CN115208846B - Session interaction method and system - Google Patents

Abstract

The embodiment of the specification provides a session interaction method and a session interaction system, wherein the method comprises the steps of responding to a session request input by a user through an interaction terminal, triggering the interaction terminal to present a session interface, and determining the display content of the session interface based on a message body, wherein the message body comprises a history message part, a system message part and a current session message part; acquiring an interaction instruction input by a user through an interaction terminal; updating a message body in response to the interaction instruction; and triggering the interactive terminal to update the session interface based on the updated message body.

Description

A conversational interaction method and system

technical field

This specification relates to the technical field of human-computer interaction, and in particular to a conversational interaction method and system.

Background technique

In order to meet user needs, more and more platforms will provide users with intelligent question-and-answer services. Through the intelligent question-and-answer service, users can have conversations with the question-and-answer robot on terminals such as mobile phones and computers, so as to raise service requests and questions. Traditional question-and-answer platforms usually store different information, such as user information, system information, and robot reply information, separately, resulting in complex data storage structures and low data management efficiency. Since the conversational interaction between the user and the Q&A robot determines the structure of the data stored in the Q&A platform, when there is a large change in the conversational interaction, the data structure will need to be adjusted. When the data structure is complex, the process of data structure adjustment is also extremely difficult, and there are some repetitive tasks.

Therefore, it is desired to provide a session interaction method and system, which can uniformly process and manage session interaction related data, and improve data storage and management efficiency.

Contents of the invention

One of the embodiments of this specification provides a session interaction method. The conversational interaction method includes: in response to a conversational request input by a user via an interactive terminal, triggering the interactive terminal to present a conversational interface, the display content of the conversational interface is determined based on a message body, wherein the message body includes a historical message part, A system message part and a current session message part; acquiring an interaction instruction input by the user via the interaction terminal; updating the message body in response to the interaction instruction; triggering the interaction terminal to update the message body based on the updated message body Describe the session interface.

One of the embodiments of this specification provides a conversational interaction system, the conversational interaction system includes: a triggering module, configured to trigger the interactive terminal to present a conversational interface in response to a conversational request input by a user via an interactive terminal, and the conversational interface The display content is determined based on the message body, wherein the message body includes a historical message part and/or a system message part and/or a current session message part; an acquisition module, configured to acquire the interaction instruction input by the user via the interaction terminal; An update module, configured to update the message body in response to the interaction instruction; a trigger module, configured to trigger the interaction terminal to update the session interface based on the updated message body.

One of the embodiments of the present specification provides a computer-readable storage medium, the storage medium stores computer instructions, and after the computer reads the computer instructions in the storage medium, the computer executes the session interaction method.

In some embodiments of the present specification, the message body may be used to store historical messages, system messages, current session messages, etc. involved in the conversation process between the user and the question-answering robot. Since various messages in the message body are stored based on a preset arrangement, various messages can be quickly located, and the display content of the conversation interface can be determined based on the message body, without obtaining different types of messages from different sources. Moreover, based on the message body, relevant data related to session interaction can be processed and managed in a unified manner, and then the data structure can be flexibly adjusted to improve the efficiency of data storage and management. In the traditional conversation interaction method, since different messages correspond to different message sources or categories, and user questions and system replies are stored separately, it is necessary to link related messages through logic. During the conversation, different information needs to be retrieved from different sources, and logical judgments are required to find related data. Compared with the traditional conversational interaction method, the message body-based conversational interaction method has higher data management and storage efficiency, and can provide better conversational services.

Description of drawings

This specification will be further illustrated by way of exemplary embodiments, which will be described in detail with the accompanying drawings. These examples are non-limiting, and in these examples, the same number indicates the same structure, wherein:

FIG. 1 is a schematic diagram of an application scenario of a conversational interaction system according to some embodiments of this specification;

Fig. 2 is an exemplary module diagram of a conversational interaction system according to some embodiments of this specification;

Fig. 3 is an exemplary flow chart of a session interaction method according to some embodiments of this specification;

Fig. 4 is an exemplary flow chart of a session interaction method according to some embodiments of this specification;

Fig. 5 is an exemplary schematic diagram of a message body according to some embodiments of this specification;

Fig. 6 is an exemplary schematic diagram of a data body of a single message according to some embodiments of the present specification;

Fig. 7A is an exemplary schematic diagram of a first conversation interface according to some embodiments of this specification;

Fig. 7B is an exemplary schematic diagram of a second conversation interface according to some embodiments of this specification;

Fig. 7C is an exemplary schematic diagram of a third conversation interface shown according to some embodiments of this specification;

Fig. 7D is an exemplary schematic diagram of a fourth conversation interface according to some embodiments of this specification.

Detailed ways

In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the following briefly introduces the drawings that need to be used in the description of the embodiments. Apparently, the accompanying drawings in the following description are only some examples or embodiments of this specification, and those skilled in the art can also apply this specification to other similar scenarios. Unless otherwise apparent from context or otherwise indicated, like reference numerals in the figures represent like structures or operations.

It should be understood that "system", "device", "unit" and/or "module" as used herein is a method for distinguishing different components, elements, parts, parts or assemblies of different levels. However, the words may be replaced by other expressions if other words can achieve the same purpose.

As indicated in the specification and claims, the terms "a", "an", "an" and/or "the" are not specific to the singular and may include the plural unless the context clearly indicates an exception. Generally speaking, the terms "comprising" and "comprising" only suggest the inclusion of clearly identified steps and elements, and these steps and elements do not constitute an exclusive list, and the method or device may also contain other steps or elements.

The flowchart is used in this specification to illustrate the operations performed by the system according to the embodiment of this specification. It should be understood that the preceding or following operations are not necessarily performed in the exact order. Instead, various steps may be processed in reverse order or simultaneously. At the same time, other operations can be added to these procedures, or a certain step or steps can be removed from these procedures.

Fig. 1 is a schematic diagram of an application scenario of a conversation interaction system according to some embodiments of this specification. As shown in FIG. 1 , an application scenario 100 of a conversation interaction system may include one or more of a processing device 110 , an interaction terminal 120 , a storage device 130 and the like.

In some embodiments, the user can interact with the question-answering robot via the interactive terminal 120, for example, submitting a session request, submitting an interaction instruction, and the like. The interactive instruction can be used to request a reply to a question, request to load historical messages, request to provide a certain service (such as online consultation), etc. A question-answering robot refers to a computer program that uses artificial intelligence, natural language processing and other technologies to conduct conversations with users. In some embodiments, the question answering bot may be executed or realized by the processing device 110 .

The interactive terminal 120 may refer to one or more terminals used by the user, including but not limited to one or any of other devices with input and/or output functions such as mobile devices, tablet computers, laptop computers, and desktop computers. combination. In some embodiments, the interactive terminal 120 may include other smart terminals, such as wearable smart terminals.

The interactive terminal 120 may include a conversational interface 121 . The conversational interface 121 can display the interaction record between the user and the question-answering robot, and the like. In some embodiments, a user may input a session request and/or an interaction instruction via an interaction terminal. The interaction terminal 120 may send the input session request and/or interaction instruction to the processing device 110 . The processing device 110 may process data related to the session request and the interaction instruction, and trigger the interaction terminal 120 to update the session interface 121 based on the processing result. In some embodiments, the interactive terminal 120 may be installed with a small program, APP, etc. for conversation. When the user opens the applet, APP, etc., the interactive terminal 120 may present the conversation interface 121 . In some embodiments, the interactive terminal 120 may perform data transmission with the processing device 110 and the storage device 130 through the network 120 . In some embodiments, processing device 110 may be part of interactive terminal 120 .

In some embodiments, the user using the interactive terminal 120 may be one or more users performing operations such as consultation. Different users have their own unique identification numbers. Users may include various types of users, for example, registered users, tourists, and so on. The processing device 110 can confirm the type of the user according to the unique identification numbers of different users. In some embodiments, the user can also use the interactive terminal 120 to have a conversation with a human customer service. For example, when the Q&A robot cannot reply to the user, a human customer service can reply. For another example, when a user needs to consult a human customer service, the human customer service can reply. The above examples are only used to illustrate the interactive terminal 120 and not to limit it.

The storage device 130 may be used to store data and/or instructions related to the application scenario 100 of the conversational interaction system. In some embodiments, the storage device 130 may store data and/or information obtained from the processing device 110, the interactive terminal 120, and the like. For example, the storage device 130 may store the message body and the message body's historical message part, system message part, current session message part, and the like.

In some embodiments, the storage device 130 may include one or more storage components. For example, one or more storage components may respectively correspond to a database, for example, a historical message database 131, a question-and-answer database 132, and the like. Alternatively, the historical message database 131 and the problem database 132 may correspond to the same storage component, for example, be stored in different partitions of the same memory. In some embodiments, the storage device 130 may be disposed in the processing device 110 . In some embodiments, the storage device 130 may include random access memory (RAM), read only memory (ROM), mass storage, removable memory, volatile read-write memory, etc., or any combination thereof. Exemplarily, mass storage may include magnetic disks, optical disks, solid-state disks, and the like. In some embodiments, storage device 130 may be implemented on a cloud platform.

The historical message database 131 can be used to store historical messages in the session interaction system. For example, historical messages may include historical questions sent by different users and historical replies to these historical questions by the question answering robot. In some embodiments, the historical messages in the historical message database 131 may be stored in a question-and-answer format. For example, in the historical message database 131, a historical question of a user and a historical reply of its corresponding question answering robot will be associated as a message record and stored together. A history reply of the question answering robot may include one or more pieces of information regarding the history question. In some embodiments, only historical valid replies are stored in the historical message database 131 , while invalid historical replies are not stored in the historical message database 131 . For more information about valid replies and invalid replies, please refer to the related description in FIG. 4 . In some embodiments, the historical message database 131 may store each historical question and the historical reply to the historical question. Regardless of whether the historical reply is valid or not, the historical reply will be stored in the historical message database 131 . In some embodiments, the historical messages in the historical message database 131 may also be stored in other ways.

Question and answer database 132 may be used to store questions and corresponding answers. When the user asks a question, the processing device 110 may query whether there is a corresponding answer through the question-and-answer database 132, and then reply to the user's question. In some embodiments, the questions and corresponding answers in the question-and-answer database 132 may be stored in a question-and-answer format. In some embodiments, the processing device 110 may make updates to the answers database 132, or the like. For example, according to the instruction of the manager of the question answering platform, the processing device 110 may add the new question and its corresponding answer to the question answering database 132 after being paired. For another example, according to the instruction of the manager of the question-and-answer platform, the processing device 110 may update the questions and corresponding answers in the question-and-answer database (for example, delete or change some answers) according to actual needs.

In some embodiments, the processing device 110 may process information and/or data related to the application scenario 100 of the conversational interaction system to perform one or more functions described in this specification. For example, the processing device 110 may process a session request, an interaction instruction (such as asking a question, loading a history message), etc. input by a user via the interaction terminal 120, and trigger the interaction terminal 120 to present a corresponding session interface. For another example, the processing device 110 may update the historical message database 131 in the storage device 130 based on the user's session record.

In some embodiments, the processing device 110 may use the message body to drive the interaction between the user and the interactive terminal 120 (or the question-and-answer robot). A message body may refer to a collection of various messages, where different messages are stored in a specific form. The content in the message body can be used to determine the display content of the conversation interface 121 of the interactive terminal 120 . For example, at the beginning of a session, the content in the message body is empty. As the session progresses, one or more of historical messages, system messages, session messages, etc. will be updated in the message body. At the same time, the display content of the conversation interface 121 will be constantly updated based on the message body.

As an example only, as shown in FIG. 5 , the message body (msgBody) 500 includes a history message part (msgHistory) 510, a system message part (msgSystem) 520, a current session message part (msgCurrent) 530 and so on. Different message parts of the message body display different content.

The history message part 510 is mainly used to store the history messages between the user and the Q&A robot. In some embodiments, the historical message section 510 may store a certain number of historical messages, for example, 10, 20 recent historical messages. In some embodiments, as mentioned above, the historical messages of the user and the Q&A robot can be stored in the historical message database 131 in the form of message records, and each message record includes the user's historical questions and the historical replies (or valid messages) of the Q&A robot. historical responses). The historical message section 510 can be used to store a specific number of message records (ie, a specific number of historical question and historical reply pairs). For example, historical message section 510 may be used to store 10 pairs of historical questions and historical replies.

The system message part 520 can be used to store system messages sent by the question answering robot, for example, "Hi, I am Xiao U, an intelligent customer service. May I help you?", "Little U guesses what you want to ask: 1. Film printing 2. Film printing color difference..." etc. In some embodiments, the system message section 520 will only store system messages sent by the Q&A robot in the current session.

The current session message section 530 may be used to store messages of the current session between the user and the question answering robot. In some embodiments, according to the interaction between the user and the question-answering robot, the content of the message part of the current conversation can be updated in real time.

In some embodiments, the history message section 510, the system message section 520, and the current session message section 530 may be arranged based on a preset order. The preset sequence may refer to a preset sequence of different message parts in the message body 500 . For example, as shown in FIG. 5 , the history message part 510 , the system message part 520 and the current session message part 530 in the message body 500 are arranged in sequence. The system message part 520 is located in the middle of the message body 500 and can separate the history message part 510 from the current session message part 530 . By arranging the historical message part 510, the system message part 520 and the current session message part 530 in a preset order, it can help to distinguish and locate messages quickly, and improve data management and control efficiency. For example, historical messages stored above system messages and current session messages stored below system messages can be quickly located based on the system messages section 520 .

It should be noted that the preset order shown in FIG. 5 is provided for example purposes only, and the history message part 510 , the system message part 520 and the current session message part 530 may be arranged in other orders. For example, the preset order may be that the system message part 520, the history message part 510 and the current session message part 530 are arranged in sequence.

In some embodiments, data related to a single message in message body 500 may be stored on a data body basis.

As an example only, FIG. 6 shows an exemplary schematic diagram of a data body of a single message. As shown in FIG. 6 , the data body (msgSingle) 600 may include userType 610, uid 620, type 630, content 640, time 650, netError/loading 660, and so on.

In some embodiments, userType 610 may be used to determine whether the data body corresponds to a system message or a non-system message. For example, when userType 610 is "system", the data body is a system message. The uid 620 can determine the sender of the message. For example, when uid 620 is "0", the data body corresponds to a message sent by the user; when uid 620 is "1", the data body corresponds to a message sent by the system. type 630 can determine the message type, for example, at least one of text messages, pictures, ratings, shared content, and the like.

content 640 may encapsulate message content. content 640 may include id, text, type, evaluate, etc. The id is the identification number of the message corresponding to the data body, and the identification numbers corresponding to different data bodies are different. text is the message content corresponding to the data body. type is the content category of the data body. evaluate can be used to determine whether the message corresponding to the data body is an evaluation message. For example, when the data body is an evaluation type message, evaluate is "true". When the data body is a non-evaluation type message, evaluate is empty.

The time 650 can record the message sending time corresponding to the data body. netError/loading 660 can determine the status of the message. The message status may include message network status and message loading status. The message network status can indicate the network abnormality when the message corresponding to the data body is sent. The message loading status may indicate that the message corresponding to the data body is being loaded. In some embodiments, a single data body will only have one of the message network state and the message loading state.

In some embodiments, data stored in a single data volume may include external state data and internal state data. External state data can be used to determine associations of data bodies with other data bodies. The internal state data may represent the content of the message itself corresponding to the data body. For example, content 640 in data body 600 belongs to internal state data, and other data belongs to external state data.

In some embodiments, the data bodies of different types of messages (eg, historical messages, system messages, current session messages, etc.) may be the same or different. For example, the netError/loading660 part can be omitted from the message body corresponding to the system message. By setting the data body, the data related to the message can be stored in a more standardized and unified form, which is conducive to improving the efficiency of subsequent data management and processing.

In some embodiments, processing device 110 may include one or more processing engines (eg, single-chip processing engines or multi-chip processing engines). By way of example only, processing device 110 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), an application specific instruction processor (ASIP), a graphics processing unit (GPU), a physical processing unit (PPU), a digital signal processor ( DSP), field programmable gate array (FPGA), programmable logic circuit (PLD), controller, microcontroller unit, reduced instruction set computer (RISC), microprocessor, etc. or any combination of the above.

It should be noted that the application scenario 100 of conversational interaction is provided for the purpose of illustration only, and is not intended to limit the scope of this description. For those skilled in the art, various modifications or changes can be made according to the description in this specification. For example, the application scenario 100 of session interaction may implement similar or different functions on other devices. However, these changes and modifications do not depart from the scope of this specification.

For example, message body 500 shown in FIG. 5 and data body 600 shown in FIG. content, or omit parts of it. For example, the content contained in the message body 500 is not limited to the history message part 510, the system message part 520 and the current session message part 530, and the processing device 110 can be configured according to actual needs.

For another example, the application scenario 100 of the conversation interaction system may also include a network and the like. A network can connect various components of a system and/or connect parts of the system with external resources. A network enables communication between components and with other parts outside the system. For example, the processing device 110 acquires historical messages from the storage device 130 through a network.

Fig. 2 is an exemplary module diagram of a conversational interaction system according to some embodiments of this specification.

In some embodiments, the session interaction system 200 may include a trigger module 210 , an acquisition module 220 , an update module 230 and a determination module 240 .

In some embodiments, the triggering module 210 may be configured to trigger the interactive terminal to present the session interface in response to a session request input by the user via the interactive terminal, and the display content of the session interface is determined based on the message body, wherein the message body includes a history message part and/or or the system messages section and/or the current session messages section.

In some embodiments, the acquisition module 220 may be used to acquire an interaction instruction input by a user via an interaction terminal.

In some embodiments, the update module 230 may be configured to update the message body in response to the interaction instruction.

In some embodiments, the triggering module 210 may also be configured to trigger the interactive terminal to update the session interface based on the updated message body.

In some embodiments, the obtaining module 220 may also be used to obtain target system information in response to the session request. In some embodiments, the update module 230 may also be configured to update the system message part based on the target system message, so that the target system message is displayed on the conversation interface.

In some embodiments, the determining module 240 may be used to determine whether there are historical messages for the user.

In some embodiments, in response to the determination result that the user has historical messages, the triggering module 210 may also be configured to trigger the interactive terminal to display a historical message loading prompt.

In some embodiments, the interaction instruction includes a historical message loading instruction, and the acquiring module 220 may also be configured to acquire the target historical message based on the historical message database in response to the historical message loading instruction. In some embodiments, the update module 230 may also be configured to update the historical message part based on the target historical message, so that the target historical message is displayed on the conversation interface.

In some embodiments, the interaction instruction includes a user question, and the acquiring module 220 may also be configured to acquire a question reply in response to the user question. In some embodiments, the updating module 230 may also be configured to update the message part of the current conversation based on the reply to the question, so that the reply to the question is displayed on the conversation interface.

In some embodiments, the determination module 240 can also be used to determine whether the question reply includes a valid reply; in response to the judgment result that the question reply includes a valid reply, store the user question and the question reply pair in the historical message database.

In some embodiments, the obtaining module 220 can also be used to determine whether the network is abnormal. In some embodiments, the triggering module 210 may also be configured to trigger the interactive terminal to display a message abnormality prompt in response to the judgment result of the network abnormality. In some embodiments, the obtaining module 220 may also be configured to obtain a reply to the question in response to the judgment result that the network is normal.

It should be understood that the system and its modules shown in FIG. 2 can be implemented in various ways. For example, in some embodiments the system and its modules may be implemented by hardware, software or a combination of software and hardware.

It should be noted that the above description of the system and its modules is only for convenience of description, and does not limit this description to the scope of the illustrated embodiments. It can be understood that for those skilled in the art, after understanding the principle of the system, it is possible to combine various modules arbitrarily, or form a subsystem to connect with other modules without departing from this principle. In some embodiments, the trigger module 210, acquisition module 220, update module 230, and determination module 240 disclosed in FIG. 2 may be different modules in one system, or one module may implement the above-mentioned two or more modules. function. For example, each module may share one storage module, or each module may have its own storage module. Such deformations are within the protection scope of this specification.

Fig. 3 is an exemplary flow chart of a session interaction method according to some embodiments of this specification. As shown in FIG. 3 , the process 300 includes the following steps. In some embodiments, the process 300 may be executed by the processing device 110 .

Step 310, in response to a session request input by the user via the interactive terminal, the interactive terminal is triggered to present a session interface, and the display content of the session interface is determined based on the message body, wherein the message body includes a history message part, a system message part and a current session message part. In some embodiments, step 310 may be performed by the triggering module 210 .

The session request may refer to a request sent by a user that requires a session. In some embodiments, the triggering module 210 may receive a session request input by a user through an interactive terminal. For example, when a user opens a question-and-answer platform, a question-and-answer applet, a question-and-answer webpage, etc. through an interactive terminal, the trigger module 210 may confirm that the user inputs a session request. The interactive terminal will automatically send the session request to the processing device 110 . For another example, there are options or buttons for sending session requests in the question-and-answer platform, question-and-answer applet, question-and-answer web page, and the like. When the user clicks on the option or button, the trigger module 210 may confirm that the user enters a session request.

A conversational interface may refer to an interface for a conversation between a user and a question-answering robot.

In some embodiments, in response to a session request input by the user via the interactive terminal, the triggering module 210 may acquire the target system message; and based on the target system message, update the system message part of the message body, so that the target system message is displayed on the session interface. For example, a targeted system message could be used to greet the user. For another example, the target system message may provide the user with some commonly used service options.

In some embodiments, the content of the target system message may be the same or different for different users. For example, the target system message may be a unified greeting message for the user. For another example, the trigger module 210 may determine different target system messages according to the information registered by the user (such as gender, age, past medical history, allergy history, etc.). Exemplarily, if the user is male, age 50, and past medical history includes high blood pressure, the trigger module 210 may send a system message about consulting high blood pressure. Exemplarily, if the user is female, age 50, and past medical history includes stomach disease, the trigger module 210 may send a system message about consulting about stomach disease. For another example, if the user is an old user, the trigger module 210 may determine the content of the system message part according to the content of the user's historical message part. Obtain different target system messages for different users, and then make the conversation interface display the target system messages, so that users can obtain the required system messages, and improve the user's experience in using the Q&A applet.

In some embodiments, in response to a session request input by a user, the triggering module 210 may determine whether the user has a history message. When the user has historical messages, the triggering module 210 may trigger the interactive terminal to display a historical message loading prompt. In some embodiments, the trigger module 210 may determine whether the user has historical messages based on the message body or the historical message database. For more information about determining whether there is a history message for the user, please refer to the related description in FIG. 4 .

Step 320, acquiring an interaction instruction input by the user via the interaction terminal. In some embodiments, step 320 may be performed by the acquisition module 220 .

Interaction instructions can be used to ask the question-answering robot to perform specific actions, for example, answering questions, loading historical messages, providing certain services (online consultation, query report), etc. For example, as shown in FIG. 7D , the interactive instruction may require the question-and-answer robot to reply to the question "has the result of the CT examination come out?". For another example, the interactive instruction may require the question-answering robot to load historical messages. In some embodiments, the user may input the interaction instruction through the interactive terminal by means of typing, voice, gesture operation and the like.

Step 330, updating the message body in response to the interaction instruction. In some embodiments, step 330 may be performed by the update module 230 .

In some embodiments, the update module 230 may update the message body based on the interaction instruction. For example, the update module 230 may obtain a reply corresponding to the interaction instruction through a database in the storage device 130 (such as the historical message database 131 , the question-and-answer database 132 , etc.) based on the interaction instruction input by the user. The update module 230 may add the interaction instruction and the reply corresponding to the interaction instruction into the message body, and update the message body. In some embodiments, the update module 230 can update different parts of the message body based on different interaction instructions. For example, the interaction instruction is a historical message loading instruction. Based on the target historical message, the update module 230 may update the historical message part of the message body so that the target historical message is displayed on the conversation interface. For details, please refer to the related description in FIG. 4 . For another example, if the interaction instruction is a user question, the update module 230 may update the current conversation message part of the message body, so that the question reply is displayed on the conversation interface. For details, please refer to the related description in FIG. 4 .

Step 340, based on the updated message body, trigger the interactive terminal to update the session interface. In some embodiments, step 340 may be performed by the triggering module 210 .

For example, as described above, in response to a user question input by a user, a question reply may be obtained and the current conversation message part in the message body may be updated based on the question reply. Based on the updated message body, the triggering module 210 may trigger the interactive terminal to display the reply to the question. For another example, in response to a historical message loading instruction input by the user, the target historical message may be acquired and the historical message part in the message body may be updated based on the target historical message. Based on the updated message body, the triggering module 210 can trigger the interactive terminal to display the target history message.

In some embodiments of the present specification, the message body may be used to store historical messages, system messages, current session messages, etc. involved in the conversation process between the user and the question-answering robot. Since various messages in the message body are stored based on a preset arrangement, the processing device 110 can quickly locate various messages and determine the display content of the conversation interface based on the message body without obtaining different types of messages from different sources. Moreover, the processing device 110 can uniformly process and manage data related to session interaction, and then can flexibly adjust a data structure to improve data storage and management efficiency. In the traditional conversation interaction method, since different messages correspond to different message sources or categories, and user questions and system replies are stored separately, it is necessary to link related messages through logic. During the conversation, different information needs to be retrieved from different sources, and logical judgments are required to find related data. Compared with the traditional conversational interaction method, the message body-based conversational interaction method has higher data management and storage efficiency, and can provide better conversational services.

Fig. 4 is an exemplary flowchart of a session interaction method according to some embodiments of this specification. As shown in Fig. 4, the process 400 includes the following steps. In some embodiments, the process 400 may be performed by the processing device 110 . In some embodiments, process 400 may be an embodiment of process 300 .

Step 401, in response to a session request, acquire target system information. In some embodiments, step 401 may be performed by the acquiring module 220 .

The target system message may refer to a message sent automatically by the system for a session request. Target system messages can be used to greet users, provide users with some common service options, and so on. For more description about the target system message, refer to step 310 in FIG. 3 .

Step 402: Update the system message part based on the target system message, so that the target system message is displayed on the conversation interface. In some embodiments, step 402 may be performed by the update module 230 .

For example, as shown in FIG. 7A, after the user inputs a session request via the interactive terminal, the system message part of the message body may be updated, and the trigger module 210 may trigger the interactive terminal to present the session interface 121-1 based on the updated message body. The conversational interface 121-1 can display the target system message sent by the question-and-answer robot U, for example, "Hi, I am U, the smart customer service. May I help you?", "U guesses what you want to ask: 1. Film Printing 2. Color difference of film printing..." etc. Users can select the corresponding needs or send their own needs according to the part of the system message sent by the question-and-answer robot U.

Step 403, in response to the session request, determine whether the user has historical messages. In some embodiments, step 403 may be performed by the determining module 240 .

The historical messages may refer to previous interaction messages between the user and the question answering robot. In some embodiments, in response to the session request, the determination module 240 may determine whether the user has history messages by determining whether the user is a newly registered user. In some embodiments, in response to the session request, the determining module 240 may determine whether the user has historical messages through the historical message database. For example, the determining module 240 can determine whether there is a historical message of the user in the historical message database 131 based on the user's identity information (such as ID number, social security number, the unique identification number of the user in the conversational interaction system, etc.), and then determine the user's Whether there are historical messages.

In some embodiments, in response to the session request, the determination module 240 may determine whether the user has historical messages based on the data body. Exemplarily, after the user starts a conversation, the determining module 240 may read the historical messages between the user and the question-answering robot from the historical message database 131, and update the historical message part based on the read historical messages. Further, the determination module 240 may determine whether the user has historical messages based on the current data volume. For example, since the historical message part, the system message part and the current session message part in the data body are arranged sequentially, if the message body only includes the system message part or the first message is a system message, it means that the user has no historical message. The first message may refer to the first message in the data body, for example, the first message appearing in the data body.

Step 404 may be executed in response to the judging result that the user has historical messages. In response to the judging result that there is no historical message for the user, step 407 may be executed.

Step 404, triggering the interactive terminal to display a prompt for loading historical messages. In some embodiments, step 404 may be performed by the triggering module 210 .

The historical message loading prompt may be a prompt for reminding the user to choose whether to load the historical message. In some embodiments, the history message loading prompt can be expressed in various ways, for example, text, picture, voice and so on. As shown in FIG. 7B , the conversational interface 121 - 2 displays the text "Drop down to load historical messages" as a prompt for loading historical messages. A variety of different historical message loading prompts can be preset in advance and stored in the storage device 130 . In some embodiments, different users may have different forms of history message loading prompts. For example, when the user is over 60 years old, the triggering module 210 may trigger the interactive terminal to display a historical message loading prompt with a larger text or voice. When the judgment result of whether the user has historical messages is no, the message loading prompt will not be displayed on the session interface of the interactive terminal.

When the user inputs a historical message loading instruction, step 405 may be performed. The history message loading instruction can be input in various ways. For example, when the prompt for loading historical messages is "load historical messages by pulling down", the user can input historical message loading instructions by pulling down the screen. For another example, the user may issue a voice such as "load historical messages", "yes" and so on to input the historical message loading instruction. In some embodiments, the user may input the historical message loading instruction multiple times. For example, the user may pull down the screen twice to input two historical information loading instructions.

Step 405, in response to the historical message loading instruction, acquire the target historical message based on the historical message database. In some embodiments, step 405 may be performed by the obtaining module 220 .

In some embodiments, the target historical messages may include all or part of the user's historical messages. For example, the target historical messages may include N (eg, 10, etc.) historical messages closest to the current session. The acquisition module 220 can set the value of N according to actual needs, for example, N is 10, 15, and so on. In some embodiments, as described in part in FIG. 1 , the historical messages in the historical message database 131 are stored in the form of one question and one answer. The target historical message may include a specific number of user historical question and historical reply pairs. For example, the target historical message may include 10 pairs of historical questions and historical replies of the user. Every time the user inputs a historical message loading instruction, the obtaining module 220 will obtain 10 pairs of historical questions and historical replies. Loading only a specific number of target historical messages and/or loading them in batches can improve responsiveness while meeting loading requirements. In some embodiments, the target historical message may include at least one pair of historical question and valid historical reply. For example, the target historical message will only include historical questions and their corresponding valid historical replies. Invalid replies, likes, dislikes, comments, system messages, etc. can not be loaded, and invalid information can be filtered to improve the user's viewing efficiency and experience.

In some embodiments, the targeted historical messages do not include current session messages. For example, each session of a user has a corresponding digital identification framework (openID) as the identifier of the session. Assume that the openID of the current session is A. The acquisition module 220 can use the openID to distinguish whether different messages belong to the current session or the historical session, etc., and only select the historical messages contained in the historical sessions whose openID is not A as the target historical messages. The user is already familiar with the messages generated in the current session (for example, user messages sent by the user, system replies, etc.), if the current session message appears again in the target historical message, it will reduce the user's experience and query effect of using the Q&A applet Efficiency of target history messages.

In some embodiments, the acquiring module 220 can acquire the target historical message by querying the historical message database. In some embodiments, the acquiring module 220 can acquire the target history message based on the message body. For example, after the user starts a conversation, the obtaining module 220 may read the target historical message between the user and the chat robot from the historical message database, and update the historical message part based on the read target historical message. When the subsequent user chooses to load historical messages, the obtaining module 220 may directly obtain the target historical messages from the historical message part. By obtaining the historical message in advance to update the message body, and loading the target historical message based on the updated message body, the efficiency of historical message loading can be improved.

In some embodiments, when the acquiring module 220 acquires the target historical message, the conversation interface may be positioned to the last one of the target historical message (ie, the closest one among the target historical messages to the current time point). Users can view other historical messages in chronological order by scrolling up. It is convenient for the user to check from the 1 historical message closest to the current time point. And, with this design, there is no need to calculate the scroll height when positioning the screen.

In some embodiments, when the acquisition module 220 acquires the target history message, there may be a network abnormality. For example, the acquisition module 220 can determine whether the network is abnormal. In response to the judgment result of the network abnormality, the triggering module 210 may trigger the interactive terminal to display an abnormality prompt indicating failure to record the historical message. Optionally, the triggering module 210 may confirm with the user whether to switch accounts. When the user chooses to switch accounts, the system is initialized and the session is resumed. When the user refuses to switch accounts, the session interaction ends. In response to the judgment result that the network is normal, step 406 may be executed.

Step 406: Based on the target historical message, update the historical message part, so that the target historical message is displayed on the conversation interface. In some embodiments, step 406 may be performed by the update module 230 .

In some embodiments, the update module 230 may add the target historical message to the historical message part, so that the target historical message is displayed on the conversation interface. For example, as shown in FIG. 7C , the upper half of the conversation interface 121-3 may display history messages including the target. In some embodiments, when the user inputs multiple message loading instructions for multiple times, the acquisition module 220 can acquire multiple target historical messages, the update module 230 can update the historical message part multiple times, and the target message can be displayed multiple times on the conversation interface. historical news. For example, after the user performs a pull-down operation for the first time, the obtaining module 220 may obtain 10 target historical messages, and the conversation interface may display the above 10 target historical messages. After the second pull-down operation, the acquiring module 220 can acquire 10 target historical messages again, and the conversation interface can display the aforementioned 20 target historical messages in total. The 10 target historical messages acquired again may be the 10 target historical messages closest to the time points of the 10 target historical messages acquired for the first time.

When the user does not input a historical message loading instruction but inputs a question, step 407 may be performed. Step 407, in response to the user's question, obtain a question reply. In some embodiments, step 407 may be performed by the acquiring module 220 .

The user question may refer to a question that needs to be consulted by the user to the question answering robot. For example, as shown in FIG. 7D , the user's question may be "has the CT examination result been released?".

The question reply may refer to a reply about a user's question. For example, as shown in FIG. 7D , the question answering robot's reply to the user's question may be "the result is out, do you want to print the film? If you want to print, please click the "film print" button" and so on. In some embodiments, when the question answering robot is unable to answer the user's question, the question reply can be the system's default reply, for example, "Sorry, Xiao U is still learning, so I can't answer your question temporarily".

In some embodiments, there is a one-to-one correspondence between user questions and question replies. A user's question may have a first ID number, and a corresponding question reply may have a second ID number, and there is a corresponding relationship between the first and second ID numbers. Alternatively, user questions and corresponding question replies may have the same ID number. During the session interaction process, multiple user questions and multiple question replies may be included. The obtaining module 220 can make a one-to-one correspondence between these user questions and question replies. In some embodiments, for a user question, the corresponding question reply may include one or more pieces of information. One or more pieces of information are question replies to the user's question, for example, may have the same second ID number. In some embodiments, the question reply to the user's question may include a large amount of information, which is not suitable for being directly displayed on the conversation interface. Question replies can be displayed in the form of folds, links, etc. Users can click to expand the question reply, or click the link to jump to a new page to view the question reply.

In some embodiments, in response to user questions, the obtaining module 220 may obtain question replies in various ways. For example, the acquiring module 220 can acquire the question replies through the question-answer database 132 . Exemplarily, the acquisition module 220 may acquire the same or similar questions through the question-and-answer database based on the user's questions, and determine the replies corresponding to the same or similar questions as question replies. For another example, in response to a user's question, the obtaining module 220 may obtain a reply to the question through the network. Exemplarily, the user asks "has the CT inspection result been released?", and the acquisition module 220 may determine whether the user's inspection result exists in the inspection results of the CT device through the network. When there is the user's inspection result, the corresponding question reply can be "The result is out, do you want to print the film? If you want to print it, please click the "film print" button". When there is no inspection result for the user, the corresponding question reply can be "the inspection result has not yet come out, please wait patiently". In some embodiments, the acquisition module 220 may query the question-and-answer database or search the Internet in a fuzzy matching manner based on the user's question to obtain the answer to the question. When a corresponding question reply is matched, the acquisition module 220 may mark the reply type field flag, for example, mark it as a valid reply. When no question reply is matched, the obtaining module 220 may mark the reply type field, for example, as an invalid reply. For relevant content about valid replies and invalid replies, please refer to the related description of step 409 .

During the interactive session between the user and the Q&A robot, there may be network anomalies. In some embodiments, the acquiring module 220 can determine whether the network is abnormal. Furthermore, different processing methods are determined according to different judgment results (eg, abnormal, normal) of the network. In response to the judgment result of network abnormality, the triggering module 210 may trigger the interactive terminal to display a message abnormality prompt. In response to the judgment result that the network is normal, the obtaining module 220 may obtain a reply to the question.

The message exception prompt may refer to a message prompting the user that the current network is abnormal. For example, the message exception prompt is the text of "network abnormality" and the like. For another example, a red exclamation mark may be displayed near the message box corresponding to the question sent by the user, indicating that the sending of the message failed, and the Q&A robot will not reply to the question. When the user clicks on the text or icon of "Network Abnormality", the corresponding "Network Abnormality" disappears, and the current session message part deletes the question. After the network recovers, the triggering module 210 may re-add the user's question at the top of the message part of the current conversation.

When the network is normal, the triggering module 210 can trigger the interactive terminal to display the words or icons of "typing" in the conversation interface. The obtaining module 220 may obtain question replies through the question-answer database 132 and the like. When the obtaining module 220 obtains the corresponding question reply, the above-mentioned "inputting" text or icon disappears.

In some embodiments, the acquisition module 220 can judge whether the network is abnormal by whether the question answering robot responds to the question within the time threshold. The time threshold may refer to the maximum time for the question answering robot to reply to the user's question. For example, the time threshold is 3 seconds, etc. When the user asks a question and the question-answering robot fails to reply within the time threshold, the acquisition module 220 may determine the judgment result of the network abnormality. In some embodiments, the obtaining module 220 may analyze the connection status between the interactive terminal 120 and the processing device 110, so as to determine whether the network is abnormal.

Step 408, based on the reply to the question, update the message part of the current conversation, so that the reply to the question is displayed on the conversation interface. In some embodiments, step 408 may be performed by update module 230 .

In some embodiments, the update module 230 may add the reply to the question to the message part of the current conversation, so that the reply to the question is displayed on the conversation interface. For example, as shown in FIG. 7D, the conversation interface 121-4 may display a reply to a question.

Step 409 , judging whether the reply to the question includes a valid reply. In some embodiments, step 409 may be performed by the determining module 240 .

In some embodiments, the question replies may include valid replies or invalid replies and the like. A valid reply may refer to a reply that can answer a user's question. In some embodiments, the determining module 240 can determine whether it is a valid reply or an invalid reply according to the type of the question reply. If the question reply is a reply obtained by searching in a question-and-answer database or on the Internet, then the question reply is a valid reply. If the question reply is the default system reply, such as "Xiao U is still studying and cannot answer your question for the time being", the question reply will be an invalid reply. If the reply to the question is a like, dislike, comment, etc., the reply to the question will be an invalid reply.

In some embodiments, the determination module 240 may determine whether the reply to the question includes a valid reply according to the user's feedback on the reply to the question. For example, an evaluation option may be displayed on the conversation interface, allowing the user to evaluate the validity of the reply to the question. If the user provides a positive comment on the reply to the question, it may be determined that the reply to the question is a valid reply. For another example, if the user sends a reply message such as "OK, I understand" and "Got it" after receiving the reply to the question, the determining module 240 may determine the reply to the question as a valid reply. If the user's reply information is "I didn't ask this", "I don't understand what you mean", etc., the determination module 240 can determine the question reply as an invalid reply.

Step 410, in response to the judgment result that the question reply includes a valid reply, store the user question and the question reply pair in the historical message database. In some embodiments, step 410 may be performed by the determining module 240 .

By pairing user questions and effective question replies and storing them in the historical message database, invalid data can be avoided from occupying storage space, thereby saving data storage space and improving the efficiency of subsequent data management and retrieval. For example, when a subsequent user chooses to load historical information, the user's question and the corresponding effective question reply can be quickly searched from the historical database, thereby improving the efficiency of the user's review of historical information and the user's experience of using the question-and-answer robot.

The possible beneficial effects of the embodiments of this specification include but are not limited to: (1) In some embodiments of this specification, the message body can be used to store historical messages, system messages, current conversation messages, etc. Since various messages in the message body are stored based on a preset arrangement, the processing device 110 can quickly locate various messages and determine the display content of the conversation interface based on the message body without obtaining different types of messages from different sources. Moreover, the processing device 110 can uniformly process and manage data related to session interaction, and then can flexibly adjust a data structure to improve data storage and management efficiency. In the traditional conversation interaction method, since different messages correspond to different message sources or categories, and user questions and system replies are stored separately, it is necessary to link related messages through logic. During the conversation, different information needs to be retrieved from different sources, and logical judgments are required to find related data. Compared with the traditional conversational interaction method, the message body-based conversational interaction method has higher data management and storage efficiency, and can provide better conversational services. (2) Obtain different target system messages for different users, and then make the conversation interface display the target system messages, so that users can obtain the required system messages, and improve the user's use of the Q&A applet. (3) By arranging the historical message part 510 , the system message part 520 and the current session message part 530 in a preset order, it can help to quickly distinguish and locate messages, and improve data management and control efficiency. (4) In some embodiments, the target historical message only includes historical questions and their corresponding valid historical replies. Invalid replies, likes, dislikes, comments, system messages, etc. can not be loaded, and invalid information can be filtered to improve the user's viewing efficiency and experience. (5) The user can view other historical messages in chronological order by scrolling up. It is convenient for the user to check from the 1 historical message closest to the current time point. And, with this design, there is no need to calculate the scroll height when positioning the screen. (6) By pairing user questions and effective question replies and storing them in the historical message database, invalid data can be avoided from occupying storage space, thereby saving data storage space and improving the management and retrieval efficiency of subsequent data. For example, when a subsequent user chooses to load historical information, the user's question and the corresponding effective question reply can be quickly searched from the historical database, thereby improving the efficiency of the user's review of historical information and the user's experience of using the question-and-answer robot.

It should be noted that, the above descriptions about the process are only for illustration and description, and do not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to the flow under the guidance of this specification. However, such modifications and changes are still within the scope of this specification.

The basic concept has been described above, obviously, for those skilled in the art, the above detailed disclosure is only an example, and does not constitute a limitation to this description. Although not expressly stated here, those skilled in the art may make various modifications, improvements and corrections to this description. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

Meanwhile, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and/or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that two or more references to "an embodiment" or "an embodiment" or "an alternative embodiment" in different places in this specification do not necessarily refer to the same embodiment . In addition, certain features, structures or characteristics in one or more embodiments of this specification may be properly combined.

In addition, unless explicitly stated in the claims, the order of processing elements and sequences described in this specification, the use of numbers and letters, or the use of other names are not used to limit the sequence of processes and methods in this specification. While the foregoing disclosure has discussed by way of various examples some embodiments of the invention that are presently believed to be useful, it should be understood that such detail is for illustrative purposes only and that the appended claims are not limited to the disclosed embodiments, but rather, the claims The claims are intended to cover all modifications and equivalent combinations that fall within the spirit and scope of the embodiments of this specification. For example, although the system components described above may be implemented by hardware devices, they may also be implemented by a software-only solution, such as installing the described system on an existing server or mobile device.

In the same way, it should be noted that in order to simplify the expression disclosed in this specification and help the understanding of one or more embodiments of the invention, in the foregoing description of the embodiments of this specification, sometimes multiple features are combined into one embodiment, drawings or descriptions thereof. This method of disclosure does not, however, imply that the subject matter of the specification requires more features than are recited in the claims. Indeed, embodiment features are less than all features of a single foregoing disclosed embodiment.

In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of the embodiments use the modifiers "about", "approximately" or "substantially" in some examples. grooming. Unless otherwise stated, "about", "approximately" or "substantially" indicates that the stated figure allows for a variation of ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximations that can vary depending upon the desired characteristics of individual embodiments. In some embodiments, numerical parameters should take into account the specified significant digits and adopt the general digit reservation method. Although the numerical ranges and parameters used in some embodiments of this specification to confirm the breadth of the range are approximations, in specific embodiments, such numerical values are set as precisely as practicable.

Each patent, patent application, patent application publication, and other material, such as article, book, specification, publication, document, etc., cited in this specification is hereby incorporated by reference in its entirety. Application history documents that are inconsistent with or conflict with the content of this specification are excluded, and documents (currently or later appended to this specification) that limit the broadest scope of the claims of this specification are also excluded. It should be noted that if there is any inconsistency or conflict between the descriptions, definitions, and/or terms used in the accompanying materials of this manual and the contents of this manual, the descriptions, definitions and/or terms used in this manual shall prevail .

Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other modifications are also possible within the scope of this description. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims (9)

1. A conversational interaction method, characterized in that the method comprises: In response to a session request input by the user via the interactive terminal, the interactive terminal is triggered to present a session interface, and the display content of the session interface is determined based on a message body, wherein the message body includes a historical message part, a system message part, and a current session message part; obtaining target system messages in response to the session request; and updating the system message part based on the target system message, so that the target system message is displayed on the conversation interface; Acquiring an interaction instruction input by the user via the interaction terminal; updating the message body in response to the interaction instruction; Based on the updated message body, the interactive terminal is triggered to update the session interface. 2. The session interaction method according to claim 1, wherein the method further comprises: In response to the session request, determine whether the user has historical messages; In response to the judgment result that the user has historical messages, the interactive terminal is triggered to display a historical message loading prompt. 3. The session interaction method according to claim 2, wherein the interaction instruction includes a historical message loading instruction, and the method further includes: Responding to the historical message loading instruction, acquiring the target historical message based on the historical message database; Based on the target historical message, the historical message part is updated, so that the target historical message is displayed on the conversation interface. 4. The conversational interaction method according to claim 3, wherein the target historical message includes at least a pair of historical questions and valid historical replies. 5. The conversation interaction method according to claim 1, wherein the interaction instruction includes user questions, and the method further comprises: Responding to said user question, obtaining a question reply; and Based on the reply to the question, update the message part of the current conversation, so that the reply to the question is displayed on the conversation interface. 6. The conversational interaction method according to claim 5, wherein the method further comprises: judging whether the reply to the question includes a valid reply; In response to a determination result that the question reply includes a valid reply, the user question and the question reply pair are stored in a history message database. 7. The session interaction method according to claim 1, wherein the history message part, the system message part and the current session message part are arranged based on a preset order. 8. A conversational interaction system, characterized in that the system comprises: A triggering module, configured to trigger the interactive terminal to present a session interface in response to a session request input by the user via the interactive terminal, the display content of the session interface is determined based on a message body, wherein the message body includes a historical message part, a system message section and the current session message section; an acquisition module, configured to acquire target system messages in response to the session request; An update module, configured to update the system message part based on the target system message, so that the target system message is displayed on the conversation interface; The acquisition module is further configured to acquire an interaction instruction input by the user via the interaction terminal; The update module is further configured to update the message body in response to the interaction instruction; A triggering module, configured to trigger the interactive terminal to update the session interface based on the updated message body. 9. A computer-readable storage medium, wherein the storage medium stores computer instructions, and when the computer instructions are executed by a processor, the conversation interaction method according to any one of claims 1 to 7 is implemented.

Images