Commodity demand prediction method and system for live broadcast platform

Abstract

The invention relates to the field of commodity demand prediction, in particular to a commodity demand prediction method and system for a live broadcast platform, and the method comprises the steps: recognizing the video content which is being played in a live broadcast room according to the moment when a user enters the live broadcast room; analyzing the video content, and extracting keywords of the video content according to an analysis result; obtaining the popularity degree of the commodities in the live broadcast room by counting the frequency of searching the commodities by the users in the live broadcast room; on the basis of the keywords of the extracted video content and the popularity degree of the commodities, analyzing the completed transactions in the live broadcast room to determine a purchase reason proportion, and on the basis of the purchase reason proportion and the number of the commodities in the live broadcast room, predicting the future transaction volume of the live broadcast commodities; and recording the purchase reason of the user completing the transaction so as to determine the pushing mode of the future live broadcast. And the shopping experience and the purchasing efficiency of the user in the live broadcasting room are improved.

Description

Technical Field

[0001] The present invention relates to the field of commodity demand forecasting, and in particular to a commodity demand forecasting method and system for a live broadcast platform. Background Art

[0002] With the development of technology, especially the application of AI and big data technology, the live streaming e-commerce industry will continue to grow and is expected to maintain a steady growth trend in the next few years, and demand forecasting will become more refined and intelligent.

[0003] The patent document with Chinese publication number CN102118060A discloses a commodity demand forecasting system based on big data analysis, which includes: a user data acquisition module, a user behavior analysis module, a demand forecasting module, a personalized management module and a commodity recommendation generation module, wherein the output end of the user data acquisition module is respectively connected to the demand forecasting module and the input end of the personalized management module, the output end of the user behavior analysis module is connected to the input end of the demand forecasting module, the output end of the demand forecasting module is connected to the personalized management module, and the personalized management module is bidirectionally connected to the commodity recommendation generation module; the user data acquisition module is responsible for acquiring various types of user data from e-commerce platforms and other channels, and integrating, cleaning and classifying the data; The user behavior analysis module extracts, transforms and analyzes the user's operation data to identify the user's shopping preferences and demand tendency information, and generates shopping demand data; the demand prediction module calculates the user's shopping demand degree based on the shopping demand data and the user classification results, and performs demand prediction to obtain the user's specific demand commodity data; the personalized management module customizes the user's personalized shopping needs based on the demand prediction results and the user's acceptance of commodity quality and price, and generates a simplified shopping list; the commodity recommendation generation module generates the user's purchase plan based on the precise shopping list and the commodity recommendation list, including the purchased commodities, purchase quantity, and price information, to assist the one-click shopping function.

[0004] The existing technology uses big data to analyze user behavior, predict product demand trends and push purchase plans, but does not concentrate consumer groups on a certain platform to prioritize the required products. Summary of the invention

[0005] To this end, the present invention provides a commodity demand prediction method for a live broadcast platform, which can solve the problem of commodity demand prediction on the live broadcast platform.

[0006] To achieve the above object, the present invention provides a commodity demand forecasting method for a live broadcast platform, comprising:

[0007] According to the time when the user enters the live broadcast room, identifying the video content being played in the live broadcast room;

[0008] Analyze the video content and extract keywords from the video content according to the analysis results;

[0009] Obtain the popularity of products in the live broadcast room by counting the frequency of product searches by users in the live broadcast room;

[0010] Analyze the completed transactions in the live broadcast room based on the keywords extracted from the video content and the popularity of the products to determine the proportion of purchase reasons, and predict the future transaction volume of the live broadcast products based on the proportion of purchase reasons and the number of products in the live broadcast room;

[0011] The purchase reason of the user who has completed the transaction is recorded to determine the push mode of future live broadcasts.

[0012] Furthermore, the step of identifying the video content played in the live broadcast room includes:

[0013] Use system identification to track each ID that enters the live broadcast room;

[0014] The video content being played when the user enters the live broadcast room is queried based on the ID.

[0015] Furthermore, the step of extracting keywords according to the video content includes:

[0016] Preprocessing the video content, including format conversion and resolution adjustment;

[0017] Based on the preprocessed video content, identifying visual content and audio content in the video content;

[0018] Keywords of the video content are extracted based on the visual content and the audio content.

[0019] Furthermore, the step of extracting keywords of the video content according to the visual content and the audio content includes:

[0020] By using visual recognition and voice recognition to analyze visual content and audio content respectively, words and phrases related to the product are extracted based on the analysis results;

[0021] The repeatedly appearing words and phrases related to the product are used as keywords for the video content.

[0022] Furthermore, the step of obtaining the popularity of the product according to the statistical result includes:

[0023] Collect and record the names of products searched by users in the live broadcast room;

[0024] According to the collected and recorded results, the number of searches for each product is counted;

[0025] According to the statistical results, a ranking algorithm is applied to obtain the popularity of the products.

[0026] Further, the step of analyzing the completed transactions to determine the proportion of purchase reasons includes:

[0027] Calculate the number of purchases of each product based on the keyword and the popularity of the product through an algorithm;

[0028] The ratio of the purchase quantity of the keyword and the popularity of the product to the completed transactions is calculated, and the ratio is used to determine the purchase reason ratio.

[0029] Furthermore, the step of predicting the future transaction volume of the live broadcast commodity based on the proportion of the purchase reasons and the number of commodities in the live broadcast room includes:

[0030] Draw a historical curve based on the proportion of the purchase reasons of the keyword and the popularity of the product, determine the prediction function based on the curve analysis, and build a prediction model based on the prediction function, where the prediction function is a linear regression prediction function in, is the transaction volume of the future live broadcast product, x 1 is the proportion of the keyword, x 2 is the percentage of the popularity of the product, a, b 1 and b 2 and are the unknown regression parameters, calculated using the least squares method right Find partial derivatives, set each partial derivative to zero, and form a system of equations to solve parameters a and b 1 and b 2 , wherein is the actual transaction volume of the live broadcast commodity. Further, the step of determining the push mode of future live broadcast according to the purchase reason of the user who has completed the transaction includes:

[0031] Collect and record users’ purchase data;

[0032] Determine the reasons for the user's purchase based on the purchase data analysis;

[0033] Determine the push mode for future live broadcasts based on the stated purchase reasons and analysis of user usage.

[0034] Furthermore, the steps of determining the push mode of future live broadcast by analyzing the purchase reason of the user and the user's usage include:

[0035] Determine the best time to push by analyzing the peak active hours of users in the live broadcast room;

[0036] By analyzing the viewing habits and preferences of the users in the live broadcast room, a personalized push method is determined, and the push method includes the anchor's oral broadcast and pop-up window reminder;

[0037] Push is performed based on the optimal timing and the personalized push method.

[0038] On the other hand, the present invention also provides a commodity demand forecasting system for a live broadcast platform, the system comprising:

[0039] A video content recognition module, which recognizes the video content being played in the live broadcast room according to the time when the user enters the live broadcast room;

[0040] A keyword extraction module analyzes the video content and extracts keywords from the video content according to the analysis results;

[0041] An acquisition determination module is used to obtain the popularity of products in the live broadcast room by counting the frequency of product searches by users in the live broadcast room;

[0042] A prediction module, which analyzes completed transactions in the live broadcast room based on the keywords extracted from the video content and the popularity of the products to determine the proportion of purchase reasons, and predicts the future transaction volume of the live broadcast products based on the proportion of purchase reasons and the number of products in the live broadcast room;

[0043] The push module records the purchase reason of the user who has completed the transaction to determine the push mode of future live broadcasts. Compared with the prior art, the beneficial effect of the present invention is that by tracking the ID of the user entering the live broadcast room, the video content that initially attracted the user to enter the live broadcast room is identified, the user's preferences can be accurately located and the content strategy can be optimized, and the video content is analyzed, and the keywords of the video content are extracted according to the analysis results. Through the extraction of keywords, the main content and theme of the video can be quickly understood, which is convenient for the platform to quickly grasp the core information of the video; for users who have entered the live broadcast room, by counting the frequency of the user's search for goods, the popularity of the goods is calculated according to the statistical results, and it is understood which goods are more popular, providing data support for market trend analysis, and based on the keywords of the extracted video content and the popularity of the goods, the completed transactions in the live broadcast room are analyzed to determine the proportion of purchase reasons, and the live broadcast content and product display strategy are adjusted according to the keywords and popularity to attract more potential users, and the future transaction volume of the live broadcast goods is predicted based on the proportion of purchase reasons and the number of goods in the live broadcast room; data support is provided for future predictions, helping merchants make more scientific decisions and record the purchase reasons of the users who have completed the transaction to determine the push mode of future live broadcasts, find the goods that users are interested in more quickly, and improve the shopping experience and purchase efficiency of users in the live broadcast room.

[0044] In particular, by tracking the ID of each user entering the live broadcast room and tracking and querying the video content being played when the user enters, the user's preferences can be accurately identified.

[0045] In particular, the video content is preprocessed to provide a basis for visual content and audio content, and scenes and objects in the video are visually identified and converted into potential keywords. Speech recognition is used to convert the speech in the video into text. By combining visual and speech information, a more comprehensive content analysis is provided. By extracting keywords, the content and theme of the video can be quickly understood, thereby grasping the core information of the video.

[0046] In particular, by collecting and recording the results of the product names, the number of searches for each product is counted to gain insight into the market demand for different products. Based on the number of searches, a sorting algorithm is applied to obtain the popularity of the products, which clearly shows the products that are most popular among consumers in the current live broadcast.

[0047] In particular, the algorithm is used to calculate the number of purchases of each product based on the keywords and the popularity of the product, thereby achieving precision marketing, and based on the ratio of the purchase number calculated for the keywords and the popularity of the product to the completed transactions, the ratio of purchase reasons is determined to optimize sales channel allocation.

[0048] In particular, a more accurate prediction model is established by drawing a historical curve based on the proportion of purchase reasons of the keywords and the popularity of the products, and future transaction volumes are predicted based on the prediction model to help merchants make more scientific decisions.

[0049] In particular, the purchase reasons of users who have completed transactions are recorded to determine the push mode for future live broadcasts to improve users' shopping experience and purchasing efficiency in the live broadcast room.

[0050] In particular, analyzing the peak hours of user activity and determining the best time to push content can help increase the exposure of pushed content, improve user engagement and interaction rates, and determine the push method based on user preferences and usage habits, thereby increasing the probability of users noticing and receiving information. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A schematic diagram of the structure of a method for predicting commodity demand for a live broadcast platform provided by an embodiment of the present invention;

[0052] Figure 2 A schematic diagram of the structure of a method for predicting commodity demand on a live broadcast platform for identifying video content played in a live broadcast room provided by an embodiment of the present invention;

[0053] Figure 3A structural schematic diagram of extracting keywords from video content of a commodity demand prediction method for a live broadcast platform provided by an embodiment of the present invention;

[0054] Figure 4 A schematic diagram of a structure for obtaining the popularity of a product from the statistical results of a product demand prediction method for a live broadcast platform provided by an embodiment of the present invention;

[0055] Figure 5 A structural diagram of determining the proportion of purchase reasons in a method for predicting commodity demand on a live broadcast platform provided by an embodiment of the present invention;

[0056] Figure 6 A structural diagram of a method for predicting commodity demand for a live broadcast platform provided in an embodiment of the present invention for determining a push mode for future live broadcasts. DETAILED DESCRIPTION

[0057] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0058] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0059] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be understood as a limitation on the present invention.

[0060] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0061] See also Figure 1 As shown, the present invention provides a commodity demand forecasting method and system for a live broadcast platform, the method comprising:

[0062] Step 100, identifying the video content being played in the live broadcast room according to the time when the user enters the live broadcast room;

[0063] Step 200, analyzing the video content and extracting keywords of the video content according to the analysis result;

[0064] Step 300, obtaining the popularity of the products in the live broadcast room by counting the frequency of product searches by users in the live broadcast room;

[0065] Step 400, analyzing completed transactions in the live broadcast room based on the keywords of the extracted video content and the popularity of the product to determine the proportion of purchase reasons, and predicting the future transaction volume of the live broadcast product based on the proportion of purchase reasons and the number of products in the live broadcast room;

[0066] Step 500, recording the purchase reason of the user who has completed the transaction to determine the push mode of future live broadcasts. Specifically, the embodiment of the present invention can accurately locate user preferences and optimize content strategies by tracking the ID of the user entering the live broadcast room and identifying the video content that initially attracted the user to enter the live broadcast room, and analyze the video content and extract keywords of the video content according to the analysis results. By extracting keywords, the main content and theme of the video can be quickly understood, which is convenient for the platform to quickly grasp the core information of the video; for users who have entered the live broadcast room, by counting the frequency of the user's search for goods, the popularity of the goods is calculated according to the statistical results, and it is understood which goods are more popular, providing data support for market trend analysis, and analyzing the completed transactions in the live broadcast room based on the keywords of the extracted video content and the popularity of the goods to determine the proportion of purchase reasons, and adjusting the live broadcast content and product display strategy according to the keywords and popularity to attract more potential users, and predicting the future transaction volume of the live broadcast goods based on the proportion of purchase reasons and the number of goods in the live broadcast room; providing data support for future predictions, helping merchants make more scientific decisions and record the purchase reasons of the users who have completed the transaction to determine the push mode of future live broadcasts, find the goods that users are interested in more quickly, and improve the shopping experience and purchase efficiency of users in the live broadcast room.

[0067] See also Figure 2 As shown, the step of identifying the video content played in the live broadcast room includes:

[0068] Step S110, using system identification to track each ID entering the live broadcast room;

[0069] Step S120, query the video content being played when the user enters the live broadcast room according to the ID.

[0070] Specifically, this embodiment tracks the ID of each user entering the live broadcast room and tracks and queries the video content being played when the user enters, thereby accurately locating the user's preferences.

[0071] See also Figure 3 As shown, the step of extracting keywords according to the video content includes:

[0072] Step S210, identifying visual content and audio content in the video content based on the preprocessed video content;

[0073] Step S220, identifying visual content and audio content in the video content based on the preprocessed video content;

[0074] Step S230: extracting keywords of the video content based on the visual content and the audio content.

[0075] Specifically, the step of extracting keywords of video content according to visual content and audio content in the embodiment of the present invention includes:

[0076] By using visual recognition and voice recognition to analyze visual content and audio content respectively, words and phrases related to the product are extracted based on the analysis results;

[0077] The repeatedly appearing words and phrases related to the product are used as keywords for the video content.

[0078] Specifically, the embodiment of the present invention pre-processes the video content to provide a basis for visual content and audio content, and visually identifies scenes and objects in the video and converts them into potential keywords. It uses speech recognition to convert the speech in the video into text, combines visual and speech information, and provides a more comprehensive content analysis. By extracting keywords, the content and theme of the video can be quickly understood, thereby grasping the core information of the video.

[0079] See also Figure 4 As shown, the steps of obtaining the popularity of a product according to the statistical results include:

[0080] Step S310, collecting and recording the names of the products searched by the users in the live broadcast room;

[0081] Step S330, counting the number of searches for each product based on the collected and recorded results;

[0082] Step S340: applying a ranking algorithm to obtain the popularity of the products according to the statistical results.

[0083] Specifically, the embodiment of the present invention collects and records the results of the names of commodities, counts the number of searches for each commodity, and gains insight into the market demand for different commodities. Based on the number of searches, a sorting algorithm is applied to obtain the popularity of the commodities, thereby clearly displaying the commodities that are most popular among consumers in the current live broadcast.

[0084] See also Figure 5 As shown, the steps of analyzing the completed transactions to determine the proportion of purchase reasons include:

[0085] Step S410, calculating the purchase quantity of each product according to the keyword and the popularity of the product through an algorithm;

[0086] Step S420, calculating the ratio of the purchase quantity of the keyword and the popularity of the product to the completed transactions, wherein the ratio is used to determine the purchase reason ratio.

[0087] Specifically, the embodiment of the present invention calculates the number of purchases of each product based on the keywords and the popularity of the product through an algorithm to achieve precision marketing, and determines the proportion of purchase reasons based on the calculated number of purchases of the keywords and the popularity of the product and the proportion of completed transactions, thereby optimizing sales channel allocation.

[0088] Specifically, the step of predicting the future transaction volume of the live broadcast commodity based on the proportion of the purchase reasons and the number of commodities in the live broadcast room according to the embodiment of the present invention includes:

[0089] Draw a historical curve based on the proportion of the purchase reasons of the keyword and the popularity of the product, determine the prediction function based on the curve analysis, and build a prediction model based on the prediction function, where the prediction function is a linear regression prediction function in, is the transaction volume of the future live broadcast product, b 1 is the percentage of the keyword, b 2 is the percentage of the popularity of the product, a, b 1 and b 2 is the unknown regression parameter, and is calculated using the least squares method right Find partial derivatives, set each partial derivative to zero to form a system of equations to solve parameters a and b 1 and b 2 , where is the actual transaction volume of live-streamed products.

[0090] Specifically, the embodiment of the present invention draws a historical curve based on the proportion of purchase reasons of the keywords and the popularity of the products to establish a more accurate prediction model, and predicts future transaction volumes based on the prediction model to help merchants make more scientific decisions.

[0091] See also Figure 6 As shown, the step of determining the push mode of future live broadcast according to recording the purchase reason of the user who has completed the transaction includes:

[0092] Step S510, collecting and recording the user's purchase data;

[0093] Step S520, analyzing the purchase data to determine the reasons for the user's purchase;

[0094] Step S530, determining the push mode for future live broadcasts based on the purchase reason and analyzing the user's usage.

[0095] Specifically, the embodiment of the present invention determines the push mode of future live broadcasts based on the purchase reasons of the users who have completed the transactions, and improves the shopping experience and purchase efficiency of the users in the live broadcast room.

[0096] In this embodiment, the steps of analyzing the purchase reason of the user and the user's usage to determine the push mode of future live broadcast include:

[0097] Determine the best time to push by analyzing the peak active hours of users in the live broadcast room;

[0098] By analyzing the viewing habits and preferences of the users in the live broadcast room, a personalized push method is determined, and the push method includes the anchor's oral broadcast and pop-up window reminder;

[0099] Push is performed based on the optimal timing and the personalized push method.

[0100] Specifically, the embodiment of the present invention analyzes the peak active hours of users and determines the best time for push, which helps to increase the exposure rate of pushed content, improve user participation and interaction rate, and determine the push method based on the user's preferred push method and usage habits, thereby increasing the probability of the information being noticed and received by the user.

[0101] The system in the embodiment of the present invention is applied in the commodity demand prediction method of the above-mentioned live broadcast platform, and can achieve the same technical effect, which will not be repeated here.

[0102] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0103] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A commodity demand forecasting method for a live broadcast platform, characterized in that: include According to the time when the user enters the live broadcast room, identifying the video content being played in the live broadcast room; Analyze the video content and extract keywords from the video content according to the analysis results; Obtain the popularity of products in the live broadcast room by counting the frequency of product searches by users in the live broadcast room; Analyze the completed transactions in the live broadcast room based on the keywords extracted from the video content and the popularity of the products to determine the proportion of purchase reasons, and predict the future transaction volume of the live broadcast products based on the proportion of purchase reasons and the number of products in the live broadcast room; The purchase reason of the user who has completed the transaction is recorded to determine the push mode of future live broadcasts.

2. The commodity demand forecasting method for a live broadcast platform according to claim 1, characterized in that: The step of identifying the video content played in the live broadcast room includes: Use system identification to track each ID that enters the live broadcast room; The video content being played when the user enters the live broadcast room is queried based on the ID.

3. The commodity demand forecasting method for a live broadcast platform according to claim 2, characterized in that: The step of extracting keywords according to the video content comprises: Preprocessing the video content, including format conversion and resolution adjustment; Based on the preprocessed video content, identifying visual content and audio content in the video content; Keywords of the video content are extracted based on the visual content and the audio content.

4. The commodity demand forecasting method for a live broadcast platform according to claim 3, characterized in that: The step of extracting keywords of the video content according to the visual content and the audio content comprises: By using visual recognition and voice recognition to analyze visual content and audio content respectively, words and phrases related to the product are extracted based on the analysis results; The repeatedly appearing words and phrases related to the product are used as keywords for the video content.

5. The commodity demand forecasting method for a live broadcast platform according to claim 4, characterized in that: The step of obtaining the popularity of the product according to the statistical results includes: Collect and record the names of products searched by users in the live broadcast room; According to the collected and recorded results, the number of searches for each product is counted; According to the statistical results, a ranking algorithm is applied to obtain the popularity of the products.

6. The commodity demand forecasting method for a live broadcast platform according to claim 5, characterized in that: The steps for analyzing the completed transactions to determine the percentage of purchase reasons include: Calculate the number of purchases of each product based on the keyword and the popularity of the product through an algorithm; The ratio of the purchase quantity of the keyword and the popularity of the product to the completed transactions is calculated, and the ratio is used to determine the purchase reason ratio.

7. The commodity demand forecasting method for a live broadcast platform according to claim 6, characterized in that: The step of predicting the future transaction volume of the live broadcast commodity based on the proportion of the purchase reasons and the number of commodities in the live broadcast room includes: Draw a historical curve based on the proportion of the purchase reasons of the keyword and the popularity of the product, determine the prediction function based on the curve analysis, and build a prediction model based on the prediction function, where the prediction function is a linear regression prediction function in, is the transaction volume of the future live broadcast product, x1 is the proportion of the keyword, x2 is the proportion of the popularity of the product, a, b1 and b2 are the undetermined regression parameters, and are calculated using the least squares method. ,right Find the partial derivatives and set each partial derivative to zero to form a system of equations to solve the parameters a, b1 and b2, where y is the actual transaction volume of the live broadcast product.

8. The commodity demand forecasting method for a live broadcast platform according to claim 7, characterized in that: The step of determining the push mode of future live broadcast according to the purchase reason of the user who has completed the transaction includes: Collect and record users’ purchase data; Determine the reasons for the user's purchase based on the purchase data analysis; Determine the push mode for future live broadcasts based on the stated purchase reasons and analysis of user usage.

9. The commodity demand forecasting method for a live broadcast platform according to claim 8, characterized in that: The steps of analyzing the purchase reasons of the user and the user's usage to determine the push mode of future live broadcasts include: Determine the best time to push by analyzing the peak active hours of users in the live broadcast room; By analyzing the viewing habits and preferences of the users in the live broadcast room, a personalized push method is determined, and the push method includes the anchor's oral broadcast and pop-up window reminder; Push is performed based on the optimal timing and the personalized push method.

10. A system based on the commodity demand forecasting method for a live broadcast platform according to any one of claims 1 to 9, characterized in that: include: A video content recognition module, which recognizes the video content being played in the live broadcast room according to the time when the user enters the live broadcast room; A keyword extraction module analyzes the video content and extracts keywords from the video content according to the analysis results; An acquisition determination module is used to obtain the popularity of products in the live broadcast room by counting the frequency of product searches by users in the live broadcast room; A prediction module, which analyzes completed transactions in the live broadcast room based on the keywords extracted from the video content and the popularity of the products to determine the proportion of purchase reasons, and predicts the future transaction volume of the live broadcast products based on the proportion of purchase reasons and the number of products in the live broadcast room; The push module records the purchase reason of the user who has completed the transaction to determine the push mode of future live broadcasts.

Images