Method and system for globally associating commodity information in multi-round dialogue process

Abstract

The application provides a commodity information global association method and system in a multi-round dialogue process, belongs to the technical field of data processing under the internet industry, is applied to a first terminal device of an intelligent customer service system of a merchant user of an e-commerce platform, and the method comprises the following steps: displaying first dialogue information input by a first consumer user in a first customer service interface of the merchant user for the first consumer user, and the first customer service interface is associated with a plurality of historical dialogue information of the first consumer user; sending the first dialogue information to a server, the first dialogue information is associated with second dialogue information, the second dialogue information comprises reply information of an inquiry matter for a target commodity, and the target commodity is a first commodity in a first type of commodities or a second commodity in a second type of commodities; receiving the second dialogue information from the server; and displaying the second dialogue information. The accuracy, comprehensiveness and fluency of the customer service system in processing dialogue information of the consumer user can be improved.

Description

Technical Field

[0001] This application belongs to the field of data processing technology in the Internet industry, specifically involving a method and system for globally associating product information during multi-turn dialogue. Background Technology

[0002] Currently, the intelligent customer service systems on e-commerce platforms typically involve a chatbot initially interacting with the user, answering standardized, structured questions. However, in practice, user interaction with the platform usually involves multiple rounds of dialogue. For example, when inquiring about products, a user might ask several questions and hope the platform can remember previous conversations, especially previously recommended products, so that they can continue discussing or searching for related items in subsequent conversations.

[0003] While current AI-powered customer service systems can identify the intent of some user questions and predict answers based on previous chat history, they cannot provide accurate answers for complex historical chat information or user questions that include references to related products from previous conversations. Even after transferring to a human customer service representative, it takes a considerable amount of time to review previous chat history and perform background checks to provide a targeted answer to the user's needs. The accuracy and timeliness of these responses are difficult to meet user requirements. Summary of the Invention

[0004] This application provides a method and system for globally associating product information during multi-turn dialogues, in order to improve the intelligence and applicability of the dialogue function of e-commerce machine customer service, while accurately identifying the intent of dialogue questions that include fuzzy references to products associated with historical dialogues and inquiries, thereby improving the accuracy, comprehensiveness and fluency of the customer service system in processing consumer dialogue information.

[0005] In a first aspect, embodiments of this application provide a method for globally associating product information during multi-turn dialogues, applied to a first terminal device of an intelligent customer service system for merchant users on an e-commerce platform, the method comprising:

[0006] The merchant user's first customer service interface for the first consumer displays the first dialogue information entered by the first consumer. The first customer service interface is associated with multiple historical dialogues of the first consumer. The multiple historical dialogues include first historical dialogues for a first type of product and second historical dialogues for a second type of product. The first type of product represented by the first historical dialogue has been in a transaction state, and the transaction state includes at least one of the following: purchase state, exchange state, and return state. The second type of product represented by the second historical dialogue has not been in the transaction state. The first dialogue information includes first description information and second description information. The first description information is expressed in a vague way for the historical dialogue product. The historical dialogue product includes the first type of product and the second type of product associated with the multiple historical dialogues. The second description information is the description information of the consultation matter. The consultation matter refers to the business matter associated with the sale of the product.

[0007] Send the first dialogue information to the server. The first dialogue information is associated with the second dialogue information. The second dialogue information contains the response information for the inquiry regarding the target product. The target product is either the first product in the first category of products or the second product in the second category of products.

[0008] Receive the second dialogue information from the server;

[0009] Display the second dialogue information.

[0010] Secondly, embodiments of this application provide an intelligent customer service system, including a first terminal device and a server, wherein,

[0011] The first terminal device is configured to perform the steps performed by the first terminal device as described in the first aspect above;

[0012] The server is configured to perform the steps performed by the server as described in the first aspect above.

[0013] As can be seen, in this embodiment, the first terminal device first displays the first dialogue information entered by the first consumer on the first customer service interface for the merchant user, then sends the first dialogue information to the server. The first dialogue information is associated with second dialogue information, which contains a response to the inquiry regarding the target product. Then, it receives the second dialogue information from the server and finally displays the second dialogue information. Compared to existing store customer service systems that can only predict the user's intent based on previous historical dialogue information and output a question rather than an answer, and compared to existing store customer service systems that can only provide vague answers or redirect to human customer service with long wait times when detecting vague references and inquiries regarding products associated with the dialogue, the intelligent customer service system of this application provides a function that can accurately handle dialogue questions and answers involving vague references and inquiries regarding products associated with historical dialogues without requiring a human intervention, thus improving the intelligence and applicability of the dialogue function of e-commerce machine customer service.

[0014] Furthermore, the intelligent customer service system of this application can specifically perform statistical analysis on the first and second categories of goods based on multiple historical dialogue information from the customer service interface, thereby supporting the accurate identification of the fuzzy references to goods related to historical dialogues and the intent of the dialogue questions regarding the inquiries, and then outputting response information for the inquiries regarding the target goods, thereby improving the accuracy, comprehensiveness and fluency of the customer service system in processing consumer dialogue information. Attached Figure Description

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

[0016] Figure 1 This is a schematic diagram illustrating the composition of an intelligent customer service system provided in an embodiment of this application;

[0017] Figure 2 This is a schematic diagram of the structure of a first terminal device provided in an embodiment of this application;

[0018] Figure 3 This is a flowchart illustrating a method for globally associating product information during a multi-turn dialogue, as provided in an embodiment of this application.

[0019] Figure 4 This is a schematic diagram of the first customer service interface provided in the embodiments of this application;

[0020] Figure 5This is a schematic diagram of the second customer service interface provided in the embodiments of this application. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be accurately and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0022] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or modules is not limited to the listed steps or modules, but may optionally include steps or modules not listed, or may optionally include other steps or modules inherent to such processes, methods, products, or apparatus.

[0023] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0024] Currently, for questions with complex historical chat logs and those where the user's question description includes references to related products from past conversations, automated customer service cannot provide accurate answers. Even after transferring to a human agent, it takes a considerable amount of time to review previous chat logs and perform background checks to provide a targeted answer to the user's needs, making it difficult to meet user requirements in terms of accuracy and timeliness.

[0025] To address the aforementioned issues, this application provides a method and system for globally associating product information during multi-turn dialogues. The embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0026] Please see Figure 1 , Figure 1This is a schematic diagram illustrating the composition of an intelligent customer service system provided in this application embodiment. The intelligent customer service system 10 includes a first terminal device 101 for a first consumer user, a server 103, and a second terminal device 102 for a merchant user. The server 103 is communicatively connected to both the first terminal device 101 and the second terminal device 102. The first consumer user can input dialogue information through the first terminal device 101, which may include inquiries about target products or related information about the merchant user, and this information is displayed on the first customer service interface. The merchant user can input reply information and product recommendations through the second terminal device 102. The first terminal device 101 and the second terminal device 102 can simultaneously display the first customer service interface, enabling communication between the consumer user and the merchant user based on this interface. The server 103 can determine associated second dialogue information based on the first dialogue information from the first terminal device 101 and send it to the second terminal device 102 for display.

[0027] Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of a first terminal device provided in an embodiment of this application. The first terminal device 101 includes a processor 120, a memory 130, a communication interface 140, and one or more programs 131. The one or more programs 131 are stored in the memory 130 and configured to be executed by the processor 120. The one or more programs 131 include instructions for performing any step in any of the methods described in the following embodiments. Specifically, the processor 120 is used to perform any step executed by the first terminal device in the following method embodiments, and when performing data transmission such as sending, it may selectively call the communication interface 140 to complete the corresponding operation.

[0028] It should be noted that the second terminal device 102 and the server 103 may have the same structure as the first terminal device 101, and will not be described again here.

[0029] Please see Figure 3 , Figure 3 This is a flowchart illustrating a method for globally associating product information during a multi-turn dialogue, as provided in an embodiment of this application. This method is applied to the first terminal device 101 of an intelligent customer service system 10 for merchant users on an e-commerce platform. The method includes the following steps.

[0030] S201, the first dialogue information entered by the first consumer is displayed on the first customer service interface of the merchant user for the first consumer.

[0031] The first customer service interface is associated with multiple historical dialogue records of the first consumer user. These historical dialogue records include first historical dialogue records for a first type of product and second historical dialogue records for a second type of product. Specifically, the dialogue records of the first consumer user in the second historical dialogue records can be empty. Alternatively, the dialogue records of the first consumer user in the second historical dialogue records can be non-empty. Specifically, the first type of product can be a product for which the consumer user has actively inquired and a transaction has occurred. For example, after a user directly jumps to the customer service interface through the product details page of product A, a transaction occurs after the first historical dialogue. The first type of product can also be a product for which a merchant user actively recommends a message, and the user clicks on the recommendation message to view the product, resulting in a transaction. Specifically, before the first dialogue records entered by the first consumer user are displayed on the merchant user's first customer service interface for the first consumer user, the method further includes: detecting that the product recommendation function of the first customer service interface is not turned off.

[0032] The first category of goods represented by the first historical dialogue information has been subject to a transaction, and the transaction status includes at least one of the following: purchase status, exchange status, and return status. The second category of goods represented by the second historical dialogue information has not been subject to any of these transaction statuses. To determine whether a product has been subject to a transaction, it can be first determined whether the historical dialogue for that product includes a corresponding function card, such as a "return / exchange application" function card. Consumers can click on this function card to jump to the corresponding function interface to return or exchange goods. If the corresponding function card is included, the product is determined to be a first category of goods. Alternatively, keyword matching can be performed on the historical dialogue for that product. For example, if the keyword "return" appears in the historical dialogue, the order status of the product can be queried to determine whether the product is a first category of goods. The second category of goods can refer to goods for which consumers have inquired but no transaction has occurred, or goods for which merchants have pushed information but users have not engaged in a dialogue regarding the product, or goods for which no transaction has occurred after the dialogue. For example... Figure 4 As shown, Figure 4 This is a schematic diagram of the first customer service interface provided in an embodiment of this application. The interface includes multiple historical dialogues, where the goods actively inquired about and purchased by the consumer user are classified as first-category goods, and the related historical dialogues constitute first-category historical dialogue information. The goods actively recommended by the merchant user are classified as second-category goods, and the merchant user's recommendation messages constitute second-category historical dialogue information.

[0033] The first dialogue information includes first descriptive information and second descriptive information. The first descriptive information is expressed as a vague reference to the products in the historical dialogue, and the historical dialogue products include the first type of products and the second type of products associated with the multiple historical dialogue information. The second descriptive information is a description of the inquiry matter, which refers to business matters related to the sale of the product. This vague reference means that the first dialogue information does not include explicit product names or identity information. For example, the first descriptive information could be "this mobile phone" or "the shoes I bought before." The second descriptive information can be used to inquire about product specifications, inventory, discounts, logistics, and returns / exchanges.

[0034] S202, send the first dialogue information to the server.

[0035] The first dialogue information is associated with the second dialogue information, which includes the response information to the inquiry regarding the target product. The target product is either the first product in the first category of products or the second product in the second category of products.

[0036] In a specific implementation, the method further includes: detecting that the product recommendation function of the first customer service interface is turned off; then the second dialogue information is created by the server through the following steps:

[0037] Step a1. Historical record storage.

[0038] After each round of dialogue, the system stores the user's question, the system's answer, and relevant product information (such as product ID, product name, attributes, etc.) in the history. For example, if a user asks "How much does this phone cost?" in the first round of dialogue, the system answers "This phone costs 3999 yuan" and stores the product ID (such as "12345") and product name (such as "X phone") in the history.

[0039] Step a2. Rewrite the multi-turn dialogue.

[0040] In the latest round of conversation, the system first analyzes the user input to determine if it involves products from previous conversations. If the user input includes a reference to a historical product (such as "this phone"), the system rewrites the input into a complete statement based on historical records. For example, if the user inputs "What is the battery capacity of this phone?", the system rewrites it as "What is the battery capacity of phone X?".

[0041] Step a3. Intent recognition to extract product ID.

[0042] The system uses intent recognition models (such as BERT, GPT, etc.) to analyze the rewritten user questions and extract the relevant product IDs. For example, in the user question "What is the battery capacity of mobile phone X?", the system extracts the product ID "12345".

[0043] Step a4. Obtain product information based on product ID.

[0044] The system retrieves detailed product information (such as battery capacity, price, color, etc.) from the product database based on the extracted product ID. For example, the system finds the product information corresponding to product ID "12345" as: "X mobile phone, battery capacity 3200mAh".

[0045] Step a5. Generate a system response.

[0046] The system generates an answer based on the user's question and the extracted product information. For example: User question: "What is the battery capacity of this phone?" System answer: "The battery capacity of phone X is 3200mAh."

[0047] In the case where the product recommendation function on the customer interface is turned off, the multiple historical dialogue information are all dialogue information between the first consumer user and the current store user's machine customer service and / or human customer service. This can be understood as dialogue information that is strongly related to the first consumer user. Therefore, compared with the aforementioned server-side technical solution, the server can also use the above-mentioned more efficient algorithm steps to create the first dialogue information.

[0048] S203, Receive the second dialogue information from the server.

[0049] In one possible embodiment, the target product is determined by the server by performing the following steps: creating a correspondence set based on the plurality of historical dialogue information, the correspondence set including a first correspondence between a first identity information sequence and a transaction-related tag, and a second correspondence between a second identity information sequence and a non-transaction-related tag; performing transaction-related analysis on the first description information and / or the second description information of the first dialogue information to obtain a transaction-related analysis result; performing product category analysis on the first description information of the first dialogue information to obtain a product category analysis result; and determining the target product based on the correspondence set, the transaction-related analysis result, the product category analysis result, and the inquiry matter.

[0050] In this system, the identity information sequence indicates a specific product, and the transaction association tag indicates the transaction status of that specific product. The identity information of the first type of product belongs to the first identity information sequence, and the identity information of the second type of product belongs to the second identity information sequence. Since the first type of product has not yet been traded, the first identity sequence corresponds to the transaction association tag; similarly, the second type of product has not yet been traded, so the second identity sequence corresponds to the non-transaction association tag. Specifically, transaction association analysis is performed on the first description information and / or the second description information of the first dialogue information to obtain the transaction association analysis results. This includes: performing semantic analysis on the first description information and / or the second description information to determine whether there are transaction-related semantics in the first description information and / or the second description information. If so, it is determined that the product corresponding to the first dialogue information has a transaction association. For example, if the semantic analysis results include content such as "purchased" or "returned," then the product referred to by the first dialogue information is the first type of product, and a transaction association exists.

[0051] In practice, when performing product category analysis, the first description information can be syntactically analyzed to determine the corresponding noun phrase. Then, the historical product categories included in the historical dialogue information are obtained. The noun phrase is compared with these historical product categories to determine the most relevant historical product category as the product category corresponding to the first dialogue information. For example, if the first description information is "rock climbing shoes I bought before," the noun phrase in this sentence is "rock climbing shoes." Since "climbing shoes" has appeared in historical dialogues and is the most relevant, the product category corresponding to the first dialogue information is "climbing shoes." Specifically, if the first description information does not include a noun phrase and only uses words like "this" to refer to it, then the product in the historical dialogue information that is closest to the first dialogue information is determined as the target product.

[0052] As can be seen, in this embodiment, by performing transaction correlation analysis and product type analysis on the first dialogue information and then comprehensively determining the target product, the accuracy of the target product confirmation can be improved, and the intelligence of the dialogue function of the e-commerce machine customer service can be enhanced.

[0053] In one possible embodiment, creating a set of correspondences based on the plurality of historical dialogue information includes: analyzing, encoding, and sorting the first historical dialogue information according to the dialogue time order to obtain a first identity information sequence; analyzing, encoding, and sorting the second historical dialogue information according to the dialogue time order to obtain a second identity information sequence; creating a first correspondence between the first identity information sequence and the transaction-related sub-label, and a second correspondence between the second identity information sequence and the non-transaction-related label; and merging the first correspondence and the second correspondence to obtain the set of correspondences.

[0054] In obtaining the identity information sequence based on identity information, the product to be encoded can first be determined based on historical dialogue information. Then, the ID of the product to be encoded can be obtained. Based on the transaction correlation analysis results, it can be determined whether the product to be encoded belongs to the first or second category of products. If it belongs to the first category, the corresponding product identification code is obtained. Furthermore, the sequential number of the product to be encoded within its category can be determined. This sequential number can be determined by the dialogue time in the historical dialogue information; the earlier the dialogue time, the smaller the sequential number. Then, based on the ID, product identification code, and sequential number of the product to be encoded, the identity information of the product to be encoded is generated, thus obtaining the first identity information sequence corresponding to the first category of products and the second identity information sequence corresponding to the second category of products.

[0055] As can be seen, in this embodiment of the application, determining the identity information sequence based on the first historical dialogue information and the second historical dialogue information respectively can improve the efficiency of the identity information sequence and ensure the accuracy of the determined identity information sequence.

[0056] In one possible embodiment, determining the target product based on the correspondence set, the transaction correlation analysis result, the product category analysis result, and the consultation matter includes: using the transaction correlation analysis result as a query identifier, querying the correspondence set to obtain an identity information sequence that matches the transaction correlation analysis result; removing irrelevant identity information from the matched identity information sequence based on the product category analysis result and the consultation matter to obtain a third identity information sequence after removing irrelevant identity information; and marking the product identified by the identity information in the third identity information sequence that is closest to the current system time as the target product.

[0057] The process of determining the response information for the inquiry related to the target product after identifying it specifically includes the following steps: rewriting the first dialogue information based on the target product; determining whether the logic of the first dialogue information is correct; if correct, obtaining the answer information based on the rewritten first dialogue information and the question-and-answer model; and then generating the second dialogue information based on the answer information and the target product. If the logic of the rewritten first dialogue information is incorrect, for example, if the first dialogue information is "What is the memory of climbing shoes?", then the logic of the rewritten first dialogue information can be corrected. For example, the logically incorrect first dialogue information can be corrected to "What is the lining of climbing shoes?". Then, the second dialogue information is generated based on the corrected first dialogue information and the target product.

[0058] As can be seen, in this embodiment, determining the target product by integrating information from multiple aspects can improve the accuracy of the target product determination and enhance the intelligence of the dialogue function of e-commerce machine customer service.

[0059] S204, Display the second dialogue information.

[0060] In one possible embodiment, displaying the second dialogue information includes: displaying the second dialogue information in a first adjacent dialogue area of ​​the first dialogue information displayed on the first customer service interface, and displaying third historical dialogue information or a dialogue information folding function control in a second adjacent dialogue area of ​​the first dialogue information displayed on the first customer service interface, wherein the third historical dialogue information is the historical dialogue information of the target product, and the dialogue information hidden by the dialogue information folding function control is historical dialogue information different from the historical dialogue information of the target product.

[0061] The first adjacent dialogue area and the second adjacent dialogue area are located on either side of the first dialogue information. This means that the second dialogue information can be displayed after the first dialogue information, while the historical dialogue information of the target product preceding the first dialogue information is displayed before the first dialogue information, and the historical dialogue information of non-target products preceding the first dialogue information is collapsed. For example... Figure 5 As shown, Figure 5 This is a schematic diagram of the second customer service interface provided in an embodiment of this application. In this customer service interface, "How do I return this table?" is the first dialogue information. Previous historical dialogue information related to this table is displayed on the interface, while information unrelated to the table, such as… Figure 4 The merchant user's proactive recommendation information is collapsed and not displayed. Then, after determining the second dialogue information, it appears on the other side of the first dialogue information.

[0062] As can be seen, in this embodiment, when displaying the second dialogue information, appropriately displaying the historical dialogue information of the target product can make the information displayed on the user customer service interface more intuitive and accurate, making it easier for consumers to trace the target product and improve communication efficiency.

[0063] In this embodiment, the first terminal device first displays the first dialogue information entered by the first consumer on the first customer service interface for the merchant user. Then, it sends the first dialogue information to the server. The first dialogue information is associated with second dialogue information, which includes a response to the inquiry regarding the target product. Next, it receives the second dialogue information from the server and finally displays it. Compared to existing store customer service systems that can only predict the user's intent based on previous historical dialogue information and output a question rather than an answer, and compared to existing store customer service systems that can only provide vague responses or redirect to human customer service with long wait times when detecting vague references and inquiries about products associated with the dialogue, this intelligent customer service system provides a function that can accurately handle consumers' dialogue questions involving vague references and inquiries about products associated with historical dialogues without requiring a human intervention, thus improving the intelligence and applicability of e-commerce machine customer service dialogue functions.

[0064] Furthermore, the intelligent customer service system of this application can specifically perform statistical analysis on the first and second categories of goods based on multiple historical dialogue information from the customer service interface, thereby supporting the accurate identification of the fuzzy references to goods related to historical dialogues and the intent of the dialogue questions regarding the inquiries, and then outputting response information for the inquiries regarding the target goods, thereby improving the accuracy, comprehensiveness and fluency of the customer service system in processing consumer dialogue information.

[0065] In one possible embodiment, the second dialogue information may be presented in at least one of the following forms: text information; image information; and preset function cards configured with hyperlinks.

[0066] Hyperlinks can be specifically designed for inquiries that machine-generated customer service cannot accurately retrieve, but for which users can quickly access product pages for further information. For example, on a shoe size page, inventory information dynamically displays the stock quantity for each shoe size, allowing users to check inventory information by clicking on preset function cards.

[0067] As can be seen, in this embodiment, the second dialogue information can be presented in a variety of ways, which can improve the intelligence of the response, enhance the intuitiveness of the response information, and improve the user experience.

[0068] In one possible embodiment, the text information includes hyperlink information and auxiliary descriptions, wherein the hyperlink information is indirect answer information for the inquiry of the target product; wherein the indirect answer information is data information that can be queried from the identity information of the target product in the first dataset; wherein the update cycle of the inquiry of the target product contained in the first dataset is less than a preset time.

[0069] To avoid the latency issue of real-time redirection to other system servers for querying results (users have very limited patience in conversational scenarios), the customer service system server generally needs to maintain a dedicated query dataset. This means the data in this query dataset is updated synchronously over a set period, such as a preset interval of one hour or real-time updates. For inquiries with short update cycles, a second answer can be sent to the user via hyperlinks, allowing them to directly access real-time information. For example, if the inquiry is about the inventory of a target product, since inventory changes rapidly, a hyperlink can guide the user to a more accurate real-time query page. Or, if the inquiry is about logistics, due to the complexity of logistics, a hyperlink can guide the user to a logistics page to obtain more information.

[0070] As can be seen, in this embodiment, for inquiries with a short update cycle, guiding users to view them via hyperlinks can improve the accuracy of the response information and enhance the user experience.

[0071] In one possible embodiment, the text information includes static text information, which contains a first direct answer to the inquiry regarding the target product. The first direct answer is data information retrieved in the latest data update cycle. The first direct answer is data information retrieved from the second dataset using the target product's identity information. The update cycle of the data information regarding the inquiry regarding the target product in the second dataset is greater than or equal to the preset duration.

[0072] If the inquiry concerns the product's specifications, such as color or battery capacity, the default update cycle for this type of inquiry can obviously be set to, for example, one month. For the server, it only needs to use the target product's identity information as the query identifier to perform the query operation.

[0073] As can be seen, in this embodiment, for inquiries with long update cycles, static text information can be used to directly answer users' inquiries, improving the intuitiveness of the response information.

[0074] In one possible embodiment, the text information includes static text information, which contains second direct answer information for the inquiry of the target product. The second direct answer information is data obtained by automatically performing a query operation based on the predicted indirect answer information of the inquiry at a first time point. The first time point is a time point earlier than the second time point, and the second time point is the time point at which the first dialogue information is displayed.

[0075] Once the dialogue target is confirmed to be the product in question, the system can predict potential indirect inquiries about that product and retrieve the corresponding direct answers in advance. This resolves the latency issue associated with real-time dialogue. Consequently, the customer service interface no longer displays indirect answers but instead shows the direct answers.

[0076] As can be seen, in this embodiment, the predicted user inquiry items can also be extracted and the answer information of the predicted inquiry items can be directly output, so that users can obtain more relevant information without asking again, thereby improving the intelligence of the dialogue function of e-commerce machine customer service.

[0077] This application provides an intelligent customer service system, including a first terminal device and a server, wherein the first terminal device is used to perform the steps performed by the first terminal device as described in any of the above embodiments; and the server is used to perform the steps performed by the server as described in any of the above embodiments.

[0078] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0079] This application also provides a computer storage medium storing a computer program for electronic data interchange, which causes a computer to perform some or all of the steps of any of the methods described in the above method embodiments, wherein the computer includes an electronic device.

[0080] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of any of the methods described in the above method embodiments. The computer program product may be a software installation package, and the computer may include an electronic device.

[0081] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0082] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0083] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical or other forms.

[0084] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0085] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0086] If the aforementioned integrated units are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0087] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage device, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0088] The embodiments of this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for globally associating product information during a multi-turn dialogue, characterized in that, The first terminal device of an intelligent customer service system for merchants on e-commerce platforms, the method comprising: The merchant user's first customer service interface for the first consumer displays the first dialogue information entered by the first consumer. The first customer service interface is associated with multiple historical dialogues of the first consumer. These multiple historical dialogues include first historical dialogues for a first type of product and second historical dialogues for a second type of product. The first type of product represented by the first historical dialogues includes products that have been recommended by the merchant user and whose transaction status has been generated after the first consumer clicked on the recommended message. The transaction status includes at least one of the following: purchase status, exchange status, and return status. The second type of product represented by the second historical dialogues has not generated the transaction status. The first dialogue information includes first description information and second description information. The first description information is expressed in a vague way for the historical dialogue products. The historical dialogue products include the first type of product and the second type of product associated with the multiple historical dialogues. The second description information is the description information of the consultation matter, which refers to the business matter related to the sale of the product. Send the first dialogue information to the server. The first dialogue information is associated with the second dialogue information. The second dialogue information contains the response information for the inquiry regarding the target product. The target product is either the first product in the first category of products or the second product in the second category of products. Receive the second dialogue information from the server; Display the second dialogue information; The target product is determined by the server through the following steps: A set of corresponding relationships is created based on the multiple historical dialogue information, the set of corresponding relationships including a first correspondence between a first identity information sequence and a transaction-related tag, and a second correspondence between a second identity information sequence and a non-transaction-related tag; Transaction correlation analysis is performed on the first description information and / or the second description information of the first dialogue information to obtain the transaction correlation analysis results; Perform product category analysis on the first description information of the first dialogue information to obtain product category analysis results; The target product is determined based on the set of corresponding relationships, the results of the transaction correlation analysis, the results of the product category analysis, and the consultation matters.

2. The method according to claim 1, characterized in that, The second dialogue information can be presented in at least one of the following forms: Text information; and, Image information; and, Preset function cards with hyperlink functionality.

3. The method according to claim 2, characterized in that, The text information includes hyperlinks and supplementary explanations, wherein the hyperlinks are indirect answers to the questions regarding the target product. The indirect answer information refers to the data information in the first dataset that can be retrieved through the identity information of the target product; Wherein, the update cycle of the consultation matters of the target product contained in the first dataset is less than a preset time.

4. The method according to claim 3, characterized in that, The text information includes static text information, which contains the first direct answer information for the inquiry regarding the target product. The first direct answer information is the data information retrieved in the latest data update cycle. Wherein, the first direct answer information is the data information that can be queried from the second dataset through the identity information of the target product; Wherein, the update cycle of the data information of the consultation matters of the target product contained in the second dataset is greater than or equal to the preset duration.

5. The method according to claim 3, characterized in that, The text information includes static text information, which contains the second direct answer information for the inquiry of the target product. The second direct answer information is data obtained by automatically performing a query operation based on the predicted indirect answer information of the inquiry at a first time node. The first time node is a time node earlier than the second time node, and the second time node is the time node at which the first dialogue information is displayed.

6. The method according to any one of claims 1-5, characterized in that, The display of the second dialogue information includes: The second dialogue information is displayed in the first adjacent dialogue area of ​​the first dialogue information displayed on the first customer service interface, and... The third historical dialogue information or a dialogue information collapse function control is displayed in the second adjacent dialogue area of ​​the first dialogue information displayed on the first customer service interface. The third historical dialogue information is the historical dialogue information of the target product, and the dialogue information hidden by the dialogue information collapse function control is historical dialogue information that is different from the historical dialogue information of the target product.

7. The method according to claim 1, characterized in that, The step of creating a corresponding relationship set based on the multiple historical dialogue information includes: The first historical dialogue information is analyzed, encoded, and sorted according to the dialogue time sequence to obtain the first identity information sequence. The second historical dialogue information is analyzed, encoded, and sorted according to the dialogue time sequence to obtain the second identity information sequence. Create a first correspondence between the first identity information sequence and the transaction association tag, and a second correspondence between the second identity information sequence and the non-transaction association tag; The first correspondence and the second correspondence are merged to obtain the correspondence set.

8. The method according to claim 7, characterized in that, The step of determining the target product based on the corresponding relationship set, the transaction correlation analysis results, the product category analysis results, and the consultation matter includes: Using the transaction correlation analysis result as the query identifier, query the corresponding relationship set to obtain the identity information sequence that matches the transaction correlation analysis result; Based on the product category analysis results and the consultation matter, irrelevant identity information is removed from the adapted identity information sequence to obtain a third identity information sequence after removing irrelevant identity information; and the product identified by the identity information closest to the current system time in the third identity information sequence is marked as the target product.

9. An intelligent customer service system, characterized in that, Including the first terminal device and the server, among which, The first terminal device is configured to perform the steps executed by the first terminal device in any of the methods described in claims 1-8; The server is configured to perform the steps performed by the server in any of the methods described in claims 1-8.

Images