CN115774776A - Man-machine conversation processing method and system and electronic equipment - Google Patents

Abstract

The specification discloses a man-machine conversation processing method, a man-machine conversation processing system and electronic equipment, which can fully and flexibly utilize historical data, improve the reply accuracy of logic complex sessions and long-term sessions, and optimize user experience. The method comprises the following steps: acquiring a memory data set of a target user and extracting a memory data subset corresponding to conversation input data of the target user from the memory data set; carrying out memory condition judgment on the memory data subset according to the session input data; and after the memory condition is judged to be passed, generating conversation output data matched with the conversation input data according to the memory data subset, and responding to the conversation input data. The system comprises a memory data acquisition module, a memory data screening module, a memory condition judgment module and a session response module. The electronic equipment comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor to realize the man-machine conversation processing method.

Description

Man-machine conversation processing method and system and electronic equipment

Technical Field

The invention relates to the technical field of session information processing, in particular to a man-machine conversation processing method, a man-machine conversation processing system and electronic equipment.

Background

The intelligent session system is also called as an intelligent session Agent or an intelligent chatting system. The system is a system for realizing language interaction with human beings by an artificial intelligence technology based on conversation recognition, natural language processing and conversation synthesis technology. The intelligent conversation system is mainly divided into a task-oriented conversation system and a non-task-oriented conversation system from the application scene, wherein the typical task-oriented conversation system comprises an intelligent conversation assistant and an intelligent telephone outbound system, and the typical non-task-oriented system comprises an intelligent sound box, a chat robot and the like.

Most of products of intelligent conversation robot types on the market at present have basic key point memory capacity, and most of products carry out context understanding through entity naming identification and slot filling, so that simple and multi-turn conversations can be carried out in real-time conversations. However, most intelligent dialogue robots cannot fully and flexibly utilize historical input information of users, the reply accuracy rate of dialogs with complex logics and dialogs with long time intervals is obviously reduced, and the memory ability of human-computer dialogs perceived by the users is weak, so that the user experience is influenced.

Disclosure of Invention

In view of this, embodiments of the present specification provide a human-machine conversation processing method, which can fully and flexibly utilize historical data, improve the reply accuracy of a logically complex session and a long-term session, and optimize user experience.

According to a first aspect, an embodiment of the present specification provides a human-machine conversation processing method, which is applied to an intelligent conversation device, and includes:

determining identity information of a target user, and acquiring a memory data set of the target user according to the identity information, wherein memory data in the memory data set is extracted from historical conversation contents and determined;

acquiring conversation input data of the target user in the current conversation process, and extracting a memory data subset corresponding to the conversation input data from the memory data set;

carrying out memory condition judgment on the memory data subset according to the session input data;

and after the memory condition is judged to be passed, generating conversation output data matched with the conversation input data according to the memory data subset, and responding to the conversation input data.

Optionally, the memory data set includes a plurality of user memory tags, and the user memory tag includes a plurality of fields, and each field is used for storing a piece of memory information.

Optionally, extracting a memory data subset corresponding to the session input data from the memory data set includes:

determining semantic constraints by performing semantic analysis on the session input data;

and screening the plurality of user memory labels in the memory data set according to the semantic constraint condition, and determining the memory data subset according to a screening result.

Optionally, the determining a memory condition for the memory data subset according to the session input data includes:

setting memory judgment conditions according to the session input data, wherein the memory judgment conditions comprise a memory search value condition, a memory support condition and a memory objection condition;

when the memory judgment condition is the memory search value condition, the memory condition judgment is performed on the memory data subset, and the memory condition judgment comprises the following steps:

comparing the numerical values in the memory data subset with the limited data in the memory search value condition, and judging that the memory condition is passed when the numerical values meet the requirement of the limited data;

when the memory determination condition is the memory support condition, performing memory condition determination on the memory data subset, including:

determining whether a target condition limited by the memory support condition exists in the memory data subset or not by searching and judging the memory data subset, and if so, judging that the memory condition is passed;

when the memory determination condition is the memory objection condition, performing memory condition determination on the memory data subset, including:

and determining whether a target condition limited by the memory objection condition exists in the memory data subset by carrying out search judgment on the memory data subset, and if not, judging that the memory condition is passed.

Optionally, after performing memory condition determination on the memory data subset according to the session input data, the method further includes: and updating the memory data set according to the session input data.

Optionally, updating the memory data set according to the session input data includes:

extracting key words and corresponding key information from the session input data by performing semantic analysis on the session input data;

comparing the keyword to a plurality of the fields in the memory dataset to determine whether a field corresponding to the keyword exists in the memory dataset;

responding to the fact that fields corresponding to the keywords exist in the memory data set, and updating the field contents of the fields according to the key information;

and responding to the fact that no field corresponding to the keyword exists in the memory data set, and creating a new user memory label in the memory data set according to the key information.

Optionally, updating the field content of the corresponding field according to the key information includes:

in response to the fact that the key information is numerical value type information, modifying the field content of the field according to the key information;

and in response to the fact that the key information is state type information, updating the field content of the field to be the key information or deleting the field corresponding to the keyword.

Optionally, after creating a new user memory tag in the memory data set according to the key information, the method further includes: setting time attribute information for a new user memory tag, wherein the time attribute information is used for limiting the effective time of the user memory tag.

In a second aspect, the present specification further provides a human-computer conversation processing system, where the system is applied to an intelligent conversation device, and the system includes:

the memory data acquisition module is used for determining the identity information of a target user and acquiring a memory data set of the target user according to the identity information, and memory data in the memory data set is extracted from historical conversation content and determined;

the memory data screening module is used for acquiring the session input data of the target user in the current session process and extracting a memory data subset corresponding to the session input data from the memory data set;

the memory condition judgment module is used for judging the memory condition of the memory data subset according to the session input data; and

and the conversation response module is used for generating conversation output data matched with the conversation input data according to the memory data subset and responding to the conversation input data after the memory condition is judged to be passed.

In a third aspect, the present specification also provides a human-computer conversation processing electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the human-computer conversation processing method according to the first aspect when executing the program.

As can be seen from the above, the man-machine conversation processing method, system and electronic device provided in the embodiments of the present specification have the following beneficial technical effects:

the method comprises the steps of obtaining a memory data set of a target user, selecting and screening a memory data subset associated with a current conversation theme from the memory data set, judging memory conditions of the memory data, matching corresponding conversation output data for the target user according to the memory data subset after the memory data subset is judged to pass, and replying by fully and flexibly utilizing historical data when replying for conversation input of the target user, so that the replying accuracy of logical complex conversations and long-term conversations can be improved, and user experience is optimized.

Drawings

The features and advantages of the present invention will be more clearly understood by reference to the accompanying drawings, which are illustrative and not to be construed as limiting the invention in any way, and in which:

FIG. 1 is a diagram illustrating a human-machine dialog processing method according to one or more alternative embodiments of the present disclosure;

FIG. 2 is a diagram illustrating a method for extracting a subset of memory data in a human-computer interaction processing method according to one or more alternative embodiments of the present disclosure;

FIG. 3 is a diagram illustrating a method for updating a memory data set in a human-machine interaction processing method according to one or more alternative embodiments of the present disclosure;

FIG. 4 is a block diagram of a human-machine dialog processing system according to one or more alternative embodiments of the present disclosure;

fig. 5 is a schematic diagram illustrating a structure of an electronic device for processing human-computer interaction according to one or more alternative embodiments of the present disclosure.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The intelligent session system is also called as an intelligent session Agent or an intelligent chatting system. The system is a system for realizing language interaction with human beings by an artificial intelligence technology on the basis of conversation recognition, natural language processing and conversation synthesis technology. The intelligent conversation system is mainly divided into a task-oriented conversation system and a non-task-oriented conversation system in an application scene, typical task-oriented conversation systems are intelligent conversation assistants and intelligent telephone outbound systems, and typical non-task-oriented systems are intelligent sound boxes, chat robots and the like.

Most of products of intelligent conversation robot types on the market at present have basic key point memory capacity, and most of products carry out context understanding through entity naming identification and slot filling, so that simple and multi-turn conversations can be carried out in real-time conversations. However, most intelligent dialogue robots cannot fully and flexibly utilize historical input information of users, the reply accuracy rate of dialogues with complex logics and dialogues with long time intervals can be obviously reduced, and the memory ability of man-machine dialogues which can be perceived by users is weak, so that the user experience is influenced.

In view of the above problems, an object of the embodiments of the present specification is to provide a man-machine conversation processing method, where historical memory data of a target user is obtained, and for session information input by the user in a current session process, memory condition determination is performed based on the historical memory data, and then a response is performed by matching corresponding session output data for the target user according to the historical memory data, so that the historical data can be fully and flexibly utilized, a reply accuracy of a logical complex session and a long-term session is improved, and user experience is optimized.

In view of the above, in a first aspect, the present specification provides a human-machine interaction processing method.

As shown in fig. 1, a man-machine conversation processing method provided in one or more alternative embodiments of this specification is applied to an intelligent conversation device, and includes:

s1: determining the identity information of a target user, and acquiring a memory data set of the target user according to the identity information, wherein the memory data in the memory data set is extracted from historical conversation contents and determined.

The identity information of the target user can be obtained when the target user logs in, and the identity information of the target user can comprise a user unique code, a name, a gender and the like. In the process of conversation, identity characteristic identification can be carried out on the conversation input by the target user, and therefore identity information of the target user is determined.

After determining the identity information of the target user, a memory data set corresponding to the target user may be obtained from a database according to the identity information. The memory data set is used for storing valuable historical data related to the target user, and the historical data can be valuable information related to the user extracted in the process of historical conversation, such as various related information of height, weight, occupation, physical health state and the like. It is understood that valuable information related to the target user may also be obtained during an initial login process of the user and stored in the memory data set.

S2: and acquiring conversation input data of the target user in the current conversation process, and extracting a memory data subset corresponding to the conversation input data from the memory data set.

The conversation input data of the target user in the current conversation process can be acquired, conversation semantics identification is carried out on the conversation input data, the conversation theme under the current conversation scene is determined, and therefore data content related to the conversation theme is screened out from the memory data set to serve as the memory data subset.

S3: and judging the memory condition of the memory data subset according to the session input data.

After the memory data subset associated with the conversation topic of the current conversation scene is screened and determined, further memory condition judgment needs to be performed on the memory data subset to determine the availability of the memory data subset. The specific judgment condition for judging the memory condition can be determined according to the actual semantic content of the session input data.

S4: and after the memory condition is judged to be passed, generating conversation output data matched with the conversation input data according to the memory data subset, and responding to the conversation input data.

And judging the memory condition of the memory data subset, and replying the session input data of the target user according to the memory information in the memory data subset by indicating that the memory information in the memory data subset is available.

In this case, the dialog output data conforming to the context logic can be executed in combination with the memory information of the memory data subset for reply by adopting corresponding greeting, expression or reply for different semantic contents of the dialog input data.

And if the judgment of the memory condition fails, the current session process cannot be successfully matched with the session input data of the target user, and other session contents can be executed, or response is performed on the session input data until the match is successful.

The man-machine conversation processing method comprises the steps of obtaining a memory data set of a target user, selecting and screening out a memory data subset associated with a current conversation theme from the memory data set, judging memory conditions of the memory data, and matching corresponding conversation output data for the target user according to the memory data subset to answer after the memory data subset passes the judgment.

In one or more alternative embodiments of the present specification, the memory data set includes a plurality of user memory tags, and the user memory tags include a plurality of fields, and each of the fields is used for storing a piece of memory information.

The user memory tag may be composed of a plurality of fields, each of which can store a piece of information, such as: user = xx, attribute = height, attribute value = xxx. The format of the field supports numbers, texts, non-aggregation variables, aggregation variables and arguments, and the multi-type memory storage is met. The user memory label can have various different field combination modes, and the different combinations of the fields can realize the flexibility and diversity of memory storage and extraction.

As shown in fig. 2, in a human-computer conversation processing method provided in one or more alternative embodiments of the present specification, extracting a memory data subset corresponding to the session input data from the memory data set includes:

s201: and determining semantic constraint conditions by performing semantic analysis on the session input data.

S202: and screening the plurality of user memory labels in the memory data set according to the semantic constraint condition, and determining the memory data subset according to a screening result.

By performing semantic analysis on the conversation input data, one or more subject words for characterizing the topic of the current conversation can be determined, and one or more subject words can be used as a constraint condition.

After the constraint condition is determined, searching and screening a plurality of user memory tags in the memory data set, and selecting the user memory tags meeting the constraint condition from the user memory tags to form the memory data subset.

In some optional embodiments, after the user memory tag meeting the constraint condition is selected, further screening fields in the user memory tag, and selecting the fields meeting the constraint condition to form the memory data subset.

Further, after the fields meeting the constraint condition are screened out, the field contents of the fields can be assigned to a user-defined variable. It should be noted that the specific content of the subject term constituting the constraint condition may be modified according to the actual session content.

The number of the user memory tags which are screened out according to the constraint conditions and meet the conditions can be multiple, under the condition, the field contents of the same field are respectively extracted from the multiple user memory tags, and the field contents of the same field extracted from the multiple different user memory tags are assigned to the multiple user-defined variables to form a set variable. By the method, the field contents of a plurality of fields can be searched and output at the same time, and are assigned to a plurality of user-defined variables at one time, so that the output of multiple information can be realized.

In one or more alternative embodiments of the present specification, in a human-computer conversation processing method, determining a memory condition for the memory data subset according to the session input data includes:

setting a memory determination condition according to the session input data, wherein the memory determination condition comprises a memory search value condition, a memory support condition and a memory objection condition, and then determining the memory data subset according to the set memory determination condition,

when the memory judgment condition is the memory search value condition, the memory condition judgment is performed on the memory data subset, and the memory condition judgment comprises the following steps:

and comparing the numerical values in the memory data subset with the limited data in the memory search value condition, and judging that the memory condition is passed when the numerical values meet the requirement of the limited data.

In some optional embodiments, the value assigned to the custom variable is determined according to the limit data in the memory search value condition.

According to the specific semantics of the session input data, when the numerical value of the self-defined variable is equal to the limited data, the judgment is passed; or when the value of the user-defined variable is not equal to the limit data, the judgment is passed; the determination of pass may also be made when the data of the custom variable conforms to the numerical range specified by the limiting data.

When the memory determination condition is the memory support condition, performing memory condition determination on the memory data subset, including:

and determining whether a target condition limited by the memory support condition exists in the memory data subset or not by searching and judging the memory data subset, and if so, judging that the memory condition is passed.

And memory support judgment, namely performing forward judgment on the content in the memory data subset, and adopting judgment logic of 'trigger if yes'. And when the target condition defined by the memory support condition exists in the memory data subset, judging that the memory condition is judged to be passed.

For example, if it is determined that the support of the memory "a" of the target user is required in the session to trigger the session according to the session input data, it is determined whether the memory information "a" exists in the filtered memory data subset, if so, it is determined that the memory condition is determined to be passed, and the session output data meeting the context logic is executed according to the memory information to reply.

When the memory determination condition is the memory objection condition, performing memory condition determination on the memory data subset, including:

and determining whether a target condition limited by the memory objection condition exists in the memory data subset or not by performing search judgment on the memory data subset, and if not, judging that the memory condition is passed.

And memory objection judgment, namely performing reverse judgment on the content in the memory data subset, and adopting judgment logic of 'no-event trigger'. And when the target condition defined by the memory support condition does not exist in the memory data subset, judging that the memory condition is judged to be passed.

For example, if it is determined that the target user memory object is required to trigger the session according to the session input data, whether the memory information "B" exists in the screened memory data subset is determined, if the memory information "B" does not exist, the memory condition is determined to be passed, and the session output data meeting the context logic is executed according to the memory information for replying.

As shown in fig. 3, a man-machine conversation processing method according to one or more alternative embodiments of the present specification, after performing a memory condition determination on the memory data subset according to the session input data, further includes: and updating the memory data set according to the conversation input data.

Updating the memory data set according to the session input data, comprising:

s301: and extracting keywords and corresponding key information from the session input data by performing semantic analysis on the session input data.

S302: comparing the keyword to a plurality of the fields in the memory dataset to determine whether a field corresponding to the keyword exists in the memory dataset.

S303: and responding to the existence of the field corresponding to the keyword in the memory data set, and updating the field content of the field according to the key information.

In the case that a field corresponding to the keyword exists in the memory data set, which indicates that the related information of the target user represented by the keyword is recorded in the memory data set, the key information and the field content of the field corresponding to the keyword can be compared to determine whether the two fields are consistent, and if the two fields are inconsistent, the related information of the target user is changed, which requires updating the field content according to the key information.

In some optional embodiments, when the field content is updated according to the key information, the information type of the key information may be determined first, and a corresponding update operation may be performed according to the information type.

And in response to the fact that the key information is numerical value type information, modifying the field content of the field according to the key information.

For example, the meaning of the keyword is the height of the target user, and the key information is a specific numerical value of the height of the target user, such as 175CM. And if the field content used for representing the height field in the memory data set corresponding to the target user is 170CM, the field content of the height field needs to be modified according to the key information, and the field content is updated to 175CM.

It should be noted that, the contents of the field may be modified in a self-increasing or self-decreasing manner according to the key information. For example, if the meaning of the keyword is the height of the target user and the content of the key information is '5 CM long and high', the field content of the height field is increased by 5CM on the basis of the original value according to the requirement of the key information.

And in response to the fact that the key information is state type information, updating the field content of the field to be the key information or deleting the field corresponding to the keyword.

The state class information may include dual state information and multi-state information. Wherein, the states represented by the two-state information conform to Boolean logic relationship, such as cold state and cold recovery state. The multi-status information indicates three or more statuses, such as a mild status, a moderate status, and a severe status indicating the severity of fever.

For dual-state information, if the key information is different from the field content, the field can be deleted. For example, the meaning of the keyword is a physical health state, the key information is that a cold of the target user is cured, and in the memory data set corresponding to the target user, the content of the field representing the physical health state is the cold of the user. In this case, the field may be deleted.

For multi-state information, if the key information is different from the field content, the field content may be updated to the key information.

S304: and in response to the fact that the fields corresponding to the keywords do not exist in the memory data set, creating a new user memory label in the memory data set according to the key information.

And if no field corresponding to the keyword exists in the memory data set, indicating that no related information is recorded in the memory data set, creating a new user memory tag in the memory data set to store the key information.

It should be noted that, the memory information in the new memory tag created in the memory data set is, by default, immediately valid. If the user inputs the name, the intelligent dialogue robot can answer the user name according to the data which is effective immediately when the intelligent dialogue robot inquires next round.

In some optional embodiments, after a new user memory tag is created in the memory data set according to the key information, time attribute information is further set for the new user memory tag.

The time attribute information may include an effective start time of the user memory tag. For example, when a new user memory tag is created, the user memory tag may be set to become effective after 10 minutes. The new user memory label records that the user is in the learning state, corresponding response content can not be obtained in the next round of inquiry, and the response content of the target user in the learning state can not be obtained until the inquiry is carried out after 10 minutes.

The time attribute information may further include an effective duration of the user memory tag. For example, when a new user memory tag is created, the user tag memory tag valid time may be set to 10 minutes.

For example, in the intelligent customer service conversation scenario, the user just calls to consult a certain service (writing memory, duration 10 minutes), the user calls again within 10 minutes, and the intelligent conversation robot can judge whether the problem of the last call is solved according to the memory. If the memory fails in more than 10 minutes, the intelligent conversation robot can not inquire the content of the last call and enters a normal flow.

It should be noted that the method of one or more embodiments of the present disclosure may be executed by a single device, such as a computer or a server. The method of the embodiment can also be applied to a distributed scene and completed by the mutual cooperation of a plurality of devices. In such a distributed scenario, one of the devices may perform only one or more steps of the method of one or more embodiments of the present disclosure, and the devices may interact with each other to complete the method.

It should be noted that the above description describes certain embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

Based on the same inventive concept, corresponding to the method of any embodiment, the embodiment of the specification further provides a man-machine conversation processing system.

Referring to fig. 4, the human-computer conversation processing system is applied to an intelligent conversation device, and the system includes:

the memory data acquisition module is used for determining the identity information of a target user and acquiring a memory data set of the target user according to the identity information, and memory data in the memory data set is extracted from historical conversation content and determined;

the memory data screening module is used for acquiring the session input data of the target user in the current session process and extracting a memory data subset corresponding to the session input data from the memory data set;

the memory condition judging module is used for judging the memory condition of the memory data subset according to the session input data; and

and the conversation response module is used for generating conversation output data matched with the conversation input data according to the memory data subset after the memory condition is judged to be passed and responding to the conversation input data.

In one or more alternative embodiments of the present specification, a human-computer dialog processing system is provided, where the memory data set includes a plurality of user memory tags, and the user memory tags include a plurality of fields, and each of the fields is used for storing a piece of memory information.

In one or more optional embodiments of the present disclosure, in a human-machine conversation processing system, the memory data filtering module is further configured to determine semantic constraints by performing semantic analysis on the session input data; and screening the plurality of user memory labels in the memory data set according to the semantic constraint conditions, and determining the memory data subset according to a screening result.

In one or more optional embodiments of the present specification, the memory condition determining module is further configured to set a memory determining condition according to the session input data, where the memory determining condition includes a memory search value condition, a memory support condition, and a memory objection condition; when the memory judgment condition is the memory search value condition, comparing the numerical values in the memory data subset with the limited data in the memory search value condition, and when the numerical values meet the requirement of the limited data, judging that the memory condition is passed; when the memory judgment condition is the memory support condition, determining whether a target condition limited by the memory support condition exists in the memory data subset or not by searching and judging the memory data subset, and if so, judging the memory condition to be passed; and when the memory judging condition is the memory objection condition, determining whether a target condition limited by the memory objection condition exists in the memory data subset by performing search judgment on the memory data subset, and if not, judging that the memory condition passes.

One or more optional embodiments of the present specification provide a human-machine dialog processing system, further including a memory data updating module, where the memory data updating module is configured to update the memory data set according to the session input data.

In one or more optional embodiments of the present specification, in a human-computer dialog processing system, the memory data updating module is further configured to extract keywords and corresponding key information from the conversational input data by performing semantic analysis on the conversational input data; comparing the keyword to a plurality of the fields in the memory dataset to determine whether a field corresponding to the keyword exists in the memory dataset; when a field corresponding to the keyword exists in the memory data set, updating the field content of the field according to the key information; and when the fields corresponding to the keywords do not exist in the memory data set, creating a new user memory label in the memory data set according to the key information.

In a man-machine conversation processing system provided in one or more optional embodiments of the present specification, the memory data updating module is further configured to modify, when the key information is value-type information, field content of the field according to the key information; and when the key information is state type information, updating the field content of the field into the key information or deleting the field corresponding to the keyword.

In a man-machine conversation processing system provided in one or more optional embodiments of the present specification, the memory data updating module is further configured to set time attribute information for a new user memory tag after the new user memory tag is created in the memory data set according to the key information, where the time attribute information is used to limit an effective time of the user memory tag.

For convenience of description, the above devices are described as being divided into various modules by functions, and are described separately. Of course, the functionality of the various modules may be implemented in the same one or more pieces of software and/or hardware in implementing one or more embodiments of the present description.

The apparatus of the foregoing embodiment is used to implement the corresponding method in the foregoing embodiment, and has the beneficial effects of the corresponding method embodiment, which are not described herein again.

Fig. 5 is a schematic diagram illustrating a more specific hardware structure of an electronic device according to this embodiment, where the device may include: a processor 1010, a memory 1020, an input/output interface 1030, a communication interface 1040, and a bus 1050. Wherein the processor 1010, memory 1020, input/output interface 1030, and communication interface 1040 are communicatively coupled to each other within the device via bus 1050.

The processor 1010 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits, and is configured to execute related programs to implement the technical solutions provided in the embodiments of the present disclosure.

The Memory 1020 may be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static storage device, a dynamic storage device, or the like. The memory 1020 may store an operating system and other application programs, and when the technical solution provided by the embodiments of the present specification is implemented by software or firmware, the relevant program codes are stored in the memory 1020 and called to be executed by the processor 1010.

The input/output interface 1030 is used for connecting an input/output module to input and output information. The i/o module may be configured as a component in a device (not shown) or may be external to the device to provide a corresponding function. The input devices may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output devices may include a display, a speaker, a vibrator, an indicator light, etc.

The communication interface 1040 is used for connecting a communication module (not shown in the drawings) to implement communication interaction between the present device and other devices. The communication module can realize communication in a wired mode (such as USB, network cable and the like) and also can realize communication in a wireless mode (such as mobile network, WIFI, bluetooth and the like).

Bus 1050 includes a path that transfers information between various components of the device, such as processor 1010, memory 1020, input/output interface 1030, and communication interface 1040.

It should be noted that although the above-mentioned device only shows the processor 1010, the memory 1020, the input/output interface 1030, the communication interface 1040 and the bus 1050, in a specific implementation, the device may also include other components necessary for normal operation. In addition, those skilled in the art will appreciate that the above-described apparatus may also include only those components necessary to implement the embodiments of the present description, and not necessarily all of the components shown in the figures.

The electronic device of the foregoing embodiment is used to implement the corresponding method in the foregoing embodiment, and has the beneficial effects of the corresponding method embodiment, which are not described herein again.

Based on the same inventive concept, corresponding to any of the above-described embodiment methods, the present disclosure also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the human-machine conversation processing method according to any of the above-described embodiments.

Computer-readable media, including both permanent and non-permanent, removable and non-removable media, for storing information may be implemented in any method or technology. The information may be computer readable instructions, data structures, modules of a program, 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 technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

The computer instructions stored in the storage medium of the foregoing embodiment are used to enable the computer to execute the human-computer interaction processing method described in any embodiment, and have the beneficial effects of the corresponding method embodiment, which are not described again here.

Those skilled in the art will appreciate that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium and can include the processes of the embodiments of the methods described above when executed. The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard Disk Drive (Hard Disk Drive, abbreviated as HDD), or a Solid State Drive (SSD); the storage medium may also comprise a combination of memories of the kind described above.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the various elements may be implemented in the same one or more pieces of software and/or hardware in the practice of the present application.

As will be appreciated by one skilled in the art, embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, the description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the description 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, and the like) having computer-usable program code embodied therein.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of other like elements in a process, method, article, or apparatus comprising the element.

The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

Those of ordinary skill in the art will understand that: the discussion of any embodiment above is meant to be exemplary only, and is not intended to intimate that the scope of the disclosure, including the claims, is limited to these examples; features from the above embodiments, or from different embodiments, may also be combined, steps may be implemented in any order, and there are many other variations of the different aspects of one or more embodiments of the present description, as described above, which are not provided in detail for the sake of brevity.

While the present disclosure has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations thereof will be apparent to those skilled in the art in light of the foregoing description. For example, other memory architectures, such as Dynamic RAM (DRAM), may use the discussed embodiments.

It is intended that the one or more embodiments of the present specification embrace all such alternatives, modifications and variations as fall within the broad scope of the appended claims. Therefore, any omissions, modifications, substitutions, improvements, and the like that may be made without departing from the spirit and principles of one or more embodiments of the present disclosure are intended to be included within the scope of the present disclosure.

Claims (10)

1. A man-machine conversation processing method is applied to an intelligent conversation device, and comprises the following steps:

determining identity information of a target user, and acquiring a memory data set of the target user according to the identity information, wherein memory data in the memory data set is extracted from historical conversation contents and determined;

acquiring conversation input data of the target user in the current conversation process, and extracting a memory data subset corresponding to the conversation input data from the memory data set;

carrying out memory condition judgment on the memory data subset according to the session input data;

and after the memory condition is judged to be passed, generating conversation output data matched with the conversation input data according to the memory data subset and responding to the conversation input data.

2. The method of claim 1, wherein the memory data set comprises a plurality of user memory tags, and the user memory tags comprise a plurality of fields, and each field is used for storing a piece of memory information.

3. The method of claim 2, wherein extracting a subset of the mnemonic data corresponding to the conversational input data from the set of mnemonic data comprises:

determining semantic constraints by performing semantic analysis on the session input data;

and screening the plurality of user memory labels in the memory data set according to the semantic constraint conditions, and determining the memory data subset according to a screening result.

4. The method of claim 2, wherein determining a memory condition for the subset of memory data based on the session input data comprises:

setting a memory judgment condition according to the session input data, wherein the memory judgment condition comprises a memory search value condition, a memory support condition and a memory objection condition;

when the memory judgment condition is the memory search value condition, the memory condition judgment is performed on the memory data subset, and the memory condition judgment comprises the following steps:

comparing the numerical values in the memory data subset with the limited data in the memory search value condition, and judging that the memory condition is passed when the numerical values meet the requirement of the limited data;

when the memory determination condition is the memory support condition, performing memory condition determination on the memory data subset, including:

determining whether a target condition limited by the memory support condition exists in the memory data subset or not by searching and judging the memory data subset, and if so, judging that the memory condition is passed;

when the memory determination condition is the memory objection condition, performing memory condition determination on the memory data subset, including:

and determining whether a target condition limited by the memory objection condition exists in the memory data subset or not by performing search judgment on the memory data subset, and if not, judging that the memory condition is passed.

5. The method of claim 2, further comprising, after performing a memory condition determination for the subset of memory data based on the conversational input data: and updating the memory data set according to the conversation input data.

6. The method of claim 5, wherein updating the memory data set based on the session input data comprises:

extracting key words and corresponding key information from the session input data by performing semantic analysis on the session input data;

comparing the keyword to a plurality of the fields in the memory dataset to determine whether a field corresponding to the keyword exists in the memory dataset;

responding to the fact that fields corresponding to the keywords exist in the memory data set, and updating the field contents of the fields according to the key information;

and responding to the fact that no field corresponding to the keyword exists in the memory data set, and creating a new user memory label in the memory data set according to the key information.

7. The method of claim 6, wherein updating the field content of the corresponding field according to the key information comprises:

in response to the fact that the key information is numerical value type information, modifying the field content of the field according to the key information;

and in response to the fact that the key information is state type information, updating the field content of the field to be the key information or deleting the field corresponding to the keyword.

8. The method of claim 6, after creating a new user memory tag in the memory dataset according to the key information, further comprising: setting time attribute information for a new user memory tag, wherein the time attribute information is used for limiting the effective time of the user memory tag.

9. A human-computer conversation processing system, applied to an intelligent conversation device, comprising:

the memory data acquisition module is used for determining the identity information of a target user and acquiring a memory data set of the target user according to the identity information, and memory data in the memory data set is extracted from historical conversation contents and determined;

the memory data screening module is used for acquiring the session input data of the target user in the current session process and extracting a memory data subset corresponding to the session input data from the memory data set;

the memory condition judgment module is used for judging the memory condition of the memory data subset according to the session input data; and

and the conversation response module is used for generating conversation output data matched with the conversation input data according to the memory data subset and responding to the conversation input data after the memory condition is judged to be passed.

10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1 to 8 when executing the program.

Images