A priority-based human-computer dialogue processing method, system, and electronic device

Abstract

This specification discloses a priority-based human-computer dialogue processing method, system, and electronic device, which can accurately identify user inquiry intent, provide accurate and effective responses to user questions, and optimize user experience. The method is applied to an intelligent dialogue device and includes: acquiring user input data during the current conversation; matching the user input data with semantic libraries corresponding to various conversation types according to priority to determine the target semantics suitable for the user input data; generating response content based on the target semantics; and performing a conversation response based on the response content. The system includes: an information acquisition module, a semantic matching module, and a conversation response module. The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the human-computer dialogue processing method.

Description

Technical Field

[0001] This invention relates to the field of digital information processing technology, and more specifically to a priority-based human-computer dialogue processing method, system, and electronic device. Background Technology

[0002] Chatbots are currently very common in the market, distributed in various scenarios to assist customer service in responding to user questions. Among the existing technologies, various algorithms identify knowledge base questions without prioritization. This leads to problems such as low recognition rate of user descriptions, misidentification and mismatch of questions with different levels of intent in the knowledge base, and inability to distinguish the primary and secondary intent of user inquiries. As a result, chatbots have low response rates to user inquiries and even poor accuracy, ultimately causing user churn, a negative impression of chatbot products, and reduced usage. Summary of the Invention

[0003] In view of this, the embodiments of this specification provide a priority-based human-computer dialogue processing method, system, and electronic device, which can accurately identify the user's inquiry intent, provide accurate and effective responses to user questions, and optimize the user experience.

[0004] According to a first aspect, embodiments of the present invention provide a priority-based human-computer dialogue processing method, applied to an intelligent dialogue device, the method comprising:

[0005] Get user input data during the current session;

[0006] The user input data is matched with semantic libraries corresponding to various session types in order of priority to determine the target semantics that are suitable for the user input data.

[0007] Reply content is generated based on the target semantics, and a session response is performed based on the reply content.

[0008] Optional, multiple session types include:

[0009] The first session type O indicates that the current session is in the state where the intelligent dialogue device is waiting for a user's reply, or the intelligent dialogue device is continuously expressing itself.

[0010] The second session type, U, indicates that the current session is in a user-initiated expression phase.

[0011] The third conversation type S indicates that the current conversation is in the context where the intelligent dialogue device recognizes user input and enters the corresponding context;

[0012] The fourth conversation type, H, indicates that the current conversation is in a situation where the intelligent dialogue device receives content expressed by the user in a strong context and sets a response; and

[0013] The fifth conversation type, X, indicates that the current conversation is in a state of having no topic.

[0014] Optionally, the user input data is matched with semantic libraries corresponding to various session types in order of priority, including:

[0015] According to the priority order of the first session type O, the second session type U, the third session type S, the fourth session type H, and the fifth session type X, the user input data is matched sequentially with the semantic databases corresponding to the various session types until the target semantic that matches the user input data is selected from the semantic databases corresponding to the various types.

[0016] Optionally, matching the user input data with the semantic library corresponding to the second session type U includes:

[0017] The second semantics in the second semantic library corresponding to the second session type U are compared and matched with the user input data in turn.

[0018] After determining that a certain semantic item successfully matches the user input data, it is further determined whether the adapted semantic item has a corresponding matching stop flag;

[0019] If the semantic item does not have the matching stop identifier, then continue searching for a semantic item that matches the user input data in the second semantic library until all semantic items in the second semantic library have been compared and matched or a suitable semantic item with the matching stop identifier is matched.

[0020] Optionally, after matching the user input data with the semantic library corresponding to the second session type U, the method further includes:

[0021] If there are multiple adapted semantic items in the second semantic library, the multiple adapted semantic items are combined as the target semantics.

[0022] Optionally, before matching the user input data with the semantic library corresponding to the third session type S, the method further includes:

[0023] Determine whether the user input data has already been matched with the semantic library corresponding to the third session type S;

[0024] If a match has already been made, the user input data is matched against the semantic library corresponding to the fourth session type H;

[0025] If no match has been found, the user input data is matched against the semantic database corresponding to the third session type S.

[0026] Optionally, matching the user input data with the semantic library corresponding to the third session type S includes:

[0027] Determine the matching priority values ​​of multiple topic semantic items in the corresponding third semantic library for the third conversation type S relative to the user input data;

[0028] The topic semantic item with the highest matching priority value is selected as the adapted topic corresponding to the user input data;

[0029] The user input data is updated based on the adapted topic, and the updated user input data is matched again with the semantic library corresponding to the second session type U.

[0030] Optionally, the user input data includes user identification data, input content information, topic information matched by the input content information, semantic information, and output content information associated with the input content information.

[0031] In a second aspect, embodiments of this specification also provide a priority-based human-computer dialogue processing system applied to intelligent dialogue devices, the system comprising:

[0032] The information acquisition module is used to acquire user input data during the current session;

[0033] The semantic matching module is used to match the user input data with semantic libraries corresponding to various session types in order of priority, so as to determine the target semantics that are suitable for the user input data.

[0034] The conversation response module is used to generate response content based on the target semantics and to respond to the conversation based on the response content.

[0035] The system is used to perform the priority-based human-computer dialogue processing method as described in the first aspect.

[0036] In a third aspect, embodiments of this specification also provide a priority-based human-computer dialogue processing electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the priority-based human-computer dialogue processing method as described in the first aspect.

[0037] As can be seen from the above, the priority-based human-computer dialogue processing method, system, and electronic device provided in the embodiments of this specification have the following beneficial technical effects:

[0038] The aforementioned priority-based human-computer dialogue processing method, system, and electronic device divides the overall dialogue process into various dialogue types according to different dialogue scenarios. The corresponding overall semantic knowledge base is also divided into different dialogue semantic bases for different dialogue types. In the human-computer dialogue processing process, detailed user input data during the dialogue process is first obtained. The user input data is then matched with the semantic knowledge bases of different dialogue types according to different priorities. This allows for hierarchical multi-intent and multi-dimensional analysis of user input, enabling a comprehensive understanding of the user's input questions and main intentions. Consequently, accurate and effective responses can be provided based on the target semantics determined by the matching, thereby optimizing the user experience. Attached Figure Description

[0039] The features and advantages of the invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the invention in any way. In the drawings:

[0040] Figure 1 This specification illustrates a priority-based human-computer dialogue processing method provided by one or more optional embodiments;

[0041] Figure 2 This illustration shows a method for matching user input data with semantic libraries corresponding to multiple conversation types in a priority-based human-computer dialogue processing method provided by one or more optional embodiments of this specification;

[0042] Figure 3 This diagram illustrates a method for matching user input data with a second semantic database in one or more optional embodiments of a priority-based human-computer dialogue processing method provided in this specification.

[0043] Figure 4 This diagram illustrates a method for matching user input data with a third semantic database in one or more optional embodiments of a priority-based human-computer dialogue processing method provided in this specification.

[0044] Figure 5 This specification shows a schematic diagram of the structure of a priority-based human-computer dialogue processing system provided by one or more optional embodiments;

[0045] Figure 6 This specification illustrates a schematic diagram of a priority-based human-computer dialogue processing electronic device provided by one or more optional embodiments. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] Chatbots are currently very common in the market, deployed in various scenarios to assist customer service in responding to user inquiries. Existing technologies typically use various algorithms to identify questions in a knowledge base without prioritization. A unified algorithm model is generally used to retrieve and match the entire semantic knowledge base, directly determining the semantic result corresponding to the user's input question. This approach lacks differentiation between hierarchical relationships between different knowledge points and the identification of the primary and secondary intents of the user's inquiry. This leads to interference between similar questions with different intents, resulting in misidentification and mismatch of questions with different levels of intent in the knowledge base. Consequently, the chatbot fails to distinguish the primary and secondary intents of the user's inquiry, leading to low response rates and even poor accuracy. Ultimately, this results in user churn, a negative impression of the chatbot product, and reduced usage.

[0048] To address the aforementioned issues, the purpose of this technical solution is to propose a priority-based human-computer dialogue processing method. This method divides human-computer conversations into various conversation types and sets corresponding matching priorities. Semantic matching is then performed based on the semantic knowledge bases corresponding to different priority conversation types. This method performs multi-intent and multi-dimensional analysis on user input in a hierarchical manner, thereby comprehensively understanding the user's input questions and main intentions, and achieving accurate and effective responses.

[0049] The technical solutions in this specification will be described below with reference to specific embodiments.

[0050] In one aspect, embodiments of this specification provide a priority-based human-computer dialogue processing method.

[0051] like Figure 1 As shown, one or more optional embodiments of this specification provide a priority-based human-computer dialogue processing method applied to an intelligent dialogue device, the method comprising:

[0052] S1: Get user input data during the current session.

[0053] The user input data can be obtained using the information input port of the intelligent dialogue device. The user input data may include user identification data, input content information, topic information matched by the input content information, semantic information, and output content information associated with the input content information.

[0054] The user identification data can be collected during the initial stage of the human-computer intelligent session, when the user logs in, and the user identification data can be extracted and determined from it. The user identification data may include information such as user name, user attributes, and user identification code.

[0055] The user's original input information can be directly obtained through the information data port. Based on this, a semantic recognition algorithm can be used to identify and extract relevant topic information and semantic information from the input information. In addition, the output information of the intelligent dialogue device's responses during the human-computer dialogue process can also be obtained.

[0056] The user input data includes user identification data, input content information, topic information matched by the input content information, semantic information, and output content information associated with the input content information. It can comprehensively cover all aspects of information in the human-computer dialogue process. Based on the user input information, matching processing can determine the integrity of the conversation content information.

[0057] S2: The user input data is matched with semantic libraries corresponding to various session types in order of priority to determine the target semantics that are suitable for the user input data.

[0058] In optional embodiments of this specification, the overall conversation process is divided into multiple different conversation types according to different conversation scenarios. A corresponding priority is set based on the actual meaning of the conversation scenario for each type. Then, based on the priority, the data is sequentially matched with semantic libraries corresponding to each conversation type to determine the target semantics that best match the user input data. This approach enables hierarchical, multi-intent, and multi-dimensional analysis of user input, allowing for a comprehensive understanding of the user's questions and main intentions, and the determination of the target semantics that precisely corresponds to the user input data.

[0059] In some alternative embodiments, the various session types include:

[0060] The first session type O indicates that the current session is in the state where the intelligent dialogue device is waiting for a user's reply, or the intelligent dialogue device is continuously expressing itself.

[0061] The second session type, U, indicates that the current session is in a user-initiated expression phase.

[0062] The third conversation type S indicates that the current conversation is in the context where the intelligent dialogue device recognizes user input and enters the corresponding context;

[0063] The fourth conversation type, H, indicates that the current conversation is in a situation where the intelligent dialogue device receives content expressed by the user in a strong context and sets a response; and

[0064] The fifth conversation type, X, indicates that the current conversation is in a state of having no topic.

[0065] Based on the above-mentioned multiple session types, in some optional embodiments, the user input data can be matched with the semantic databases corresponding to the multiple session types in order of priority, and the following implementation methods can be adopted:

[0066] According to the priority order of the first session type O, the second session type U, the third session type S, the fourth session type H, and the fifth session type X, the user input data is matched sequentially with the semantic databases corresponding to the various session types until the target semantic that matches the user input data is selected from the semantic databases corresponding to the various types.

[0067] Figure 2 This is a schematic diagram illustrating a method for matching user input data with semantic libraries corresponding to various session types in order of priority. For example... Figure 2 As shown in some optional embodiments of this specification, the method for matching the user input data with semantic libraries corresponding to various session types may include the following steps:

[0068] S201: Match the user input data with the first semantic library corresponding to the first session type O.

[0069] First, the first semantic database corresponding to the first session type O is considered for matching. If a semantic item matching the user input data can be determined in the first semantic database, this matching semantic item is taken as the target semantic, and a response is generated based on this to conduct the session response. The first session type O indicates that the current session is in a situation where the intelligent dialogue device is waiting for a user response or the intelligent dialogue device is continuously expressing itself. For the first session type O, semantic parsing is relatively simple in strong context.

[0070] S202: If the user input data fails to match the first semantic library, then the user input data is matched with the second semantic library corresponding to the second session type U.

[0071] If the user input data fails to match a semantic item in the first semantic database, then the second semantic database corresponding to the second session type U, which has a lower priority, is used for matching. The second session type U represents a situation where the user is actively expressing themselves, a common situation in actual conversations, and the logical meaning contained in this session type is relatively complex. Considering the potentially complex logical meaning of user input data in the case of active user expression, multi-dimensional analysis and matching of the user input data is necessary.

[0072] Figure 3 This is a schematic diagram illustrating a method for matching the user input data with a second semantic library corresponding to the second session type U. For example... Figure 3 As shown, in some optional embodiments, matching against the second semantic library can be performed using the following steps:

[0073] S301: The multiple semantic entries in the second semantic library corresponding to the second session type U are sequentially compared and matched with the user input data.

[0074] S302: After determining that a certain semantic item successfully matches the user input data, further determine whether the adapted semantic item has a corresponding matching stop flag.

[0075] S302: If the semantic item does not have the matching stop identifier, then continue to search for a semantic item that matches the user input data in the second semantic library until all semantic items in the second semantic library have been compared and matched or a suitable semantic item with the matching stop identifier is matched.

[0076] When a semantic item in the second semantic database is determined to successfully match the user input data, this semantic item is not directly output as the target semantic item. Instead, it is further determined whether the semantic item has a corresponding matching stop flag. The matching stop flag can be set by staff. If the semantic item does not have a matching stop flag, the semantic item is recorded, and matching continues for the remaining semantic items in the second semantic item database until a successfully matched semantic item has the matching stop flag, or all semantic items in the second semantic database have been matched.

[0077] In the second semantic library, multiple rounds of matching are performed on the user input data. The number of adapted semantic items that match the user input data in the matching results may be 0, 1, or more: if the number is 0, it means that the user input data fails to match the second semantic library; if the number is 1, it means that there is only one adapted semantic item in the second semantic library that successfully matches the user input data, and this adapted semantic item is used as the target semantic. If there are multiple matched semantic items, the multiple matched semantic items are combined as the target semantic.

[0078] This approach, through multi-round matching of the user input data in the second semantic database, can obtain multi-dimensional information from the user input data, thereby enabling a more comprehensive understanding of the user's complete intent. The response content generated based on the target semantics determined in this way can provide more accurate and effective conversational responses to the user.

[0079] S203: If the user input data fails to match the second semantic library, then the user input data is matched with the third semantic library corresponding to the third session type S.

[0080] If the number of adapted semantic items for the user input data in the second semantic library is 0, it indicates that the user input data fails to match the second semantic library, and the user input data is then matched with the third semantic library corresponding to the third conversation type S. The third conversation type S indicates that the current conversation is in the case where the intelligent dialogue device recognizes the user input and enters the corresponding context.

[0081] The user input data was matched with the semantic databases corresponding to the first and second session types, and both matches were successful. This indicates that the topic information matched by the user input data in the early stages may have been biased. In this case, the user input data was then matched with the third semantic database corresponding to the third session type S, which has a lower priority.

[0082] Figure 4 This is a schematic diagram illustrating a method for matching the user input data with a third semantic library corresponding to the third session type S. For example... Figure 4 As shown, in some alternative embodiments, matching against a third semantic library can be performed using the following steps:

[0083] S401: Determine the matching priority value of multiple topic semantic items in the third semantic library corresponding to the third session type S relative to the user input data.

[0084] S402: Select the topic semantic item with the highest matching priority value as the adapted topic corresponding to the user input data.

[0085] S403: Update the user input data based on the adapted topic, and match the updated user input data again with the semantic library corresponding to the second session type U.

[0086] By matching the user input data with the third semantic database, the topic semantic item with the highest matching priority is selected as the adapted topic corresponding to the user input data. This allows for topic updates of the user input data and the identification of more accurate topic content. Subsequently, based on the updated user input information, it can be matched again with the semantic database corresponding to the second conversation type U. Based on the newly determined adapted semantic item, more accurate conversation responses with intent recognition can be achieved.

[0087] It should be noted that matching the user input data with the semantic library corresponding to the third session type S serves to update and adjust the topic content matched in the user input data. Based on the updated user input information, it is then matched again with the semantic library corresponding to the second session type U. If a match still cannot be successfully found after the topic update, it will not be matched with the semantic library corresponding to the third session type S. Therefore, before matching the user input data with the semantic library corresponding to the third session type S, it is necessary to first determine whether a match has already been made.

[0088] In a specific implementation, before matching the user input data with the semantic library corresponding to the third session type S, it is determined whether the user input data has already been matched with the semantic library corresponding to the third session type S. If it has already been matched, the user input data is matched with the semantic library corresponding to the fourth session type H; if it has not been matched, the user input data is matched with the semantic library corresponding to the third session type S.

[0089] S204: If the user input data fails to match the third semantic library, then the user input data is matched with the fourth semantic library corresponding to the fourth session type H.

[0090] The fourth conversation type, H, indicates that the current conversation is in a situation where the intelligent dialogue device receives content expressed by the user in a strong context and sets a response. Matching the user input data with the fourth semantic database to determine suitable semantic items can reduce matching interference when simple and complex semantics overlap, prevent the intelligent dialogue device from forcibly responding within pre-set logic, and improve the response accuracy of the intelligent dialogue device.

[0091] S205: If the user input data fails to match the fourth semantic library, then the user input data is matched with the fifth semantic library corresponding to the fifth session type X.

[0092] The user input data is matched sequentially with the semantic libraries corresponding to the first, second, third, and fourth conversation types. If no match is found, the user input data is then matched with the fifth semantic library corresponding to the fifth conversation type X, which has a lower priority. The fifth conversation type X indicates that the current conversation is in a no-topic state.

[0093] S3: Generate response content based on the target semantics, and respond to the session based on the response content.

[0094] In the priority-based human-computer dialogue processing method, the overall dialogue process is divided into various dialogue types according to different dialogue scenarios. The corresponding overall semantic knowledge base is also divided into different dialogue semantic bases for different dialogue types. In the human-computer dialogue processing process, detailed user input data in the dialogue process is first obtained. The user input data is matched with the semantic knowledge bases of different dialogue types according to different priorities. This allows for hierarchical multi-intent and multi-dimensional analysis of user input, enabling a comprehensive understanding of the user's input questions and main intentions. Furthermore, it allows for accurate and effective responses based on the target semantics determined by the matching, thereby optimizing the user experience.

[0095] It should be noted that the methods of one or more embodiments of this specification can be executed by a single device, such as a computer or server. The methods of this embodiment can also be applied in a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the methods of one or more embodiments of this specification, and the multiple devices will interact with each other to complete the method described.

[0096] It should be noted that the above description describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims may be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0097] Based on the same inventive concept, and corresponding to any of the above embodiments, this specification also provides a priority-based human-computer dialogue processing system.

[0098] refer to Figure 5 The priority-based human-computer dialogue processing system is applied to intelligent dialogue devices, and the system includes:

[0099] The information acquisition module is used to acquire user input data during the current session;

[0100] The semantic matching module is used to match the user input data with semantic libraries corresponding to various session types in order of priority, so as to determine the target semantics that are suitable for the user input data.

[0101] The conversation response module is used to generate response content based on the target semantics and to respond to the conversation based on the response content.

[0102] In a priority-based human-computer dialogue processing system provided in one or more optional embodiments of this specification, multiple dialogue types include: a first dialogue type O, indicating that the current dialogue device is waiting for a user's reply or the intelligent dialogue device is continuously expressing itself; a second dialogue type U, indicating that the current dialogue device is actively expressing itself; a third dialogue type S, indicating that the current dialogue device recognizes user input and enters a corresponding context; a fourth dialogue type H, indicating that the current dialogue device receives content expressed by the user in a strong context and sets a response; and a fifth dialogue type X, indicating that the current dialogue device is in a topicless state.

[0103] In a priority-based human-computer dialogue processing system provided in one or more optional embodiments of this specification, the semantic matching module is further configured to match the user input data sequentially with the semantic libraries corresponding to the various conversation types according to the priority order of the first conversation type O, the second conversation type U, the third conversation type S, the fourth conversation type H, and the fifth conversation type X, until the target semantic that matches the user input data is selected from the semantic libraries corresponding to the various conversation types.

[0104] In a priority-based human-computer dialogue processing system provided in one or more optional embodiments of this specification, the semantic matching module is further configured to compare and match multiple semantic items in the second semantic library corresponding to the second conversation type U with the user input data in sequence; after determining that a certain semantic item is successfully matched with the user input data, it further determines whether the adapted semantic item has a corresponding matching stop flag; if the semantic item does not have the matching stop flag, it continues to search for a semantic item that matches the user input data in the second semantic library until multiple semantic items in the second semantic library have been compared and matched or an adapted semantic item with the matching stop flag is matched.

[0105] In a priority-based human-computer dialogue processing system provided in one or more optional embodiments of this specification, the semantic matching module is further configured to combine multiple adapted semantic items as the target semantic when there are multiple adapted semantic items in the second semantic library.

[0106] In a priority-based human-computer dialogue processing system provided in one or more optional embodiments of this specification, the semantic matching module is further configured to determine whether the user input data has already been matched with the semantic library corresponding to the third session type S before matching the user input data with the semantic library corresponding to the third session type S; if it has been matched, the user input data is matched with the semantic library corresponding to the fourth session type H; if it has not been matched, the user input data is matched with the semantic library corresponding to the third session type S.

[0107] In a priority-based human-computer dialogue processing system provided in one or more optional embodiments of this specification, the semantic matching module is further configured to determine the matching priority values ​​of multiple topic semantic items in the third semantic library corresponding to the third conversation type S relative to the user input data; select the topic semantic item with the highest matching priority value as the adapted topic corresponding to the user input data; update the user input data based on the adapted topic, and match the updated user input data again with the semantic library corresponding to the second conversation type U.

[0108] In a priority-based human-computer dialogue processing system provided in one or more optional embodiments of this specification, the user input data includes user identification data, input content information, topic information matched by the input content information, semantic information, and output content information associated with the input content information.

[0109] For ease of description, the above apparatus is described in terms of function, divided into various modules. Of course, when implementing one or more embodiments of this specification, the functions of each module can be implemented in one or more software and / or hardware.

[0110] The apparatus described above is used to implement the corresponding methods in the foregoing embodiments and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0111] Figure 6 This embodiment illustrates a more specific hardware structure of an electronic device, which may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0112] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0113] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0114] The input / output interface 1030 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.

[0115] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0116] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0117] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0118] The electronic devices described above are used to implement the corresponding methods in the foregoing embodiments and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0119] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this disclosure also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the priority-based human-computer dialogue processing method as described in any of the above embodiments.

[0120] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0121] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the priority-based human-computer dialogue processing method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0122] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium can also include combinations of the above types of memory.

[0123] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0124] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0125] Those skilled in the art will understand that embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0126] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0127] This application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0128] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0129] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this disclosure (including the claims) is limited to these examples; within the framework of this disclosure, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of one or more embodiments of this specification as described above, which are not provided in detail for the sake of brevity.

[0130] Although this disclosure has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0131] One or more embodiments of this specification are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments of this specification should be included within the scope of protection of this disclosure.

Claims

1. A priority-based human-computer dialogue processing method, characterized in that, Applied to intelligent dialogue devices, the method includes: Get user input data during the current session; The user input data is matched with semantic libraries corresponding to various session types in order of priority to determine the target semantics that are suitable for the user input data. Generate response content based on the target semantics, and conduct session responses based on the response content; The various session types include: The second session type, U, indicates that the current session is in a user-initiated communication situation. Matching the user input data with the semantic library corresponding to the second session type U includes: The second semantics in the second semantic library corresponding to the second session type U are compared and matched with the user input data in turn. After determining that a certain semantic item successfully matches the user input data, it is further determined whether the adapted semantic item has a corresponding matching stop flag; If the semantic item does not have the matching stop identifier, then continue searching for a semantic item that matches the user input data in the second semantic library until all semantic items in the second semantic library have been compared and matched or a suitable semantic item with the matching stop identifier is matched.

2. The method according to claim 1, characterized in that, Multiple session types include: The first conversation type O indicates that the current conversation is in the state where the intelligent dialogue device is waiting for a user's reply, or the intelligent dialogue device is continuously expressing itself. The third conversation type S indicates that the current conversation is in the context where the intelligent dialogue device recognizes user input and enters the corresponding context; The fourth conversation type, H, indicates that the current conversation is in a situation where the intelligent dialogue device receives content expressed by the user in a strong context and sets a response; and The fifth conversation type, X, indicates that the current conversation is in a state of having no topic.

3. The method according to claim 2, characterized in that, The user input data is matched against semantic libraries corresponding to various session types in order of priority, including: According to the priority order of the first session type O, the second session type U, the third session type S, the fourth session type H, and the fifth session type X, the user input data is matched sequentially with the semantic databases corresponding to the various session types until the target semantic that matches the user input data is selected from the semantic databases corresponding to the various types.

4. The method according to claim 1, characterized in that, After matching the user input data with the semantic library corresponding to the second session type U, the method further includes: If there are multiple adapted semantic items in the second semantic library, the multiple adapted semantic items are combined as the target semantics.

5. The method according to claim 3, characterized in that, Before matching the user input data with the semantic library corresponding to the third session type S, the method further includes: Determine whether the user input data has already been matched with the semantic library corresponding to the third session type S; If a match has already been made, the user input data is matched against the semantic library corresponding to the fourth session type H; If no match has been found, the user input data is matched against the semantic database corresponding to the third session type S.

6. The method according to claim 5, characterized in that, Matching the user input data with the semantic library corresponding to the third session type S includes: Determine the matching priority values ​​of multiple topic semantic items in the corresponding third semantic library for the third conversation type S relative to the user input data; The topic semantic item with the highest matching priority value is selected as the adapted topic corresponding to the user input data; The user input data is updated based on the adapted topic, and the updated user input data is matched again with the semantic library corresponding to the second session type U.

7. The method according to claim 1, characterized in that, The user input data includes user identification data, input content information, topic information matched by the input content information, semantic information, and output content information associated with the input content information.

8. A priority-based human-computer dialogue processing system, characterized in that, The system, applied to intelligent conversational devices, includes: The information acquisition module is used to acquire user input data during the current session; The semantic matching module is used to match the user input data with semantic libraries corresponding to various session types in order of priority, so as to determine the target semantics that are suitable for the user input data. The conversation response module is used to generate response content based on the target semantics and to respond to the conversation based on the response content; The various session types include: The second session type, U, indicates that the current session is in a user-initiated communication situation. The semantic matching module is also used for: The second semantics in the second semantic library corresponding to the second session type U are compared and matched with the user input data in turn. After determining that a certain semantic item successfully matches the user input data, it is further determined whether the adapted semantic item has a corresponding matching stop flag; If the semantic item does not have the matching stop identifier, then continue searching for a semantic item that matches the user input data in the second semantic library until all semantic items in the second semantic library have been compared and matched or a suitable semantic item with the matching stop identifier is matched.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 7.

Images