Internet e-commerce live broadcast dynamic management system

Abstract

The application relates to the technical field of electronic commerce, in particular to an internet e-commerce live broadcast dynamic management system, which comprises the following steps: establishing a flow investment plan library, automatically supplementing flow according to the flow gap of live broadcast, collecting user data to establish a label, matching goods and benefits according to the label and outputting explanation guidance for the host, monitoring orders to remind the host to adjust the rhythm and force single, guiding to add purchase after ordering, synchronously realizing after-sales and pushing the repurchase benefit. In the application, the flow investment plan library is built in advance, the real-time flow data and the gap prediction are combined in the live broadcast, the automatic matching investment scheme is supplemented, the stable flow supply is guaranteed in the whole live broadcast process, the flow utilization efficiency is improved, the user shopping broadcast preference data is collected in real time, the user exclusive label is constructed, the goods, benefits and live broadcast explanation rhythm are accurately matched for the user, the user stay time and conversion willingness are improved, the conversion link dynamic management and control are synchronously realized, the related goods are intelligently matched after ordering to guide to add purchase, the whole-link after-sales system is built, and the unit price and the repurchase rate are improved.

Description

Technical Field

[0001] This invention relates to the field of e-commerce technology, and in particular to an internet e-commerce live streaming dynamic management system. Background Technology

[0002] E-commerce technology is a field that relies on internet communication networks to achieve the digitalization of the entire process of online business transactions and operations. This field encompasses core aspects such as online product transactions, supply and demand matching, online marketing and promotion, transaction node management, user operation services, and supply chain collaborative scheduling. Overall, it achieves the digitalized operation and standardized management of the entire online business process by building online transaction platforms, integrating data across the entire operational chain, managing key transaction nodes, matching supply and demand resources, and optimizing business interactions and fulfillment links. Traditional internet e-commerce live streaming dynamic management refers to the technical aspects of scheduling and managing the entire live streaming process in e-commerce live streaming scenarios. This includes aspects such as live streaming session scheduling and personnel on-duty arrangements, verification of live streaming footage and content, updating and delisting live streaming product information, viewing and replying to user comments, connecting live streaming order fulfillment, and summarizing and recording live streaming data. Traditionally, operations personnel check live streaming time slots and personnel on-duty arrangements according to a workflow table, on-site management personnel view the live streaming footage and content in real time, product operations personnel manually modify and delist product information in the backend, on-site management personnel view and reply to user comments, order operations personnel verify orders and connect the fulfillment process, and data personnel summarize and record relevant live streaming data.

[0003] The current e-commerce live streaming process relies entirely on manual operations performed by different roles according to fixed procedures. Each role operates independently with disconnected data links, resulting in significant time lags between operational actions and data feedback. This makes real-time dynamic control of the entire live streaming process impossible. Manual operations at each stage are highly subjective and unstable, and the varying professional skills and execution efficiency of personnel in different roles easily lead to problems such as process execution deviations, omissions in node control, and untimely event responses. This directly affects the standardization and stability of live streaming operations. Furthermore, manual operations cannot adapt to the differentiated needs of different user groups in real time or deeply explore the value of individual customer purchases. This results in problems such as the inability to fill traffic gaps in a timely manner, low user conversion rates, and weak repeat purchase loyalty, ultimately lowering the overall operational efficiency and commercial revenue of the live streaming room. Summary of the Invention

[0004] To address the technical problems existing in the prior art, embodiments of the present invention provide an internet e-commerce live streaming dynamic management system, the system comprising: The dynamic streaming module pre-establishes a streaming plan library that has been approved for all scenarios. During the live stream, it automatically selects the corresponding plan from the streaming plan library to supplement the traffic based on the real-time traffic and the predicted traffic gap. The user tagging module automatically pushes interactive gameplay or warm-up messages to users after they enter the live broadcast room, and simultaneously collects data such as users' browsing preferences and source channels to build user-specific tags in real time. The precise delivery module intelligently matches and pushes exclusive products and benefits pop-ups based on users' unique tags, while providing the live streamer with explanations and pacing guidance that match the current group of users entering the room; The order conversion module monitors the number of people clicking on products in real time, predicts the number of product orders based on historical data from the same scenario, and reminds the live streamer to adjust the pace of the presentation, focus on user pain points, and launch limited-time benefits to complete the concentrated order conversion. The add-to-cart module allows users to intelligently match related products and display bundled offers based on their unique tags and purchased items after placing an order, thus guiding users to make additional purchases. The customer retention module creates a personalized service profile for each user after they place an order, synchronizes the order logistics status in real time, provides full-process after-sales follow-up, and pushes repeat purchase benefits to users after they sign for the package.

[0005] As a further aspect of the present invention, the dynamic streaming module includes: The plan library construction sub-module obtains the platform's traffic approval requirements, creates traffic plans according to different live streaming scenarios, completes the approval of all traffic plans in advance, integrates all traffic plans, and generates a traffic plan library. The traffic gap calculation submodule collects data on the number of online viewers and product changes in the live stream, combines this data with the subsequent product ranking schedule, predicts traffic changes in the next 15 minutes, calculates the number of viewers needed to fill the gap, and generates the number of viewers needed to fill the traffic gap. The traffic delivery plan execution submodule calls the corresponding traffic delivery plan from the traffic delivery plan library based on the number of people with traffic shortfall, and replenishes the corresponding traffic to the live broadcast room.

[0006] As a further aspect of the present invention, the user marking module includes: The automatic push submodule automatically pushes interactive gameplay and warm-up messages to users after they enter the live broadcast room, records the user's click time and dwell time on the pushed content, and generates interactive response data. The preference collection submodule, based on interactive response data, synchronously collects users' historical order data and source channels to generate historical feature data; The tag building submodule performs correlation calculations based on interactive response data and historical feature data to generate user-specific tags.

[0007] As a further aspect of the present invention, the precise receiving module includes: The tag matching submodule collects the attributes of products sold in the live broadcast room based on the user's exclusive tags, compares the correspondence between the user's exclusive tags and the product attributes, and generates the product matching score. The content push submodule selects the exclusive product with the highest matching degree, extracts the corresponding benefit content, combines the exclusive product and benefit information, and performs a pop-up push operation. The anchor guidance submodule calls up the corresponding script template based on the exclusive product type parameters, and adjusts the pacing parameters of the presentation.

[0008] As a further aspect of the present invention, the order conversion module includes: The order volume prediction submodule monitors the number of people clicking on products in real time, collects historical product click data in the same scenario, as well as corresponding product order data, matches the current live streaming scenario dimension parameters, and calculates the predicted order volume for the products. The order deviation verification submodule obtains the real-time actual order quantity of the product, calls the product's predicted order quantity, compares the actual order quantity with the product's predicted order quantity, assigns a value of 1 when the actual order quantity is less than the product's predicted order quantity, and assigns a value of 0 when the actual order quantity is not less than the product's predicted order quantity, and generates a forced order execution judgment value.

[0009] As a further aspect of the present invention, the order conversion module includes: The order-pushing execution submodule calls the order-pushing execution judgment value. When the order-pushing execution judgment value is 1, it pushes instructions to adjust the explanation rhythm, focus on user pain points, and launch limited-time benefits. When the order-pushing execution judgment value is 0, it pushes instructions to maintain the live broadcast rhythm.

[0010] As a further aspect of the present invention, the add-on purchase module includes: The user preference parsing submodule retrieves the user's exclusive tags and purchased items after the user places an order, calculates the similarity between the weight values ​​of each dimension in the user's exclusive tags and the attribute values ​​of the purchased items, and generates a user preference vector. The associated product matching submodule collects the attribute values ​​of all products based on the user preference vector, calculates the matching value between the user preference vector value and each product attribute tag, and filters the top three products with the highest matching values ​​to generate a candidate product list.

[0011] As a further aspect of the present invention, the add-on purchase module includes: The "Add to Cart Discount" submodule collects the unit price and available coupon information of the corresponding products from the candidate product list and pushes it to the user who places an order.

[0012] As a further aspect of the present invention, the customer retention module includes: The service profile module collects user information and order information after a user places an order, creates a unique service profile for the user, and generates a unique service profile. The logistics synchronization submodule, based on the dedicated service file, matches the corresponding order, obtains the order logistics node data, updates the logistics node data in real time, and synchronizes the logistics node data to the dedicated service file.

[0013] As a further aspect of the present invention, the customer retention module includes: The repeat purchase push submodule calls the dedicated service file, determines the user's receipt status, locks the users who have signed for receipt, matches the corresponding repeat purchase benefits, and pushes the repeat purchase benefits to the corresponding users.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, a traffic allocation plan library is built in advance. During the live stream, real-time traffic data and gap prediction are combined to automatically match the traffic allocation plan to supplement the traffic, ensuring a stable traffic supply throughout the live stream and improving traffic utilization efficiency. User browsing preference data is collected in real time to build user-specific tags and accurately match products, benefits and live stream explanation rhythm for them, increasing user dwell time and conversion intention. At the same time, dynamic control of the conversion process is realized. After placing an order, related products are intelligently matched to guide users to add to their cart. A full-link after-sales system is built to improve the average order value and repurchase rate. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a system flowchart of the present invention; Figure 2 This is a flowchart of the dynamic flow delivery module of the present invention; Figure 3 This is a flowchart of the user marking module of the present invention; Figure 4 This is a flowchart of the precise receiving module of the present invention; Figure 5 This is a flowchart of the order conversion module of the present invention; Figure 6 This is a flowchart of the accompanying add-on purchase module of the present invention; Figure 7 The flowchart is for the customer retention module of this invention. Detailed Implementation

[0017] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0018] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0019] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0020] This invention provides an internet e-commerce live streaming dynamic management system, such as... Figure 1-2 The diagram shown illustrates a dynamic management system for internet e-commerce live streaming. The system includes: The dynamic streaming module pre-establishes a streaming plan library that has been approved for all scenarios. During the live stream, it automatically selects the corresponding plan from the streaming plan library to supplement the traffic based on the real-time traffic and the predicted traffic gap. The user tagging module automatically pushes interactive gameplay or warm-up messages to users after they enter the live broadcast room, and simultaneously collects data such as users' browsing preferences and source channels to build user-specific tags in real time. The precise delivery module intelligently matches and pushes exclusive products and benefits pop-ups based on users' unique tags, while providing the live streamer with explanations and pacing guidance that match the current group of users entering the room; The order conversion module monitors the number of people clicking on products in real time, predicts the number of product orders based on historical data from the same scenario, and reminds the live streamer to adjust the pace of the presentation, focus on user pain points, and launch limited-time benefits to complete the concentrated order conversion. The add-to-cart module allows users to intelligently match related products and display bundled offers based on their unique tags and purchased items after placing an order, thus guiding users to make additional purchases. The customer retention module creates a personalized service profile for each user after they place an order, synchronizes the order logistics status in real time, provides full-process after-sales follow-up, and pushes repeat purchase benefits to users after they sign for the package.

[0021] Specifically, such as Figure 2 As shown, the dynamic streaming module includes: The plan library construction sub-module obtains the platform's traffic approval requirements, creates traffic plans according to different live streaming scenarios, completes the approval of all traffic plans in advance, integrates all traffic plans, and generates a traffic plan library. The platform's ad placement review requirements are broken down into different live streaming scenarios such as beauty, apparel, and food. The review dimensions of ad placement time, material type, and audience tags are extracted for each scenario. Ad placement plans with corresponding time-slot targeting audiences and bid-material combinations are created according to the requirements of beauty live streaming and apparel live streaming. Each ad placement plan for each scenario is submitted to the platform's review interface one by one. The compliance of the plan name, targeting range, material links, and bid range is verified. Admitted plans are marked as available. Rejected plans are re-edited according to the feedback and resubmitted until all plans pass the platform's review. The approved ad placement plans for each scenario are uniformly classified and stored according to scenario number, time period, conditions, and targeting dimensions to form a structured and callable ad placement plan library. The traffic gap calculation submodule collects data on the number of online viewers and product changes in the live stream, combines this data with the subsequent product ranking schedule, predicts traffic changes in the next 15 minutes, calculates the number of viewers needed to fill the gap, and generates the number of viewers needed to fill the traffic gap. The live stream data includes real-time online viewership, product clicks, add-to-cart and sales data. It collects peak and average online viewership per minute, as well as current product clicks, add-to-cart and sales. It also reads the product ranking sequence and corresponding estimated conversion weights for the next 15 minutes. Using the fluctuation range of online viewership per minute combined with product ranking switching points, it employs a linear trend extrapolation method to predict changes in online viewership per minute. The estimated total online viewership is obtained by summing the estimated online viewership for each minute of the next 15 minutes. The target total online viewership is obtained by multiplying the target active online viewership by 15. The estimated total online viewership is then subtracted from the target total online viewership to obtain the traffic gap value, and this positive result is taken as the number of people needed to fill the traffic gap. The traffic delivery plan execution submodule calls the corresponding traffic delivery plan from the traffic delivery plan library based on the number of people with traffic shortfall, and supplements the corresponding traffic to the live broadcast room; The system matches the number of people with the traffic gap to the corresponding ad spend tiers based on numerical ranges. According to the current live streaming scenario, time period, and product type, it retrieves available ad spend plans with the same scenario and time period from the ad spend plan library and matches them with the correct targeting. It selects an ad spend plan with the corresponding bid and exposure based on the number of people with the traffic gap, calls the platform's ad spend API to execute the ad spend plan, and delivers creative materials to the targeted audience according to the plan settings. It provides real-time feedback on ad spend exposure, clicks, and entry volume. It fine-tunes the ad spend rate and the duration of each ad spend plan based on the difference between the actual entry volume and the number of people with the traffic gap, and continuously replenishes the corresponding traffic to the live streaming room.

[0022] Specifically, such as Figure 3 As shown, the user labeling module includes: The automatic push submodule automatically pushes interactive gameplay and warm-up messages to users after they enter the live broadcast room, records the user's click time and dwell time on the pushed content, and generates interactive response data. When a user enters the live stream, a signal is triggered. Warm-up messages and interactive activities are matched according to the entry time. A welcome message is pushed for 0-10 seconds, a like and follow activity is pushed for 10-30 seconds, and a comment interaction activity is pushed for 30-60 seconds. The system records the timestamp of the user clicking the push control and the duration from push display to exit. The click time is accurate to the second, and the dwell time is the difference between the display time and the exit time. When the user clicks twice and the display time is 15 seconds, the dwell time is recorded as 15 seconds. The number of clicks, click time, and dwell time of each activity for a single user are summarized by field to generate interactive response data. The preference collection submodule, based on interactive response data, synchronously collects users' historical order data and source channels to generate historical feature data; Retrieve user order records for the past 30 days, extract order product categories, order frequency, and average order value range. Order frequency is calculated by natural day, and average order value is the average of actual payment amount. The user source channels are collected as recommendation page, search, sharing, and paid traffic. The source channels are coded and marked. Interaction click data, order data, and channel data are bound to the same user ID to form structured data containing interaction behavior, orders, and sources, generating historical feature data. The tag building submodule performs correlation calculations based on interactive response data and historical feature data to generate user-specific tags. Users with an interaction duration of more than 10 seconds are marked as high-response users, those with an order frequency of more than 3 times are marked as high-frequency users, and those whose source channel is sharing are marked as social users. High response, high frequency, and social features are assigned values ​​according to weights. Feature matching operation is used to select users who meet both high response and high frequency and assign them high-value tags. Users with short stay duration and no orders are assigned potential customer tags. Through multi-feature cross-matching operation, a unique combination identifier is generated for each user, generating user-specific tags.

[0023] Specifically, such as Figure 4 As shown, the precise receiving module includes: The tag matching submodule collects the attributes of products sold in the live broadcast room based on the user's exclusive tags, compares the correspondence between the user's exclusive tags and the product attributes, and generates the product matching score. Based on user-specific tags, we extract user preference categories, price ranges, and functional requirements. We collect attributes such as category, pricing, target audience, and core functions of products sold in the live stream. We compare user tag features with product attributes item by item. For items with the same category, we count one match; for items with the same price range, we count one match; and for items with the same functional requirements, we count one match. We divide the number of matches by the total number of matches to get the ratio. When the user tags are beauty and moisturizing and under 100 yuan, and the product is a moisturizing face cream priced at 99 yuan, the number of matches is 3, the total number of matches is 3, and the ratio is 1. This ratio is used as the product matching degree. The content push submodule selects the exclusive product with the highest matching degree, extracts the corresponding benefit content, combines the exclusive product and benefit information, and performs a pop-up push operation. Iterate through all product matching scores, filter out the product with the highest value as the target exclusive product, extract the corresponding coupons, discounts, gifts and other benefits for the product, combine the exclusive product title, image, price field and benefit information into a unified content, call the front-end pop-up control interface, load the combined content into the pop-up container, send the display instruction to the corresponding user terminal, and complete the pop-up push operation; The anchor guidance submodule calls up the corresponding script template based on the exclusive product type parameters, and adjusts the pacing parameters at the same time; Read parameters such as product category, price range, and selling point type for the exclusive product. Match the corresponding category explanation template in the script template library according to the parameters. Read the current live broadcast explanation duration, interaction frequency, and conversion rate data. Set the explanation rhythm baseline value to 60 seconds per product. When the product price is higher than 200 yuan, the explanation rhythm parameter is adjusted to 90 seconds. When the product price is lower than 100 yuan, the explanation rhythm parameter is adjusted to 45 seconds. The explanation rhythm parameter adjustment is completed.

[0024] Specifically, such as Figure 5 As shown, the order conversion module includes: The order volume prediction submodule monitors the number of people clicking on products in real time, collects historical product click data in the same scenario, as well as corresponding product order data, matches the current live streaming scenario dimension parameters, and calculates the predicted order volume for the products. The system collects the number of clicks on the current product in real time, counts the number of clicks per minute, retrieves historical live streaming data from the last three live streams of the same product category, price range, and traffic period, extracts the click volume and transaction volume for the corresponding period, calculates the historical average conversion rate, and multiplies the current number of clicks by the historical average conversion rate. With 120 clicks currently and a historical conversion rate of 8%, the predicted order volume is 9.6 orders. Rounding down to 9 orders, the predicted order volume for the product is obtained. The order deviation verification submodule obtains the real-time actual order quantity of the product, calls the product's predicted order quantity, compares the actual order quantity with the product's predicted order quantity, assigns a value of 1 when the actual order quantity is less than the product's predicted order quantity, and assigns a value of 0 when the actual order quantity is not less than the product's predicted order quantity, and generates a forced order execution judgment value. Read the current real-time actual order quantity of the product from the backend order interface, retrieve the calculated predicted order quantity, compare the two sets of values, if the actual order quantity is 6 orders, which is less than the predicted order quantity of 9 orders, then assign a judgment flag of 1; if the actual order quantity is 10 orders, which is not less than the predicted order quantity of 9 orders, then assign a judgment flag of 0. Use this flag as the judgment result to generate the forced order execution judgment value. The order-forcing execution submodule calls the order-forcing execution judgment value. When the order-forcing execution judgment value is 1, it pushes instructions to adjust the explanation rhythm, focus on user pain points, and launch limited-time benefits. When the order-forcing execution judgment value is 0, it pushes instructions to maintain the live broadcast rhythm. Read the order-pushing execution judgment value. When the judgment value is 1, send an execution instruction to the broadcaster to speed up the explanation pace, highlight user pain points, and activate limited-time benefits. When the judgment value is 0, send an instruction to the broadcaster to maintain the existing explanation speed, process, and benefits, and push the corresponding live broadcast rhythm control instruction.

[0025] Specifically, such as Figure 6 As shown, the add-on shopping module includes: The user preference parsing submodule retrieves the user's exclusive tags and purchased items after the user places an order, calculates the similarity between the weight values ​​of each dimension in the user's exclusive tags and the attribute values ​​of the purchased items, and generates a user preference vector. After a user completes an order, the user's generated exclusive tags are retrieved, and the category, price, function, and material attributes of the purchased goods are extracted. The user's exclusive tags are aligned with the product attributes item by item. Category overlap is recorded as 0.3, price range overlap as 0.3, function matching as 0.2, and material matching as 0.2. The scores of each item are added together to form a vector combination. The user's tags are beauty, moisturizing, and under 100 yuan. The purchased goods are moisturizing face cream. The scores of each item are added together to form a vector. The user's tags are beauty, moisturizing, and under 100 yuan. The purchased goods are moisturizing face cream. The scores of each item are added together to form a vector. The user's preference vector is generated by combining these multi-dimensional values. The associated product matching submodule collects the attribute values ​​of all products based on the user preference vector, calculates the matching value between the user preference vector value and each product attribute tag, and filters the top three products with the highest matching values ​​to generate a candidate product list; Read the category, price, function, and material values ​​of the user preference vector, traverse all products for sale in the live broadcast room and extract the corresponding attribute tags, calculate the overlap between the user preference vector and the product attributes item by item, sort them from high to low values, select the top three product records, remove purchased products and keep three items, and organize the three product IDs, names, and attributes to generate a candidate product list. The "Add to Cart Discount" submodule collects the unit price and available coupon information of the corresponding products from the candidate product list and pushes it to the user who has placed an order. Based on the product ID in the candidate product list, the corresponding unit price is read from the product database, and the information on applicable discount coupons and coupons for the product is retrieved from the coupon database. The product name, unit price, discount amount, and usage conditions are combined into a message content and sent to users who have placed orders through the live broadcast private message channel.

[0026] Specifically, such as Figure 7 As shown, the customer retention module includes: The service profile module collects user information and order information after a user places an order, creates a unique service profile for the user, and generates a unique service profile. After a user successfully places an order, the system collects the user ID, nickname, contact information, and delivery location information, as well as the order number, product name, specifications, actual payment amount, and order time. Using the user ID as the unique index field, the system associates and binds the user information with the order information, creates an independent profile entry for each user, formats and stores the fields, and generates a unique service profile. The logistics synchronization submodule, based on the dedicated service file, matches the corresponding order, obtains the order logistics node data, updates the logistics node data in real time, and synchronizes the logistics node data to the dedicated service file. Based on the user ID and order number in the exclusive service file, match the same order number in the logistics system, obtain the time and status data of each node of pickup, transit, delivery and signing, poll the logistics interface every 5 minutes, read the latest status and overwrite the old data, write the updated logistics node data into the corresponding field of the exclusive service file, and complete the synchronous update of logistics information. The repeat purchase push submodule calls the dedicated service file, determines the user's receipt status, locks the users who have already received the goods, matches the corresponding repeat purchase benefits, and pushes the repeat purchase benefits to the corresponding users in a targeted manner. Retrieve the logistics status field from the exclusive service file to determine if the logistics status is "signed for". Filter out users whose status is "signed for". Match corresponding repurchase coupons and gifts based on the purchased product category and usage period. Push the benefits to the corresponding users through their private messaging channels.

[0027] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A dynamic management system for internet e-commerce live streaming, characterized in that, The system includes: The dynamic streaming module pre-establishes a streaming plan library that has been approved for all scenarios. During the live stream, it automatically selects the corresponding plan from the streaming plan library to supplement the traffic based on the real-time traffic and the predicted traffic gap. The user tagging module automatically pushes interactive gameplay or warm-up messages to users after they enter the live broadcast room, and simultaneously collects data such as users' browsing preferences and source channels to build user-specific tags in real time. The precise delivery module intelligently matches and pushes exclusive products and benefits pop-ups based on users' unique tags, while providing the live streamer with explanations and pacing guidance that match the current group of users entering the room; The order conversion module monitors the number of people clicking on products in real time, predicts the number of product orders based on historical data from the same scenario, and reminds the live streamer to adjust the pace of the presentation, focus on user pain points, and launch limited-time benefits to complete the concentrated order conversion. The add-to-cart module allows users to intelligently match related products and display bundled offers based on their unique tags and purchased items after placing an order, thus guiding users to make additional purchases. The customer retention module creates a personalized service profile for each user after they place an order, synchronizes the order logistics status in real time, provides full-process after-sales follow-up, and pushes repeat purchase benefits to users after they sign for the package.

2. The Internet e-commerce live streaming dynamic management system according to claim 1, characterized in that, The dynamic streaming module includes: The plan library construction sub-module obtains the platform's traffic approval requirements, creates traffic plans according to different live streaming scenarios, completes the approval of all traffic plans in advance, integrates all traffic plans, and generates a traffic plan library. The traffic gap calculation submodule collects data on the number of online viewers and product changes in the live stream, combines this data with the subsequent product ranking schedule, predicts traffic changes in the next 15 minutes, calculates the number of viewers needed to fill the gap, and generates the number of viewers needed to fill the traffic gap. The traffic delivery plan execution submodule calls the corresponding traffic delivery plan from the traffic delivery plan library based on the number of people with traffic shortfall, and replenishes the corresponding traffic to the live broadcast room.

3. The Internet e-commerce live streaming dynamic management system according to claim 2, characterized in that, The user labeling module includes: The automatic push submodule automatically pushes interactive gameplay and warm-up messages to users after they enter the live broadcast room, records the user's click time and dwell time on the pushed content, and generates interactive response data. The preference collection submodule, based on interactive response data, synchronously collects users' historical order data and source channels to generate historical feature data; The tag building submodule performs correlation calculations based on interactive response data and historical feature data to generate user-specific tags.

4. The Internet e-commerce live streaming dynamic management system according to claim 3, characterized in that, The precise receiving module includes: The tag matching submodule collects the attributes of products sold in the live broadcast room based on the user's exclusive tags, compares the correspondence between the user's exclusive tags and the product attributes, and generates the product matching score. The content push submodule selects the exclusive product with the highest matching degree, extracts the corresponding benefit content, combines the exclusive product and benefit information, and performs a pop-up push operation. The anchor guidance submodule calls up the corresponding script template based on the exclusive product type parameters, and adjusts the pacing parameters of the presentation.

5. The Internet e-commerce live streaming dynamic management system according to claim 4, characterized in that, The order conversion module includes: The order volume prediction submodule monitors the number of people clicking on products in real time, collects historical product click data in the same scenario, as well as corresponding product order data, matches the current live streaming scenario dimension parameters, and calculates the predicted order volume for the products. The order deviation verification submodule obtains the real-time actual order quantity of the product, calls the product's predicted order quantity, compares the actual order quantity with the product's predicted order quantity, assigns a value of 1 when the actual order quantity is less than the product's predicted order quantity, and assigns a value of 0 when the actual order quantity is not less than the product's predicted order quantity, and generates a forced order execution judgment value.

6. The Internet e-commerce live streaming dynamic management system according to claim 5, characterized in that, The order conversion module includes: The order-pushing execution submodule calls the order-pushing execution judgment value. When the order-pushing execution judgment value is 1, it pushes instructions to adjust the explanation rhythm, focus on user pain points, and launch limited-time benefits. When the order-pushing execution judgment value is 0, it pushes instructions to maintain the live broadcast rhythm.

7. The Internet e-commerce live streaming dynamic management system according to claim 6, characterized in that, The add-on purchase module includes: The user preference parsing submodule retrieves the user's exclusive tags and purchased items after the user places an order, calculates the similarity between the weight values ​​of each dimension in the user's exclusive tags and the attribute values ​​of the purchased items, and generates a user preference vector. The associated product matching submodule collects the attribute values ​​of all products based on the user preference vector, calculates the matching value between the user preference vector value and each product attribute tag, and filters the top three products with the highest matching values ​​to generate a candidate product list.

8. The Internet e-commerce live streaming dynamic management system according to claim 7, characterized in that: The add-on purchase module includes: The "Add to Cart Discount" submodule collects the unit price and available coupon information of the corresponding products from the candidate product list and pushes it to the user who places an order.

9. The Internet e-commerce live streaming dynamic management system according to claim 8, characterized in that: The customer retention module includes: The service profile module collects user information and order information after a user places an order, creates a unique service profile for the user, and generates a unique service profile. The logistics synchronization submodule, based on the dedicated service file, matches the corresponding order, obtains the order logistics node data, updates the logistics node data in real time, and synchronizes the logistics node data to the dedicated service file.

10. The Internet e-commerce live streaming dynamic management system according to claim 9, characterized in that: The customer retention module includes: The repeat purchase push submodule calls the dedicated service file, determines the user's receipt status, locks the users who have signed for receipt, matches the corresponding repeat purchase benefits, and pushes the repeat purchase benefits to the corresponding users.

Images