast into several slices; Determine the number of lost users in each time period, establish a mapping between slices and time periods, find slices with a loss number greater than the threshold, and define inefficient segments.

[0011] Furthermore, the method further comprises: Opening the interface of the funnel ladder table; The live broadcast data from different platforms is received via the interface, and the inefficient segments are adjusted.

[0012] Furthermore, the system comprises: A configuration module is used to collect the interactive behaviors of users during the e-commerce live broadcast process, wherein the interactive behaviors at least include: watching, commenting, liking, browsing products and placing orders, and configure the corresponding relationship between the interactive behaviors and user IDs; A generation module is used to determine the order between the interactive behaviors, count the number of participants in each interactive behavior, create ladder items corresponding to the interactive behaviors one by one, and query a preset comparison table based on the number of participants, wherein the comparison table includes: a number item and a length item, determine the length of the ladder item, arrange and integrate all the ladder items in the order, and generate a funnel ladder table, wherein the top ladder item in the funnel ladder table corresponds to the viewing in the interactive behavior; A linking module, used to collect the behavior data of all users in the funnel ladder table, and mount the behavior data to the corresponding user ID, align the user ID in the funnel ladder table, and link to obtain a behavior chain; A hanging module is used to compare two adjacent ladder items, determine the lost personnel, count the number of lost people, and based on the comparison table, cut the lost items from the ladder items, traverse the key nodes of each lost person, where the key nodes are used to characterize the reasons affecting the conversion of lost personnel, and hang the key nodes down to the lost items; The push module is used to input the behavior chain and the churn item into a preset diagnostic analysis model, output the analysis result, integrate the funnel ladder table into the analysis result, and push it to a preset terminal.

[0013] Furthermore, the configuration module includes: An insertion unit, used for extending the interactive behavior, constructing additional items, and inserting the additional items into the ladder items; The obtaining unit is used to obtain the identity information of the user, establish a user portrait based on the interactive behavior, and use the user portrait to classify the user IDs in the funnel ladder table to obtain several levels.

[0014] Furthermore, the generation module includes: The traversal unit is used to embed the association rule mining mechanism between the ladder items and traverse the mutual relations; An integration unit, configured to receive the real-time data of the interactive behavior uploaded by the user, predict the real-time data by using the mutual relationship, and integrate the prediction result into the analysis result; A numbering unit, used to number the sessions of the e-commerce live broadcast and insert labels generated by the numbers into the funnel ladder table; The correction unit is used to compare the change trend of the step items in different sessions and correct the change trend using a preset adjustment rule.

[0015] Compared with the prior art, the present invention has the following beneficial effects: By determining interactive behaviors, it is possible to conduct accurate data analysis on e-commerce live broadcasts, collect more detailed user behavior data, and understand how users perform in each interactive behavior. By constructing ladder items, it is possible to intuitively display the number of participants in each interactive behavior, greatly improving the pertinence of live broadcast content and providing data support for the optimization decisions of e-commerce live broadcasts. By constructing a funnel ladder table, it is possible to integrate ladder items and improve the efficiency of live broadcast data analysis. By collecting behavioral data, it is possible to analyze the specific behaviors of users in live broadcasts, thereby accurately understanding user needs and improving user experience. By determining the lost personnel, it is possible to accurately locate the problem links and greatly improve the content pertinence and live broadcast conversion rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flowchart of an e-commerce live streaming traffic data analysis method provided by an embodiment of the present invention; Figure 2 A schematic diagram of an example of a funnel ladder table in the e-commerce live broadcast traffic data analysis method provided by an embodiment of the present invention; Figure 3 A first sub-process flowchart of the e-commerce live streaming traffic data analysis method provided by an embodiment of the present invention; Figure 4 A second sub-process diagram of the e-commerce live streaming traffic data analysis method provided by an embodiment of the present invention; Figure 5 A third sub-process flowchart of the e-commerce live streaming traffic data analysis method provided by an embodiment of the present invention; Figure 6 A fourth sub-process flowchart of the e-commerce live streaming traffic data analysis method provided by an embodiment of the present invention; Figure 7 A block diagram of the composition of the e-commerce live streaming traffic data analysis system provided by an embodiment of the present invention; Figure 8 A block diagram of the configuration module in the e-commerce live streaming traffic data analysis system provided by an embodiment of the present invention; Fig. 9A block diagram of the composition of a generation module in an e-commerce live streaming traffic data analysis system provided by an embodiment of the present invention; Fig.10 A block diagram of the composition of the link module in the e-commerce live broadcast traffic data analysis system provided by an embodiment of the present invention; Fig.11 A block diagram of the composition of the downstream modules in the e-commerce live broadcast traffic data analysis system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] In Example 1, Figure 1 The implementation process of the e-commerce live broadcast traffic data analysis method provided by the embodiment of the present invention is shown and described in detail below: S100: Collecting the interactive behaviors of users during the e-commerce live broadcast process, wherein the interactive behaviors at least include: watching, commenting, liking, browsing products and placing orders, and configuring the corresponding relationship between the interactive behaviors and the user ID.

[0019] Collect users' interactive behaviors during e-commerce live broadcasts and record the interactive behaviors of each user ID. For example, if a user ID is X, and after entering the live broadcast room, he only watches the live broadcast for 3 minutes and does not make any comments or likes, then X's interactive behavior is watching the live broadcast for 3 minutes.

[0020] S200: Determine the order between the interactive behaviors, count the number of participants in each of the interactive behaviors, create ladder items corresponding to the interactive behaviors one by one, and query a preset comparison table based on the number of participants, wherein the comparison table includes: a number item and a length item, determine the length of the ladder item, arrange and integrate all the ladder items in the order, and generate a funnel ladder table, wherein the top ladder item in the funnel ladder table corresponds to the viewing in the interactive behavior.

[0021] The order of interactive behaviors is watch-comment-like-browse products-order. It should be noted that not all users will complete all the steps in the above order; it is also necessary to determine the number of participants in each interactive behavior; for example, determine all users who comment in the live broadcast room, or all users who watch the live broadcast, and classify the user IDs of these users into the corresponding ladder items. By determining the length of the ladder items according to the number of participants, and arranging the ladder items according to the order of interactive behaviors, all ladder items are integrated to generate a funnel ladder table. (Instruction Manual Attachment Figure 2The following is an example of a funnel ladder table. It should be noted that the length of each ladder item is determined by the number of participants in it. In a certain e-commerce live broadcast room, there are 8 users A, B, C, D, E, F, G and H, but there should be more in fact. During the live broadcast, all these 8 users must have watched the live broadcast, and these 8 users were classified into the ladder items corresponding to the viewing. However, among them, there were F, H and G, and these 3 users made comments. F, H, A and G viewed the detail page, and finally only H and A placed orders to purchase the products, and the users were classified into the corresponding ladder items.

[0022] S300: Collecting the behavior data of all users in the funnel ladder table, and mounting the behavior data to the corresponding user ID, aligning the user ID in the funnel ladder table, and linking to obtain a behavior chain.

[0023] Collect the behavior data of all users in the funnel ladder table, where the behavior data is the specific time the user watched the live broadcast and the specific comments posted, etc., mount the behavior data to the corresponding user ID, and align the user ID in the funnel ladder table; in the appendix of the manual Figure 2 In the funnel, D, B, and H are aligned, each user is extracted from the funnel ladder table, and the corresponding behavior data is used to build a behavior chain; for example, A's behavior chain can be watching the live broadcast for 14 minutes - viewing the details page of product A for 2 minutes, and the details page of product B for 1 minute - placing an order to purchase product B.

[0024] S400: Compare two adjacent ladder items to determine the lost personnel, count the number of lost personnel, and based on the comparison table, cut the lost items from the ladder items, traverse the key nodes of each lost personnel, where the key nodes are used to characterize the reasons that affect the conversion of lost personnel, and hang the key nodes down to the lost items.

[0025] Lost personnel are users who do not perform the next interactive behavior in order. According to the number of lost personnel, the lost items are divided from the ladder items, and the key nodes of each lost personnel are found; for example, in the appendix of the manual Figure 2 In the above example, the key nodes corresponding to the five users A, B, C, D and E are comments.

[0026] S500: Input the behavior chain and the churn item into a preset diagnostic analysis model, output an analysis result, integrate the funnel ladder table into the analysis result, and push it to a preset terminal.

[0027] The diagnostic analysis model in the existing technology is used to determine the analysis results, where the analysis results should include: the slices that lead to user loss in the e-commerce live broadcast, the reasons for user loss and defective products or detail pages, etc., where the preset terminal can be the live broadcast room manager terminal.

[0028] In Example 2, Figure 3 The implementation process of the e-commerce live broadcast traffic data analysis method provided by the embodiment of the present invention is shown, and S100 is described in detail as follows: S101: Expand the interactive behavior, construct additional items, and insert the additional items into the ladder items, wherein the additional items at least include: sharing, gift giving, and participating in a lottery.

[0029] Expand the interactive behavior and insert additional items into the ladder items, where the additional items can be sharing, giving gifts, participating in lucky draws, etc.

[0030] S102: Obtain the user's identity information, establish a user portrait based on the interactive behavior, and use the user portrait to classify the user IDs in the funnel ladder table to obtain several levels, wherein the levels at least include: high-intent users, potential users and low-intent users.

[0031] Obtain the user's identity information, such as the user's place of residence, age and gender, and create a user portrait for each user. At the same time, classify the users and user IDs in the ladder items based on the user portrait. After classification, multiple levels are obtained, and users are divided into high-intention users, potential users, and low-intention users.

[0032] In Example 3, Figure 4 The implementation process of the e-commerce live broadcast traffic data analysis method provided by the embodiment of the present invention is shown, and S200 is described in detail as follows: S201: embedding association rule mining mechanism between ladder items to traverse the mutual relationships.

[0033] An association rule mining mechanism is embedded between the ladder items, where the association rule mining mechanism is the data mining technology in the prior art, which can discover the relationship between the ladder items; for example, users who frequently like posts generally have a higher order conversion rate. In other words, there is a relationship between the ladder items corresponding to the likes and the ladder items corresponding to the orders.

[0034] S202: receiving real-time data of interactive behaviors uploaded by users, predicting the real-time data using the mutual relationships, and integrating the prediction results into the analysis results.

[0035] Receive real-time data of interactive behaviors uploaded by users and make predictions about their interactive behaviors.

[0036] In Example 4, Figure 4 The implementation process of the e-commerce live broadcast traffic data analysis method provided by the embodiment of the present invention is shown, and S200 is described in detail as follows: S203: Number the sessions of the e-commerce live broadcast, and insert labels generated by the numbers into the funnel ladder table.

[0037] The e-commerce live broadcast sessions are numbered, a funnel ladder table is created for each live broadcast, and the number is synchronized to the funnel ladder table.

[0038] S204: Compare the change trends of the step items in different sessions, and modify the change trends using preset adjustment rules.

[0039] By comparing the changing trends of the step items in different e-commerce live broadcasts, we can determine the changes in the number of participants in each interactive behavior, and use the preset adjustment rules to adjust the changing trends.

[0040] For example, through comparison, it is found that the number of participants in comments in several adjacent live broadcasts is gradually decreasing; at this time, the preset adjustment rules can be used to adjust the next live broadcast, where the adjustment rules are specific coping methods. The coping method for the gradual decrease in the number of participants is to increase interaction during the live broadcast.

[0041] In Example 5, Figure 5 The implementation process of the e-commerce live broadcast traffic data analysis method provided by the embodiment of the present invention is shown, and S300 is described in detail as follows: S301: Integrate all behavior chains and construct historical data sets.

[0042] Integrate all behavior chains and construct a historical data set; the historical data set is a collection of users' past interactive behaviors when watching e-commerce live broadcasts.

[0043] S302: Collect visitor data during the e-commerce live broadcast process, mark long-term visitors, where the number of visits of the long-term visitors is greater than 1, traverse the historical data set, and determine the promotion strategy.

[0044] Find out the long-term visitors from the users of the live broadcast room. Long-term visitors are users who have visited the live broadcast room before. By querying the historical data set of long-term visitors, different promotion strategies can be formulated for each long-term visitor.

[0045] In Example 6, Figure 6 The implementation process of the e-commerce live broadcast traffic data analysis method provided by the embodiment of the present invention is shown, and S400 is described in detail as follows: S401: Define several time periods and divide the e-commerce live broadcast into several slices.

[0046] E-commerce live broadcast is divided into several slices according to time periods.

[0047] S402: Determine the number of lost users in each time period, establish a mapping between slices and time periods, find slices with a number of lost users greater than a threshold, and define inefficient segments.

[0048] Determine the number of lost people in each time period, and establish a mapping between the number of lost people, slices and time periods. Slices with a loss number greater than a threshold are identified as inefficient segments, and the inefficient segments are sent to the preset terminal for the live broadcast room manager to adjust the e-commerce live broadcast.

[0049] In Example 7, different from Example 1, in this embodiment of the present invention, the method further includes: Opening the interface of the funnel ladder table; The live broadcast data from different platforms is received via the interface, and the inefficient segments are adjusted.

[0050] In addition to using the funnel ladder table to count and analyze e-commerce live broadcasts on a single platform, you can also integrate the e-commerce live broadcast data from different platforms into a funnel ladder table, expand the data source of the funnel ladder table, and adjust inefficient segments based on the live broadcast data from different platforms.

[0051] Figure 7 The structure diagram of the e-commerce live broadcast traffic data analysis system provided by an embodiment of the present invention is shown. The e-commerce live broadcast traffic data analysis system 1 includes: Configuration module 11, used to collect the interactive behaviors of users during the e-commerce live broadcast process, wherein the interactive behaviors at least include: watching, commenting, liking, browsing products and placing orders, and configure the corresponding relationship between the interactive behaviors and user IDs; A generating module 12 is used to determine the order between the interactive behaviors, count the number of participants in each interactive behavior, create ladder items corresponding to the interactive behaviors one by one, and query a preset comparison table based on the number of participants, wherein the comparison table includes: a number item and a length item, determine the length of the ladder item, arrange and integrate all the ladder items in the order, and generate a funnel ladder table, wherein the top ladder item in the funnel ladder table corresponds to the viewing in the interactive behavior; A linking module 13 is used to collect the behavior data of all users in the funnel ladder table, and mount the behavior data to the corresponding user ID, align the user ID in the funnel ladder table, and link to obtain a behavior chain; The following module 14 is used to compare two adjacent ladder items, determine the lost personnel, count the number of lost personnel, and based on the comparison table, cut the lost items from the ladder items, traverse the key nodes of each lost personnel, wherein the key nodes are used to characterize the reasons affecting the conversion of lost personnel, and the key nodes are linked to the lost items; The push module 15 is used to input the behavior chain and the churn items into a preset diagnostic analysis model, output the analysis results, integrate the funnel ladder table into the analysis results, and push them to a preset terminal.

[0052] Figure 8 The structure diagram of the e-commerce live broadcast traffic data analysis system provided by an embodiment of the present invention is shown, and the configuration module 11 includes: An inserting unit 111, used to extend the interactive behavior, construct additional items, and insert the additional items into the ladder items; The obtaining unit 112 is used to obtain the identity information of the user, establish a user portrait based on the interactive behavior, and use the user portrait to classify the user IDs in the funnel ladder table to obtain a plurality of levels.

[0053] Fig. 9 The structure diagram of the e-commerce live broadcast traffic data analysis system provided by the embodiment of the present invention is shown, and the generation module 12 includes: A traversal unit 121, used to embed an association rule mining mechanism between adjacent ladder items to traverse the mutual relationships; An integration unit 122, configured to receive the real-time data of the interactive behavior uploaded by the user, predict the real-time data by using the mutual relationship, and integrate the prediction result into the analysis result; A numbering unit 123, used to number the sessions of the e-commerce live broadcast and insert labels generated by the numbers into the funnel ladder table; The correction unit 124 is used to compare the change trends of the step items in different sessions and correct the change trends using preset adjustment rules.

[0054] Fig.10 The structure diagram of the e-commerce live broadcast traffic data analysis system provided by an embodiment of the present invention is shown, and the link module 13 includes: A construction unit 131 is used to integrate all behavior chains and construct a historical data set; The determination unit 132 is used to collect visitor data during the e-commerce live broadcast process, mark long-term visitors, where the number of visits of the long-term visitors is greater than 1, traverse the historical data set, and determine the promotion strategy.

[0055] Fig.11The structure diagram of the e-commerce live broadcast traffic data analysis system provided by the embodiment of the present invention is shown, and the hanging module 14 includes: A division unit 141 is used to define a number of time periods and divide the e-commerce live broadcast into a number of slices; The definition unit 142 is used to determine the number of lost users in each time period, establish a mapping between slices and time periods, find slices with a number of lost users greater than a threshold, and define inefficient segments.

[0056] The configuration module 11 is mainly used to complete step S100, the generation module 12 is mainly used to complete step S200, the link module 13 is mainly used to complete step S300, the hanging module 14 is mainly used to complete step S400, and the push module 15 is mainly used to complete step S500; The inserting unit 111 is mainly used to complete step S101, and the obtaining unit 112 is mainly used to complete step S102; The traversal unit 121 is mainly used to complete step S201, the integration unit 122 is mainly used to complete step S202, the numbering unit 123 is mainly used to complete step S203, and the correction unit 124 is mainly used to complete step S204; The construction unit 131 is mainly used to complete step S301, and the determination unit 132 is mainly used to complete step S302; The dividing unit 141 is mainly used to complete step S401, and the defining unit 142 is mainly used to complete step S402.

[0057] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0058] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0059] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for analyzing e-commerce live streaming traffic data, characterized in that: The method comprises: S100: Collecting interactive behaviors of users during e-commerce live broadcasting, wherein the interactive behaviors at least include: watching, commenting, liking, browsing products, and placing orders, and configuring a corresponding relationship between the interactive behaviors and user IDs; S200: determining the order between the interactive behaviors, counting the number of participants in each of the interactive behaviors, creating ladder items corresponding to the interactive behaviors one by one, and querying a preset comparison table based on the number of participants, wherein the comparison table includes: a number item and a length item, determining the length of the ladder item, arranging and integrating all the ladder items in the order, and generating a funnel ladder table, wherein the top ladder item in the funnel ladder table corresponds to the viewing in the interactive behaviors; S300: collecting behavior data of all users in the funnel ladder table, and mounting the behavior data to corresponding user IDs, aligning the user IDs in the funnel ladder table, and linking to obtain a behavior chain; S400: comparing two adjacent ladder items, determining lost personnel, counting the number of lost personnel, and based on the comparison table, segmenting lost items from the ladder items, traversing the key nodes of each lost personnel, wherein the key nodes are used to characterize the reasons affecting the conversion of lost personnel, and attaching the key nodes to the lost items; S500: Input the behavior chain and the churn item into a preset diagnostic analysis model, output an analysis result, integrate the funnel ladder table into the analysis result, and push it to a preset terminal.

2. The e-commerce live broadcast traffic data analysis method according to claim 1 is characterized in that: The S100 includes: Expanding the interactive behavior, constructing additional items, and inserting the additional items into the ladder items, wherein the additional items at least include: sharing, gift giving, and participating in a lucky draw; The user's identity information is obtained, and a user portrait is established based on the interactive behavior. The user portrait is used to classify the user IDs in the funnel ladder table to obtain several levels, wherein the levels include at least: high-intent users, potential users, and low-intent users.

3. The e-commerce live broadcast traffic data analysis method according to claim 1 is characterized in that: The S200 includes: Embed association rule mining mechanism between ladder items to traverse the mutual relationships; Receive the real-time data of the interactive behavior uploaded by the user, use the mutual relationship to predict the real-time data, and integrate the prediction result into the analysis result.

4. The e-commerce live broadcast traffic data analysis method according to claim 1 is characterized in that: The S200 includes: Numbering the sessions of the e-commerce live broadcasts, and inserting labels generated by the numbers into the funnel ladder table; The changing trends of the step items in different sessions are compared, and the changing trends are corrected using preset adjustment rules.

5. The e-commerce live broadcast traffic data analysis method according to claim 4 is characterized in that: The S300 includes: Integrate all behavior chains and construct historical data sets; Collect visitor data during the e-commerce live broadcast process, mark long-term visitors, where the number of visits of the long-term visitors is greater than 1, traverse the historical data set, and determine the promotion strategy.

6. The e-commerce live broadcast traffic data analysis method according to claim 1 is characterized in that: The S400 includes: Define several time periods and split the e-commerce live broadcast into several slices; Determine the number of lost users in each time period, establish a mapping between slices and time periods, find slices with a loss number greater than the threshold, and define inefficient segments.

7. The e-commerce live broadcast traffic data analysis method according to claim 6 is characterized in that: The method further comprises: Opening the interface of the funnel ladder table; The live broadcast data from different platforms is received via the interface, and the inefficient segments are adjusted.

8. An e-commerce live streaming traffic data analysis system, characterized in that: The system comprises: A configuration module is used to collect the interactive behaviors of users during the e-commerce live broadcast process, wherein the interactive behaviors at least include: watching, commenting, liking, browsing products and placing orders, and configure the corresponding relationship between the interactive behaviors and user IDs; A generation module is used to determine the order between the interactive behaviors, count the number of participants in each interactive behavior, create ladder items corresponding to the interactive behaviors one by one, and query a preset comparison table based on the number of participants, wherein the comparison table includes: a number item and a length item, determine the length of the ladder item, arrange and integrate all the ladder items in the order, and generate a funnel ladder table, wherein the top ladder item in the funnel ladder table corresponds to the viewing in the interactive behavior; A linking module, used to collect the behavior data of all users in the funnel ladder table, and mount the behavior data to the corresponding user ID, align the user ID in the funnel ladder table, and link to obtain a behavior chain; A hanging module is used to compare two adjacent ladder items, determine the lost personnel, count the number of lost people, and based on the comparison table, cut the lost items from the ladder items, traverse the key nodes of each lost person, where the key nodes are used to characterize the reasons affecting the conversion of lost personnel, and hang the key nodes down to the lost items; The push module is used to input the behavior chain and the churn item into a preset diagnostic analysis model, output the analysis result, integrate the funnel ladder table into the analysis result, and push it to a preset terminal.

9. The e-commerce live streaming traffic data analysis system according to claim 8 is characterized in that: The configuration module includes: An insertion unit, used for extending the interactive behavior, constructing additional items, and inserting the additional items into the ladder items; The obtaining unit is used to obtain the identity information of the user, establish a user portrait based on the interactive behavior, and use the user portrait to classify the user IDs in the funnel ladder table to obtain several levels.

10. The e-commerce live streaming traffic data analysis system according to claim 8, characterized in that: The generation module comprises: The traversal unit is used to embed the association rule mining mechanism between the ladder items and traverse the mutual relationships; An integration unit, configured to receive the real-time data of the interactive behavior uploaded by the user, predict the real-time data by using the mutual relationship, and integrate the prediction result into the analysis result; A numbering unit, used to number the sessions of the e-commerce live broadcast and insert labels generated by the numbers into the funnel ladder table; The correction unit is used to compare the change trend of the step items in different sessions and correct the change trend using a preset adjustment rule.

Images