Information push method and system based on live transaction order data

Abstract

The present invention discloses an information push method and device based on live transaction order data, which relates to the field of data processing technology, and the main purpose is to solve the problem of poor effectiveness of information push of existing live transaction order data. It includes: obtaining live transaction order data to be paid and marked commodity information of target users; determining the intent category of the marked commodity information according to the order commodity category in the live transaction order data, and counting the number of marks matching different intent categories in the marked commodity information; determining the correlation coefficient between the order commodity category and the intent category, and searching for push information matching the order commodity in the live transaction order data according to the marked commodity information, the number of marks, and the correlation coefficient; if the push information is searched, the push information is pushed in the page to be paid.

Description

Technical Field

[0001] The present invention relates to a field of data processing technology, and in particular to an information push method and system based on live broadcast transaction order data. Background Art

[0002] With the rapid development of live broadcast platforms, the live broadcast process is usually associated with transaction links of various commodities to achieve the purpose of online sales. At present, when a live broadcast transaction order is generated, the generated order information is usually sent directly to the user, and the user is immediately returned to the live broadcast room to participate in the introduction of the next commodity. However, there is usually no correlation between the commodities traded live in the live broadcast room. For example, when a user places an order for hand sanitizer, the live broadcast room will broadcast the introduction of tableware commodities. At this time, if the user chooses other hand sanitizers later, the ordered commodities need to be canceled, which greatly reduces the effectiveness of order generation. Sometimes, even after the commodity has generated logistics consumption, it still needs to be canceled, which greatly increases the consumption of manpower and material resources. Therefore, there is an urgent need for an information push method based on live broadcast transaction order data to solve the above problems. Summary of the invention

[0003] In view of this, the present invention provides an information push method and device based on live broadcast transaction order data, the main purpose of which is to solve the problem of poor effectiveness of existing information push of live broadcast transaction order data.

[0004] According to one aspect of the present invention, a method for pushing information based on live broadcast transaction order data is provided, comprising:

[0005] Obtain the live broadcast transaction order data to be paid and the target user's marked product information;

[0006] Determine the intent category of the marked product information according to the order product category in the live transaction order data, and count the number of times the marked product information matches different intent categories;

[0007] Determine the correlation coefficient between the order product category and the intention category, and search for push information matching the order product in the live transaction order data according to the marked product information, the marking times, and the correlation coefficient;

[0008] If the push information is found, the push information is pushed on the payment page.

[0009] According to another aspect of the present invention, there is provided an information push device based on live transaction order data, comprising:

[0010] The acquisition module is used to obtain the live broadcast transaction order data to be paid and the marked product information of the target user;

[0011] A determination module, used to determine the intent category of the marked product information according to the order product category in the live transaction order data, and to count the number of marks matching different intent categories in the marked product information;

[0012] A matching module, used to determine the correlation coefficient between the order product category and the intention category, and search for push information of the order product in the live transaction order data that matches the marked product information, the marking times, and the correlation coefficient;

[0013] The push module is used to push the push information on the payment page if the push information is found.

[0014] According to another aspect of the present invention, a storage medium is provided, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables a processor to execute operations corresponding to the above-mentioned information push method based on live transaction order data.

[0015] According to another aspect of the present invention, there is provided a terminal, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0016] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the above-mentioned information push method based on live transaction order data.

[0017] By means of the above technical solution, the technical solution provided by the embodiment of the present invention has at least the following advantages:

[0018] The present invention provides an information push method and device based on live transaction order data. Compared with the prior art, the embodiment of the present invention obtains the live transaction order data to be paid and the marked product information of the target user; determines the intention category of the marked product information according to the order product category in the live transaction order data, and counts the number of marks matching different intention categories in the marked product information; determines the correlation coefficient between the order product category and the intention category, and searches for push information matching the order product in the live transaction order data according to the marked product information, the number of marks, and the correlation coefficient; if the push information is searched, the push information is pushed in the page to be paid, so as to achieve the purpose of information push during the live broadcast process, greatly reduce the resource waste of the live broadcast order, thereby achieving the purpose of improving order transactions based on push information, so as to improve the effective information push effect.

[0019] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0021] Figure 1 A flow chart of an information push method based on live broadcast transaction order data provided by an embodiment of the present invention is shown;

[0022] Figure 2 A block diagram showing a composition of an information push device based on live broadcast transaction order data provided by an embodiment of the present invention;

[0023] Figure 3 A schematic diagram of the structure of a terminal provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0024] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0025] The embodiment of the present invention provides a method for pushing information based on live transaction order data, such as Figure 1 As shown, the method includes:

[0026] 101. Obtain the live broadcast transaction order data to be paid and the marked product information of the target user.

[0027] In the embodiment of the present invention, the current execution end, as the server end of the live broadcast platform, can obtain the specific live broadcast transaction order data generated in the live broadcast room in real time. At this time, the live broadcast transaction order data includes live broadcast product information, transaction amount, order time and other information. This live broadcast transaction order information is unpaid order information. Therefore, the current execution end only needs to obtain the marked product information for the target user corresponding to the unpaid live broadcast transaction order data. Among them, the marked product information is used to represent the products that are marked after being liked or followed by the user under the target user's account, including but not limited to shopping cart marked product information, favorite marked product information and browsed marked product information, which is not specifically limited in the embodiment of the present invention.

[0028] In another embodiment of the present invention, for further definition and explanation, the step of obtaining the live broadcast transaction order data to be paid and the marked product information of the target user includes:

[0029] Obtain the live broadcast push permission of the target user;

[0030] When it is detected that live broadcast transaction order data is generated and the payment operation of the live broadcast transaction order data has not been completed, the marked product information of the target user is obtained according to the live broadcast push authority.

[0031] In order to ensure the effectiveness of obtaining the marked product information and ensure the security of information acquisition, the current execution end first obtains the live push permission of the target user, and judges whether the information push of the live transaction order data can be executed according to the live push permission. Among them, the live push permission is used to represent the permission to access the account information of the target user. The live push permission includes access permission 1 and no access permission 0. When the current execution end determines that it has access permission 1, it obtains the generated live transaction order data and detects whether the payment operation of the live transaction order data has been completed. When it is detected that the live transaction order data is generated and the payment operation of the live transaction order data has not been completed, the marked product information of the target user is obtained according to the live push permission. At this time, the marked product information includes shopping cart marked product information, favorite marked product information and browsed marked product information. After having access permission, the current execution end obtains the product information marked as shopping cart products, the product information marked as favorite products and the product information marked as browsed products from the target user's account information. The embodiment of the present invention does not make specific limitations.

[0032] It should be noted that the live push permission can be a permission configuration instruction issued to the target user when the target user enters the live broadcast room. The target user configures this live push permission in the initial interface of entering the live broadcast room, and can choose to configure it as 1 or 0, so that the current execution end confirms whether to broadcast steps 101-104. The embodiment of the present invention does not specifically limit it. If it is configured as not having access right 0, the information push method in the embodiment of the present invention can be directly skipped, and the live broadcast room can be directly entered.

[0033] 102. Determine the intent category of the marked product information according to the order product category in the live transaction order data, and count the number of marks matching different intent categories in the marked product information.

[0034] In an embodiment of the present invention, the order commodity categories include but are not limited to household appliances, daily necessities, clothing and shoes, luggage, etc. After determining the marked commodity information, the current execution end determines the intent category of the marked commodity information according to the order commodity category in the live transaction order data. At this time, the order commodity category in the live transaction order data can be extracted based on the commodity information, which is not specifically limited in the embodiment of the present invention. Among them, the intent category is used to characterize the purchase demand intention of the target user when purchasing the commodity, for example, household cleaning intention, beauty intention, home decoration intention, etc., which is not specifically limited in the embodiment of the present invention. Furthermore, after the current execution end determines the intent category, it counts the number of marks that match different intent categories in the marked commodity information. At this time, since 2 or 4 intent categories may be determined in 100 marked commodity information, at this time, the number of marks of different intent categories, such as how many times the commodities corresponding to 2 intent categories are marked in total, is not specifically limited in the embodiment of the present invention.

[0035] In another embodiment of the present invention, for further definition and explanation, the step of determining the intended category of the marked product information according to the order product category in the live transaction order data includes:

[0036] Extracting order keywords from the live transaction order data, and classifying the order keywords based on the trained natural language classification model to obtain order commodity categories;

[0037] The intended category of the marked product information is determined based on the order product category and a classification threshold.

[0038] In order to improve the accuracy of identifying the intent category, thereby improving the accuracy of identifying the push information, in a specific implementation scenario, when the current execution end determines the intent category, it first extracts order keywords from the live transaction order data for classification. Among them, the order keywords can be extracted based on the natural semantic division of the product name to obtain keywords in the form of names. At this time, the natural semantic division can be based on the noun vocabulary or the natural semantic processing model, which is not specifically limited in the embodiment of the present invention. After the order keywords are obtained, the order keywords are classified based on the trained natural language classification model to obtain the order product category. At this time, the natural language classification model is obtained by training the textCNN model based on the full amount of natural language training samples with classification marks. The classification mark can be a classification label marked according to the order product category, which can be marked manually or by a marking algorithm. The embodiment of the present invention does not make specific limitations. Furthermore, the intent category of the marked product information is determined based on the obtained order product category and the classification threshold. At this time, the classification threshold is used to characterize the correlation between the classification of different order product categories, thereby determining the intent category. The intent category is used to characterize the purchase demand intention of the target user when purchasing the product. The embodiment of the present invention does not make specific limitations.

[0039] It should be noted that the classification threshold can be configured based on the similarity of the meaning of words between different order product categories, and the configuration range is 0-1. For example, the classification threshold of cleaning products and decontamination products is configured as 0.8 according to the semantic meaning similarity, which is not specifically limited in the embodiment of the present invention. In addition, the semantic meaning similarity can be calculated based on the semantic vectors of different order product categories, and configured based on the mapping relationship between the similarity value and the threshold, which is not specifically limited in the embodiment of the present invention.

[0040] In another embodiment of the present invention, for further definition and explanation, the step of determining the intended category of the marked product information based on the order product category and the classification threshold includes:

[0041] Calculating a classification confidence value corresponding to the order product category and the classification threshold by using a summation relationship;

[0042] Based on a preset intention-classification mapping relationship, an intention category matching the product tag information and the classification confidence value is queried.

[0043] In order to accurately identify the intent category and determine the recommended content for the order based on the intent category, in a specific implementation scenario, when the current execution end determines the intent category based on the order product category and the classification threshold, the current execution end first uses the sum relationship to calculate the classification confidence value corresponding to the order product category and the classification threshold. Among them, the classification confidence value is used to characterize the degree of trust under the matching order product category, and the sum relationship is used to add the order product category to the classification threshold to obtain the classification confidence value. Since the classification threshold is obtained based on the similarity between the order product categories, the formula corresponding to the sum relationship is: , is the order product category coefficient, is the number of product categories that are similar to the order product category, that is, the number of other order product categories used to calculate the classification threshold. is other order commodity category coefficients. In the embodiment of the present invention, the order commodity category coefficient is a numerical representation of the order commodity category, and the embodiment of the present invention does not make any specific limitation.

[0044] Furthermore, after the classification confidence value is calculated, the preset intention classification mapping relationship queries the intent category that matches the product tag information and the classification confidence value, wherein the preset intention classification mapping relationship stores the intent classifications corresponding to different tagged product information and different classification confidence values, thereby determining the intent category. At this time, the intent classifications corresponding to different tagged product information and different classification confidence values ​​can be configured based on the needs for intent classification, and the embodiments of the present invention do not make specific limitations.

[0045] 103. Determine the correlation coefficient between the order product category and the intention category, and search for push information of the order product in the live transaction order data that matches the marked product information, the marking times, and the correlation coefficient.

[0046] In the embodiment of the present invention, after determining the number of markings, the current execution end determines the correlation coefficient between the order product category and the intent category. At this time, the correlation coefficient is used to characterize the probability value of whether the order product category and the intent category are the same, so as to combine the marked product information, the number of markings and the determined correlation coefficient to search for the push information corresponding to the order product that can be pushed to the target user. Among them, the push information includes but is not limited to product information, description information, marketing information, etc. The push information is pre-configured in the current execution end, and the embodiment of the present invention does not make specific limitations.

[0047] In another embodiment of the present invention, for further definition and explanation, the step of determining the correlation coefficient between the order product category and the intention category, and searching for push information matching the order product in the live transaction order data according to the marked product information, the marking times, and the correlation coefficient includes:

[0048] Performing semantic analysis on the order product category and the intention category to determine a semantic vector matrix, and calculating the vector similarity between the semantic vector matrices to determine the correlation coefficient;

[0049] Searching for matched reference push information based on the marked product information, the reference push information including reference product information, reference description information, and reference marketing information;

[0050] The push evaluation value of the reference push information is calculated according to the number of markings and the correlation coefficient, and the push evaluation values ​​are sorted to determine the push information matching the order product.

[0051] In order to accurately screen the information that can be pushed to the user, thereby improving the efficiency of information push based on live transaction order data, when the current execution end determines the push information, specifically, first semantically parse the order product category and the intent category, determine the semantic vector matrix respectively, and calculate the vector similarity between the semantic vector matrices to determine the correlation coefficient. At this time, the order product category and the intent category can be the same or different. If they are the same, the calculated vector similarity is 1. If they are different, the similarity calculation is based on the corresponding semantic vector matrices. The embodiment of the present invention does not make specific limitations. After determining the correlation coefficient, the current execution end can search for matching reference push information in the push information library based on the marked product information. At this time, the reference push information includes reference product information, reference description information, and reference marketing information. The product information is used to represent information such as the product name or product image, and the description information is used to represent the parameters, evaluations, and other information of the order product. The marketing information is used to represent marketing strategy information associated with the order product, such as discounts, red envelope activities, and other information. The embodiment of the present invention does not make specific limitations. In addition, when searching for reference push information, matching reference push information can be searched based on the name, keywords, parameter information and other contents in the marked product information. The push information library stores the search relationship between different reference push information and different marked product information for effective query.

[0052] Furthermore, after the current execution end determines the reference push information, it calculates the push evaluation value of the reference push information according to the number of marks and the correlation coefficient, and sorts according to the push evaluation value to determine the push information of the matching order product. Among them, the push evaluation value is calculated based on the product of the number of marks and the correlation coefficient divided by the push value of the reference push information, and all the reference push information are sorted in turn, and a preset number of push information is selected according to the sorting size, which is not specifically limited in the embodiment of the present invention. In addition, the push value is a numerical value pre-configured for each reference push information to represent the priority of pushing to the target user, which is configured as 0-1, and 1 indicates the best recommendation, which is not specifically limited in the embodiment of the present invention.

[0053] 104. If the push information is found, the push information is pushed on the payment page.

[0054] In an embodiment of the present invention, after the current execution end is able to search for the push information, since the order corresponding to the live transaction order data has not been paid, the current execution end will display the push information on the payment page so that the target user can jump to other pages of interest in time to place an order, thereby reducing the waste of transaction resources for ordering goods in the live broadcast room.

[0055] In another embodiment of the present invention, for further definition and explanation, the step of pushing the push information in the payment page includes:

[0056] When generating a payment webpage for the live transaction order data, determining a push method for the push information;

[0057] The push information is displayed in the payment webpage according to the push method.

[0058] In order to improve the effectiveness of pushing push information and reduce the resource consumption of live order transactions, the current execution end first detects whether to generate a payment webpage when pushing, and determines the push method of the push information while generating the payment webpage. At this time, the push method includes at least one of image push, video push, voice push and link push. Different push methods can be selected according to different push information. For example, when the push information is a red envelope, the push method is connection push, which is not specifically limited in the embodiment of the present invention. Furthermore, the current execution end can display the push information in the payment webpage according to the push method, which is not specifically limited in the embodiment of the present invention.

[0059] In another embodiment of the present invention, for further limitation and explanation, after the step of pushing the push information in the payment page, the method further includes:

[0060] When it is detected that the order transaction is completed through the push information, other users who are similar to the target user are searched, and the push information is pushed to the other users.

[0061] In order to push the push information effectively and flexibly, after completing the push information, the current execution end detects whether the target user completes a new order transaction based on the push information. If a new order transaction is completed based on the push information, it means that the target user is more inclined to choose the push information rather than directly placing an order transaction through the live broadcast room. At this time, the current execution end searches for other users who are similar to the target user, and recommends this push information to other users. Among them, users with similarities can be users who have placed orders for the same live broadcast product in the live broadcast room but have not paid, so that the push information is more targeted, thereby improving the push effect of the push information.

[0062] In another embodiment of the present invention, for further definition and explanation, the step of searching for other users having similarities with the target user includes:

[0063] Obtain all user information that accesses the same live broadcast room, retrieve all transaction orders that have been ordered but not paid, and determine the reference user. The current execution end then obtains the live broadcast push permission of the reference user. At this time, if the live broadcast push permission has account information access permission, the current execution end retrieves the transaction goods that the reference user has ordered, and calculates the commodity category similarity between the transaction goods and the marked commodity information. At the same time, according to the shopping habit information between the reference user and the target user, a weight coefficient is configured for the commodity category similarity, and the similarity and the weight coefficient are summed and calculated. If the total similarity value obtained is greater than the preset similarity threshold, it means that this reference user is another user with similarity. The preset similarity threshold is not specifically limited in the embodiment of the present invention. In addition, the corresponding shopping habit information includes order payment speed, order payment type, payment commodity type, etc., and then the commodity category similarity configuration weight coefficient corresponding to the shopping habit information is determined according to the preset habit similarity mapping relationship. The habit similarity mapping relationship includes weight coefficients corresponding to the similarity of different shopping habit information and different commodity categories. This mapping habit can be configured based on the diversity requirements of the commodity types traded in the live broadcast room, and the embodiment of the present invention does not make specific limitations.

[0064] The embodiment of the present invention provides an information push method based on live transaction order data. Compared with the prior art, the embodiment of the present invention obtains the live transaction order data to be paid and the marked product information of the target user; determines the intention category of the marked product information according to the order product category in the live transaction order data, and counts the number of marks matching different intention categories in the marked product information; determines the correlation coefficient between the order product category and the intention category, and searches for push information matching the order product in the live transaction order data according to the marked product information, the number of marks, and the correlation coefficient; if the push information is searched, the push information is pushed in the page to be paid, thereby achieving the purpose of information push during the live broadcast process, greatly reducing the waste of resources of the live broadcast order, thereby achieving the purpose of improving order transactions based on push information, and improving the effective information push effect.

[0065] Furthermore, as a response to the above Figure 1 The implementation of the method shown in the figure, the embodiment of the present invention provides an information push device based on live transaction order data, such as Figure 2 As shown, the device comprises:

[0066] The acquisition module 21 is used to acquire the live broadcast transaction order data to be paid and the marked product information of the target user;

[0067] A determination module 22, configured to determine the intent category of the marked product information according to the order product category in the live transaction order data, and to count the number of marks in the marked product information that match different intent categories;

[0068] A matching module 23, used to determine the correlation coefficient between the order product category and the intention category, and search for push information of the order product in the live transaction order data that matches the marked product information, the marking times, and the correlation coefficient;

[0069] The push module 24 is used to push the push information in the payment page if the push information is found.

[0070] The embodiment of the present invention provides an information push device based on live transaction order data. Compared with the prior art, the embodiment of the present invention obtains the live transaction order data to be paid and the marked product information of the target user; determines the intention category of the marked product information according to the order product category in the live transaction order data, and counts the number of marks matching different intention categories in the marked product information; determines the correlation coefficient between the order product category and the intention category, and searches for push information matching the order product in the live transaction order data according to the marked product information, the number of marks, and the correlation coefficient; if the push information is searched, the push information is pushed in the page to be paid, thereby achieving the purpose of information push during the live broadcast process, greatly reducing the waste of resources of the live broadcast order, thereby achieving the purpose of improving order transactions based on push information, and improving the effective information push effect.

[0071] According to one embodiment of the present invention, a storage medium is provided, wherein the storage medium stores at least one executable instruction, and the computer executable instruction can execute the information push method based on live transaction order data in any of the above method embodiments.

[0072] Figure 3 A schematic diagram of the structure of a terminal provided according to an embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the terminal.

[0073] like Figure 3 As shown, the terminal may include: a processor (processor) 302 , a communication interface (Communications Interface) 304 , a memory (memory) 306 , and a communication bus 308 .

[0074] The processor 302 , the communication interface 304 , and the memory 306 communicate with each other via the communication bus 308 .

[0075] The communication interface 304 is used to communicate with other devices such as clients or other servers.

[0076] Processor 302 is used to execute program 310, and specifically can execute relevant steps in the above-mentioned information push method embodiment based on live transaction order data.

[0077] Specifically, the program 310 may include program codes, which include computer operation instructions.

[0078] The processor 302 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiment of the present invention. The one or more processors included in the terminal may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0079] The memory 306 is used to store the program 310. The memory 306 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0080] The program 310 may be specifically configured to enable the processor 302 to perform the following operations:

[0081] Obtain the live broadcast transaction order data to be paid and the target user's marked product information;

[0082] Determine the intent category of the marked product information according to the order product category in the live transaction order data, and count the number of times the marked product information matches different intent categories;

[0083] Determine the correlation coefficient between the order product category and the intention category, and search for push information matching the order product in the live transaction order data according to the marked product information, the marking times, and the correlation coefficient;

[0084] If the push information is found, the push information is pushed on the payment page.

[0085] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0086] 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 method for pushing information based on live transaction order data, characterized in that: include: Obtain the live broadcast transaction order data to be paid and the target user's marked product information; Determine the intent category of the marked product information according to the order product category in the live transaction order data, and count the number of times the marked product information matches different intent categories; Determine the correlation coefficient between the order product category and the intention category, and search for push information matching the order product in the live transaction order data according to the marked product information, the marking times, and the correlation coefficient; If the push information is found, the push information is pushed on the payment page; The determining of the intended category of the marked commodity information according to the order commodity category in the live transaction order data comprises: Extracting order keywords from the live transaction order data, and classifying the order keywords based on the trained natural language classification model to obtain order commodity categories; Determining the intended category of the marked product information based on the order product category and a classification threshold, wherein the classification threshold is used to characterize the degree of association between classifications of different order product categories; The determining the intended category of the marked product information based on the order product category and the classification threshold comprises: Calculating a classification confidence value corresponding to the order product category and the classification threshold by using a summation relationship; Querying the intent category matching the marked product information and the classification confidence value based on a preset intent classification mapping relationship, wherein the preset intent classification mapping relationship stores intent classifications corresponding to different marked product information and different classification confidence values; The determining of the correlation coefficient between the order product category and the intention category, and searching for push information matching the order product in the live transaction order data according to the marked product information, the marking times, and the correlation coefficient includes: Performing semantic analysis on the order product category and the intention category to determine a semantic vector matrix, and calculating the vector similarity between the semantic vector matrices to determine the correlation coefficient; Searching for matched reference push information based on the marked product information, the reference push information including reference product information, reference description information, and reference marketing information; The push evaluation value of the reference push information is calculated according to the number of markings and the correlation coefficient, and the push evaluation values ​​are sorted to determine the push information matching the order product.

2. The method according to claim 1, characterized in that The step of obtaining the live broadcast transaction order data to be paid and the marked product information of the target user includes: Acquire the live broadcast push permission of the target user, where the live broadcast push permission is used to represent the permission to access the account information of the target user; When it is detected that live transaction order data is generated and the payment operation of the live transaction order data has not been completed, the marked product information of the target user is obtained according to the live broadcast push authority, and the marked product information includes shopping cart marked product information, favorite marked product information and browsed marked product information.

3. The method according to claim 1, characterized in that: The pushing of the push information in the payment page includes: When generating a payment webpage for the live transaction order data, determining a push method for the push information, wherein the push method includes at least one of image push, video push, voice push, and link push; The push information is displayed in the payment webpage according to the push method.

4. The method according to claim 1, characterized in that: After pushing the push information in the payment page, the method further includes: When it is detected that the order transaction is completed through the push information, other users who are similar to the target user are searched, and the push information is pushed to the other users.

5. An information push system based on live transaction order data, characterized in that: include: The acquisition module is used to obtain the live broadcast transaction order data to be paid and the marked product information of the target user; A determination module, used to determine the intent category of the marked product information according to the order product category in the live transaction order data, and to count the number of marks matching different intent categories in the marked product information; A matching module, used to determine the correlation coefficient between the order product category and the intention category, and search for push information of the order product in the live transaction order data that matches the marked product information, the marking times, and the correlation coefficient; A push module, for pushing the push information on the payment page if the push information is found; The determination module is specifically used to extract order keywords from the live transaction order data, and classify the order keywords based on the trained natural language classification model to obtain order commodity categories; Determining the intended category of the marked product information based on the order product category and a classification threshold, wherein the classification threshold is used to characterize the degree of association between classifications of different order product categories; The determination module is further configured to calculate a classification confidence value corresponding to the order commodity category and the classification threshold by using a summation relationship; Querying the intent category matching the marked product information and the classification confidence value based on a preset intent classification mapping relationship, wherein the preset intent classification mapping relationship stores intent classifications corresponding to different marked product information and different classification confidence values; The matching module is used to perform semantic analysis on the order product category and the intention category to determine a semantic vector matrix, and calculate the vector similarity between the semantic vector matrices to determine the correlation coefficient; Based on the marked product information, matching reference push information is searched, and the reference push information includes reference product information, reference description information, and reference marketing information; a push evaluation value of the reference push information is calculated according to the marking times and the correlation coefficient, and the information is sorted according to the push evaluation value to determine the push information that matches the order product.

6. A storage medium storing at least one executable instruction, wherein the executable instruction enables a processor to execute operations corresponding to the information push method based on live transaction order data as described in any one of claims 1-4.

7. A terminal, comprising: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the information push method based on live transaction order data as described in any one of claims 1-4.

Images