CN115424624A

CN115424624A - Man-machine interaction service processing method and device and related equipment

Info

Publication number: CN115424624A
Application number: CN202211374769.0A
Authority: CN
Inventors: 王曦; 冯丽云
Original assignee: Shenzhen Renma Interactive Technology Co Ltd
Current assignee: Shenzhen Renma Interactive Technology Co Ltd
Priority date: 2022-11-04
Filing date: 2022-11-04
Publication date: 2022-12-02
Anticipated expiration: 2042-11-04
Also published as: CN115424624B

Abstract

The application relates to a method and a device for processing man-machine interaction service and related equipment. The method comprises the following steps: and if the fact that the user reply sentence of the user aiming at the first machine inquiry sentence in the currently processed man-machine conversation event is detected to be a positive reply is detected, whether reference historical user conversation events which have an association relation with the first event topic exist in the pre-stored historical user conversation events or not is inquired according to the first event topic of the currently processed man-machine conversation event. If the reference historical user conversation event exists, determining a target service item adapting to the requirements of the user according to the reference historical user conversation event and the currently processed man-machine conversation event; generating a target machine output statement according to the target service project; and sending the target machine output statement to the terminal equipment. By adopting the method, the user experience in the human-computer interaction process can be improved.

Description

Man-machine interaction service processing method and device and related equipment

Technical Field

The invention relates to the technical field of general data processing in the Internet industry, in particular to a man-machine interaction service processing method, a man-machine interaction service processing device and related equipment.

Background

With the continuous development of computer technology, the human-computer interaction technology has been gradually applied to various aspects of people's life, such as voice assistants in terminal devices, voice control engines in intelligent home appliances, and the like.

In the existing human-computer interaction process, a machine receives a demand or a question provided by a user and then gives a corresponding answer to the demand or the question. However, the question-and-answer process is too mechanized, and the machine replies with a single content and has low accuracy, so the user experience of the existing human-computer interaction technology is poor. Therefore, how to improve the user experience of the human-computer interaction technology has become one of the technical problems to be solved urgently.

Disclosure of Invention

The embodiment of the application provides a method and a device for processing a man-machine interaction service and related equipment.

In a first aspect, an embodiment of the present invention provides a service processing method for human-computer interaction, which is applied to a server of a human-computer interaction system, where the human-computer interaction system includes a terminal device for human-computer interaction between the server and a user, and the server includes a human-computer conversation engine; the method comprises the following steps: invoking the human-machine conversation engine to perform the following operations. Detecting whether a user answer sentence of the user for a first machine inquiry sentence in a currently processed human-machine dialog event is a positive answer, wherein the first machine inquiry sentence contains first service recommendation information for the requirement of the user in the currently processed human-machine dialog event. If the answer is positive, inquiring whether reference historical user conversation events which have an association relation with the first event theme exist in prestored historical user conversation events or not according to the first event theme of the currently processed man-machine conversation events, wherein the association relation comprises a time sequence and/or a place adaptation relation, the historical user conversation events comprise historical man-machine conversation events and/or historical call records, and the historical call records refer to call records which are answered by the terminal equipment and are authorized by the user to be cached and used. If the reference historical user dialogue event exists, determining a target service item which is adapted to the requirement of the user according to the reference historical user dialogue event and the currently processed man-machine dialogue event; generating a target machine output statement according to the target service project; and sending the target machine output statement to the terminal equipment. And if the reference historical user dialogue event does not exist, sending a conventional machine reply sentence aiming at the positive reply to the terminal equipment. If the answer is negative, second service recommendation information aiming at the requirements of the user is generated according to the user answer sentence; and generating a second machine query statement according to the second service recommendation information, and sending the second machine query statement to the terminal device.

In the embodiment of the application, a man-machine conversation engine is called to carry out man-machine interaction with a user, whether a user response sentence is a positive response or not is judged in the man-machine interaction process, and then a corresponding answer is given according to the requirement of the user by combining historical user conversation events. By adopting the method, in the process of man-machine interaction between the terminal equipment and the user, more accurate and flexible answers can be given according to the requirements of the user by combining historical user conversation events, so that the user experience is improved.

With reference to the first aspect, in a possible implementation manner, the determining, according to the reference historical user dialog event and the currently processed human-machine dialog event, a target service item that is adapted to requirements of the user includes: and detecting whether a first target service item needing to be updated exists in the service items preset by the user or not according to the reference historical user conversation event and the currently processed man-machine conversation event. If the first target service item is detected to exist, marking the first target service item as a target service item which is adaptive to the requirement of the user; determining a target service scene adapting to the requirements of the user according to the reference historical user conversation event and the currently processed man-machine conversation event; and judging whether a second target service item which is not subjected to conversation processing exists in a candidate service item set corresponding to the pre-stored target service scene. And if the second target service item exists, marking the second target service item as a target service item which is adaptive to the requirement of the user. And if the second target service item does not exist, not processing. If the first target service item is detected to be absent, determining a target service scene which is adapted to the requirements of the user according to the reference historical user dialogue event and the currently processed man-machine dialogue event; and judging whether a second target service item which is not subjected to conversation processing exists in a candidate service item set corresponding to the pre-stored target service scene. And if the second target service item exists, marking the second target service item as a target service item which is adaptive to the requirement of the user. And if the second target service item does not exist, not processing.

With reference to the first aspect, in one possible implementation manner, the reference historical user session event includes a reference historical call record; the step of detecting whether a first target service item needing to be updated exists in the service items preset by the user according to the reference historical user conversation event and the currently processed man-machine conversation event comprises the following steps: analyzing the currently processed human-machine conversation event to determine that the user is scheduled to watch a movie at a target theater during a target time period. If the service items scheduled by the user are determined to include the business overtime in the target time period by analyzing the reference historical call records, the detection result indicates that a first target service item needing to be updated exists in the service items scheduled by the user, and the target cinema watching movie is determined to be the first target service item needing to be updated. And if the reference historical call record is analyzed to determine that the reference historical call record does not contain the arrangement of the target time interval, or the arrangement of the target time interval is determined to be that the target cinema is used for watching the movie, the detection result indicates that the first target service item needing to be updated does not exist in the service items preset by the user.

With reference to the first aspect, in a possible implementation manner, the determining, according to the reference historical user dialog event and the currently processed human-machine dialog event, a target service scenario that is adapted to requirements of the user includes: a second event topic for the reference historical user dialog event is determined. Screening service scenes simultaneously containing the second event topic and the first event topic from a pre-stored service scene set, and determining the service scenes simultaneously containing the second event topic and the first event topic as target service scenes matched with the requirements of the user, wherein the service scene set comprises the corresponding relation between the service scenes and an event topic subset, and the event topic subset comprises a plurality of reference event topics involved in the corresponding service scenes.

With reference to the first aspect, in a possible implementation manner, before the determining whether a second target service item that is not subjected to dialog processing exists in a candidate service item set corresponding to the pre-stored target service scenario, the method further includes: and creating the candidate service item set according to the event topic subset of the target service scene.

With reference to the first aspect, in a possible implementation manner, the creating the set of candidate service items according to the event topic subset of the target service scenario includes: for each reference event topic in the event topic subset of the target service scenario, performing a candidate service item determination operation as follows. And acquiring a reference historical user dialog event which has an association relation with the currently processed reference event topic. And performing statistical analysis on the reference historical user conversation events which have the incidence relation with the currently processed reference event topic, and determining the reference service items which are received by the user at high frequency in the currently processed reference event topic. And determining candidate service items of the currently processed reference event topic according to the reference service items and/or the recommendation priority of the merchant belonging to the reference event topic. Creating the set of candidate service items according to a result of performing the candidate service item determination for each reference event topic in the subset of event topics for the target service scenario.

With reference to the first aspect, in a possible implementation manner, after determining, according to the reference historical user dialog event and the currently processed human-machine dialog event, a target service item that is adapted to the user's requirement, the method further includes: and analyzing the target service item as the lighting equipment is turned off. Determining a number of positive responses by the user to a third machine query statement from the reference historical user dialog event, wherein the third machine query statement contains third service recommendation information associated with the turn off lighting device; and judging whether the number of positive answers exceeds a preset number. If the number of times of the positive answers is judged to be equal to or greater than the preset number of times, generating a target control instruction for turning off the lighting equipment; and sending the target control instruction to the terminal equipment. And if the number of times of the positive answers is smaller than the preset number of times, not processing.

In a second aspect, an embodiment of the present invention provides a service processing apparatus for human-computer interaction, where the apparatus includes a human-computer dialog engine unit of a calling unit: and the calling unit is used for calling the man-machine conversation engine unit to execute the following operations. The system comprises a man-machine conversation engine unit, a first service recommendation unit and a second service recommendation unit, wherein the man-machine conversation engine unit is used for detecting whether a user reply sentence of a first machine inquiry sentence in a currently processed man-machine conversation event is a positive reply or not, and the first machine inquiry sentence contains first service recommendation information required by the user in the currently processed man-machine conversation event. If the answer is positive, whether reference historical user conversation events which have an association relation with the first event theme exist in prestored historical user conversation events or not is inquired according to the first event theme of the currently processed man-machine conversation events, wherein the association relation comprises a time sequence and/or a place adaptation relation, the historical user conversation events comprise historical man-machine conversation events and/or historical call records, and the historical call records refer to call records which are received by the terminal equipment and are subjected to user authorization caching and used. If the reference historical user dialogue event exists, determining a target service item which is adapted to the requirement of the user according to the reference historical user dialogue event and the currently processed man-machine dialogue event; generating a target machine output statement according to the target service project; and sending the target machine output statement to the terminal equipment. And if the reference historical user dialogue event does not exist, sending a conventional machine reply sentence aiming at the positive reply to the terminal equipment. If the answer is negative, second service recommendation information aiming at the requirements of the user is generated according to the user answer sentence; and generating a second machine query statement according to the second service recommendation information, and sending the second machine query statement to the terminal device.

With reference to the second aspect, in one possible implementation manner, the apparatus includes: and the man-machine conversation engine unit is used for detecting whether a first target service item needing to be updated exists in the service items predetermined by the user or not according to the reference historical user conversation event and the currently processed man-machine conversation event. If the first target service item is detected to exist, marking the first target service item as a target service item which is adaptive to the requirement of the user; determining a target service scene adapting to the user requirement according to the reference historical user dialogue event and the currently processed man-machine dialogue event; and judging whether a second target service item which is not subjected to conversation processing exists in a candidate service item set corresponding to the pre-stored target service scene. And if the second target service item exists, marking the second target service item as a target service item which is adaptive to the requirement of the user. And if the second target service item does not exist, not processing. If the first target service item is detected to be absent, determining a target service scene which is adapted to the requirements of the user according to the reference historical user dialogue event and the currently processed man-machine dialogue event; and judging whether a second target service item which is not subjected to session processing exists in a pre-stored candidate service item set in the target service scene. And if the second target service item exists, marking the second target service item as a target service item which is adaptive to the requirement of the user. And if the second target service item does not exist, not processing.

With reference to the second aspect, in one possible implementation manner, the apparatus includes: a human-machine dialog engine unit for analyzing the currently processed human-machine dialog events to determine that the user is scheduled to watch a movie at a target theater during a target time period. If the reference historical call records are analyzed to determine that the service items scheduled by the user include the business overtime in the target time period, the detection result indicates that a first target service item needing to be updated exists in the scheduled service items, and the target cinema watching movie is determined as the first target service item needing to be updated. And if the reference historical call record is analyzed to determine that the reference historical call record does not include the arrangement of the target time interval, or the arrangement of the target time interval is determined to be that the target cinema is used for watching the movie, the detection result indicates that the first target service item needing to be updated does not exist in the preset service items.

With reference to the second aspect, in one possible implementation manner, the apparatus includes: and the man-machine conversation engine unit is used for determining a second event theme of the reference historical user conversation event. Screening service scenes simultaneously containing the second event topic and the first event topic from a pre-stored service scene set, and determining the service scenes simultaneously containing the second event topic and the first event topic as target service scenes matched with the user requirements, wherein the service scene set comprises the corresponding relation between the service scenes and an event topic subset, and the event topic subset comprises a plurality of reference event topics involved in the corresponding service scenes.

With reference to the second aspect, in one possible implementation manner, the apparatus includes: and the man-machine conversation engine unit is used for creating the candidate service item set according to the event topic subset of the target service scene.

With reference to the second aspect, in one possible implementation manner, the apparatus includes: the man-machine conversation engine unit is used for executing the following candidate service item determination operation aiming at each reference event topic in the event topic subset of the target service scene, wherein the specific operation content comprises the following steps: and acquiring a reference historical user dialog event which has an association relation with the currently processed reference event topic. And performing statistical analysis on the reference historical user conversation events with the incidence relation in the currently processed reference event theme, and determining the reference service items which are received by the user at high frequency in the currently processed reference event theme. And determining candidate service items of the currently processed reference event topic according to the reference service items and/or the recommendation priority of the merchant belonging to the reference event topic. Creating the set of candidate service items according to a result of performing the candidate service item determination operation on each reference event topic in the subset of event topics of the target service scenario.

With reference to the second aspect, in one possible implementation manner, the apparatus includes: a human machine dialog engine unit for determining a number of positive answers by the user to a third machine query statement from the reference historical user dialog event, wherein the third machine query statement contains third service recommendation information associated with the turn-off lighting device; and judging whether the number of positive answers exceeds a preset number. If the positive answer times are judged to be equal to or larger than the preset times, generating a target control instruction for turning off the lighting equipment; and sending the target control instruction to the terminal equipment. And if the number of times of the positive answers is smaller than the preset number of times, not processing.

In a third aspect, an embodiment of the present application provides a computer-readable storage medium, where the computer-readable storage medium is used to store a computer program, and when the computer program runs on a computer, the computer is enabled to execute the service processing method for human-computer interaction provided in any one of the possible implementation manners of the first aspect, and also can achieve beneficial effects of the service processing method for human-computer interaction provided in the first aspect.

In a fourth aspect, embodiments of the present application provide an electronic device, which may include a processor and a memory, where the processor and the memory are connected to each other. The memory is used for storing a computer program, and the processor is configured to execute the computer program to implement the human-computer interaction service processing method provided by the first aspect, and can also implement the beneficial effects of the human-computer interaction service processing method provided by the first aspect.

By implementing the embodiment of the invention, in the process of man-machine interaction between a man-machine interaction system and a user, a man-machine conversation engine is called to detect whether a user response sentence of the user aiming at a first machine inquiry sentence is an affirmative response, if the user response sentence is the affirmative response, a server can inquire a reference historical user conversation event which has an incidence relation with a first event topic according to the currently processed man-machine conversation event topic, and further determine a service item which is adapted to the requirement of the user according to the reference historical user conversation event and the currently processed man-machine conversation event, so as to generate a target machine output sentence; if the answer is negative, the server can generate second service recommendation information aiming at the requirement of the user, so as to generate a target machine inquiry statement. By adopting the method, in the process of man-machine interaction, the server can give corresponding answers aiming at the requirements of the user by combining with reference historical user conversation events, so that the flexibility and the accuracy of machine output sentences are improved, and the user experience is improved.

Drawings

In order to more clearly illustrate the embodiments or technical solutions in the prior art of the present invention, the drawings used in the description of the embodiments or prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.

Fig. 1 is a schematic structural diagram of a human-computer interaction system provided in an embodiment of the present application;

fig. 2 is a schematic flowchart of a service processing method for human-computer interaction according to an embodiment of the present disclosure;

FIG. 3 is a schematic diagram of dialog contents of a human-computer interaction provided by an embodiment of the present application;

FIG. 4 is a schematic diagram of another structure of a human-computer interaction system according to an embodiment of the present application;

FIG. 5 is a schematic diagram of a triplet provided by an embodiment of the present application;

FIG. 6 is a schematic diagram illustrating still another dialog content of a human-computer interaction provided in an embodiment of the present application;

fig. 7 is another schematic flowchart of a service processing method for human-computer interaction according to an embodiment of the present application;

FIG. 8 is a schematic structural diagram of a human-computer interaction service processing apparatus according to an embodiment of the present disclosure;

fig. 9 is a schematic structural diagram of an electronic device provided in an embodiment of the present application.

Detailed Description

In order to make the technical solutions of the present application better understood, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. 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 application.

The terms "first," "second," and the like in the description and claims of the present application and in the foregoing drawings are used for distinguishing between different objects and not for describing a particular sequential order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps is not limited to only those steps recited, but may alternatively include other steps not recited, or may alternatively include other steps inherent to such process, method, article, or apparatus.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein may be combined with other embodiments.

In the existing human-computer interaction process, a machine receives a demand or a question provided by a user and then provides a corresponding answer for the demand or the question, however, the process of asking and answering is too mechanized and strongly depends on the user to provide the demand or the question actively, so that the conversation process is not flexible and intelligent enough, and the user experience is poor. Therefore, the technical problem to be solved by the application is as follows: how to improve the user experience of the human-computer interaction process.

Referring to fig. 1, fig. 1 is a schematic structural diagram of a human-computer interaction system according to an embodiment of the present disclosure. As shown in fig. 1, the human-computer interaction system 10 includes a server 11 and a terminal device 12. The server 11 is in communication connection with the terminal device 12, the server 11 comprises a man-machine conversation engine supporting man-machine interaction, and the server 11 can call the man-machine conversation engine to perform man-machine interaction with a user through the terminal device 12. In this embodiment, the server 11 may be a server, or a server cluster composed of a plurality of servers, or a cloud computing service center, and the terminal device 12 may be an intelligent terminal that can be used by a user for human-computer interaction, such as a smart phone, a tablet computer, a portable notebook computer, a wearable device, a head-mounted device, and a vehicle-mounted terminal, and the implementation forms of the server 11 and the terminal device 12 are not particularly limited.

It should be noted that the "man-machine conversation event" referred to in the embodiment of the present application may be a man-machine conversation recorded by the server 11 and performed with the user through the terminal device 12, and the man-machine conversation content corresponding to the man-machine conversation event may include at least one machine output sentence and at least one user reply content. The "event topic" of a certain man-machine conversation event may be the intention of the user involved in this man-machine conversation event. The "service scenario" may be the specific life scenario in which the user was at the time the "human-machine dialog event" occurred. A "service item" is a specific event that the server has drawn up for a certain event topic. The "service recommendation information" may be specific recommendation information that the server plans for a certain service item, and the specific recommendation information may include information such as a specific occurrence time and a specific occurrence location of a certain service item. For convenience of understanding, the concepts of "event topic", "service scenario", "service item", "service recommendation information", and the like, which are referred to in the embodiments of the present application, are described exemplarily. For example, it is assumed here that a service scene corresponding to a certain man-machine interaction event is an appointment scene, the event topic of the man-machine interaction event may be eating, watching a movie, and the like, the service items that may exist in the man-machine interaction may be a dinner reservation, a movie ticket reservation, and the like, and the service recommendation information that may exist in the man-machine interaction event may be "two-person position for reserving six o 'clock to restaurant" and "movie ticket for reserving eight o' clock movie theater".

The following describes a method for processing a service through human-computer interaction provided by an embodiment of the present application.

Referring to fig. 2, fig. 2 is a flowchart illustrating a man-machine interaction service processing method according to an embodiment of the present application, where the method is applied to a server 11 in a man-machine interaction system 10 shown in fig. 1, and specifically may be executed by the server 11 invoking a man-machine conversation engine. As shown in fig. 2, the method may specifically include the steps of:

s201, detecting whether a user reply sentence of the user to the first machine inquiry sentence in the currently processed man-machine conversation event is a positive reply.

If yes, go to step S202;

if the answer is negative, step S205 is executed.

In some possible embodiments, for the above step S201, the server 11 may first detect whether the user reply sentence of the user for the first machine inquiry sentence in the currently processed man-machine interaction event is an affirmative reply. It should be noted that the first machine query statement includes first service recommendation information for a user's request in the current human-machine interaction event.

Specifically, the server 11 may first obtain a user reply sentence of the user to the first machine query sentence in the currently processed man-machine dialog event, then detect the user reply sentence of the user to the first machine query sentence by using a technology such as semantic recognition, and determine whether the user reply sentence includes a positive word such as "may", "good", or "row", or determine whether the user reply sentence includes a negative word such as "not", or "not using". If the server 11 determines that the word indicating affirmative is included, it may determine that the user response sentence is an affirmative response; if the word indicating negative is judged to be contained, the user response sentence can be determined as a negative response.

For example, please refer to fig. 3, fig. 3 is a schematic diagram of a dialog content of a human-computer interaction according to an embodiment of the present application. As shown in fig. 3, the first machine query statement in the currently processed man-machine interaction event is "recommend movie theater B, and optionally movie ticket of movie C at eight clicks, the unit price is 298 yuan, and you need to buy it. "and, the user reply statement for the first machine inquiry statement by the user is" buy bar ". Further, the server 11 may determine that the user reply sentence to the first machine query sentence is a positive reply by detecting the user reply sentence of the user to the first machine query sentence in the currently processed man-machine dialog event. It should be understood that, in this man-machine conversation event, the first service recommendation information for the user's needs includes "go to B theater to see eight dots of C movie".

It should be understood that the first machine query statement may be any one of the machine query statements present in the currently processed human machine dialog event that contains a service recommendation that is adapted to the user's needs. Preferably, the first machine query statement is a first machine query statement in the currently processed human-machine interaction event.

S202, according to a first event theme of the currently processed man-machine conversation event, whether a reference historical user conversation event which has an association relation with the first event theme exists in the pre-stored historical user conversation events or not is inquired.

If the user dialog event with the reference history exists, executing step S203;

if it is found that the reference historical user dialog event does not exist, step S204 is executed.

In some possible embodiments, for the step S202, the server 11 may query whether there is a reference historical user dialog event in the pre-stored historical user dialog events that has an association relationship with the first event topic according to the first event topic of the currently processed human-computer dialog event. It should be noted that the historical user session events may include historical man-machine session events and/or historical call records, where the historical call records refer to call records that are answered by the terminal device and are subject to user authorization for caching and use.

It should be noted that the association relationship may be one or more of a time sequence, a location adaptation relationship, or a person relationship. Specifically, if the information contained in the currently processed human-computer conversation event and the reference historical user conversation event has a time sequence due to the arrangement of the user, the association relationship existing between the currently processed human-computer conversation event and the reference historical user conversation event is a time sequence relationship. If the man-machine conversation event processed currently and the reference historical user conversation event both contain the same place position information, the association relationship existing between the man-machine conversation event processed currently and the reference historical user conversation event is a place adaptation relationship. If the currently processed man-machine conversation event and the reference historical user conversation event both relate to the same person, the association relationship existing between the currently processed man-machine conversation event and the reference historical user conversation event is a character relationship.

In a specific implementation, the server 11 may determine the first event topic of the currently processed man-machine conversation event by using a semantic recognition technology or the like. Further, the server 11 may query whether there is a reference historical user dialog event having the above-mentioned association relationship with the above-mentioned first event topic in the pre-stored historical user dialog events according to the above-mentioned first event topic.

For example, it is assumed that the man-machine interaction content corresponding to the currently processed man-machine interaction event includes a man-machine interaction sentence "reserve movie tickets at eight o 'clock in city D tomorrow" and the server 11 can determine the first event topic of the currently processed man-machine interaction event as "movie at eight o' clock" through the man-machine interaction sentence by using a technology such as semantic recognition. It is assumed that the pre-stored man-machine interaction content corresponding to the historical user interaction event includes a man-machine interaction sentence such as "movie tickets reserved for six points", and further, the server 11 may determine the event topic of the historical user interaction event as "movie watched at six points" through the man-machine interaction sentence by using a technology such as semantic recognition. Since the event topic of the historical user dialog event and the first event topic of the currently processed man-machine dialog event have a time sequence due to the arrangement of the user, it can be determined that the event topic of the historical user dialog event and the first event topic of the currently processed man-machine dialog event have a time sequence relationship. Further, the server 11 may determine the above-mentioned historical user dialog events as reference historical user dialog events.

For example, it is assumed here that the man-machine interaction content corresponding to the currently processed man-machine interaction event includes a man-machine interaction sentence "ticket booking 5 o 'clock to D city" in tomorrow, and the server 11 can determine the first event topic of the currently processed man-machine interaction event as "5 o' clock to D city" by using the man-machine interaction sentence by using a technique such as semantic recognition. It is assumed that the pre-stored man-machine conversation contents corresponding to the historical user conversation events include a man-machine conversation sentence "booking a network appointment of eight points in city D tomorrow", and the server 11 may further determine the event topic of the historical user conversation events as "booking a vehicle of eight points in city D" through the man-machine conversation sentence by using a technology such as semantic recognition. Since the event topic of the historical user dialog event and the first event topic of the currently processed man-machine dialog event both have the location of the city D, it can be determined that the event topic of the historical user dialog event and the first event topic of the currently processed man-machine dialog event have a location adaptation relationship. Further, the server 11 may determine the above-mentioned historical user dialog events as reference historical user dialog events.

For example, it is assumed here that the man-machine interaction content corresponding to the currently processed man-machine interaction event includes a man-machine interaction term "booking air tickets for tomorrow and X", and the server 11 can determine that the first event topic of the currently processed man-machine interaction event is "booking air tickets for and X" by using the man-machine interaction term by using a technique such as semantic recognition. It is assumed that the pre-stored man-machine conversation contents corresponding to the historical user conversation events include man-machine conversation sentences such as "reserve tomorrow and dinner of X in restaurant a", and further, the server 11 may determine the event topic of the historical user conversation events as "reserve and dinner of X and dinner" through the man-machine conversation sentences by using semantic recognition and other technologies. Since the event topic of the historical user dialog event and the first event topic of the currently processed man-machine dialog event both have the character of "X", it can be determined that the event topic of the historical user dialog event has a location adaptive relationship with the first event topic of the currently processed man-machine dialog event. Further, the server 11 may determine the above-mentioned historical user dialog events as reference historical user dialog events.

S203, determining a target service item adaptive to the requirement of the user according to the reference historical user conversation event and the currently processed man-machine conversation event; generating a target machine output statement according to the target service project; and sending the target machine output statement to the terminal device.

In some possible embodiments, the server 11 may determine a target service item that adapts to the user's needs based on the reference historical user dialog events and the currently processed human-machine dialog events. Further, the server 11 may generate a target device output sentence from the target service item, and transmit the target device output sentence to the terminal device 12.

Specifically, the server 11 may first detect whether there is a first target service item to be updated in the service items that have been scheduled by the user according to the reference historical user session event and the currently processed man-machine session event. It should be noted that the first target service item to be updated may be a service item which has a conflict and cannot be completed according to a predetermined schedule, and is determined according to the currently processed man-machine conversation event and the reference historical user conversation event.

Illustratively, the reference historical user session events include reference historical call records that are answered by the end device 12 and authorized for buffering and use by the user. It is assumed here that the server 11 can determine that the user's schedule at the target time period is to go to the target cinema for watching a movie, based on the man-machine conversation events currently processed. Further, if the server 11 analyzes the reference calendar call records to determine that the service items scheduled by the user include going to the company overtime in the target time interval, the detection result is to determine that a first target service item needing to be updated exists in the service items scheduled by the user, where the first target service item is going to a target theater to watch a movie; if the server 11 analyzes the reference historic call records to determine that the reference historic call records do not contain the schedule of the target time interval, or if the server 11 analyzes to determine that the schedule of the target time interval is that the target cinema is going to watch the movie, the detection result is that the first target service item needing to be updated does not exist in the service items preset by the user.

On the one hand, if the server 11 detects that there is a first target service item to be updated in the service items that have been predetermined by the user according to the reference historical user session event and the currently processed man-machine session event, the first target service item may be marked as a target service item that is adapted to the user's requirement.

Further, the server 11 may determine a target service scenario adapting to the user's requirement according to the reference historical user dialog event and the currently processed man-machine dialog event.

Specifically, the server 11 may determine the second event topic of the reference historical user dialog event according to the reference historical user dialog event. Then, the server 11 may screen a service scenario including both the second event topic and the first event topic of the currently processed human-computer conversation event from a pre-stored service scenario set, and determine the service scenario including both the second event topic and the first event topic as a target service scenario adapted to the user's requirement. It should be noted here that the service scenario set includes a corresponding relationship between a service scenario and an event topic subset, and the event topic subset includes a plurality of reference event topics involved in the corresponding service scenario.

For example, it is assumed that the first event topic of the currently processed human-computer interaction event is meal, and the service scenes included in the pre-stored service scene set include an appointment scene and a sleep scene, when the server 11 determines that the second event topic of the reference historical interaction event is watching a movie according to the reference historical user interaction event, if the server 11 determines that the appointment scenes in the pre-stored service scene set include both the first event topic and the second event topic, the appointment scene may be determined as a target service scene that is adapted to the user's requirement.

Further, the server 11 may determine whether a second target service item that has not undergone session processing exists in the pre-stored candidate service item set corresponding to the target service scenario. If the server 11 determines that the second target service item exists, marking the second target service item as a target service item adapted to the requirement of the user; and if the second target service item does not exist, not processing.

For example, in combination with the above example, when the server 11 detects and determines that there is a first target service item that needs to be updated in the service items that the user has subscribed to, the server 11 may determine, according to the reference historical user session event and the currently processed man-machine session event, a target service scenario that is adapted to the user's requirement as a work scenario. Further, the server 11 may determine whether a second target service item that has not been subjected to the session processing exists in the candidate service item set corresponding to the pre-stored working scenario. Assuming that the candidate service item set corresponding to the pre-stored working scenario includes a taxi-to company and a purchase target dinner, the server 11 may analyze whether there are user session events for the taxi-to company and the purchase target dinner in the reference historical user session events. If the server 11 analyzes and judges the reference historical user session event to determine that there is a user session event for the service item of the purchase target dinner but there is no user session event for the service item of the taxi-taking company, it may be determined that the service item of the taxi-taking company is the second target service item which is not subjected to the session processing, and the server 11 marks the second target service item as the target service item which meets the requirement of the user. If the server 11 analyzes and judges the reference historical user session event to determine that there is a user session event for both the taxi-to-company service and the purchase target dinner service, it may be determined that the second target service item does not exist and the server 11 does not perform the processing.

Optionally, the target service items adapted to the requirements of the user may be one or more. The processing sequence of the target service items adapted to the requirements of the users can determine the user acceptance times of the target service items adapted to the requirements of the users according to reference historical user conversation events or big data analysis, so that the target service items adapted to the requirements of the users can be sorted according to the user acceptance times from large to small. The server 11 may subsequently process the target service items in the order to meet the user's requirements.

On the other hand, if the server 11 detects that there is no first target service item to be updated in the service items predetermined by the user according to the reference historical user dialog event and the currently processed man-machine dialog event, it may determine a target service scenario adapted to the user's requirement according to the reference historical user dialog event and the currently processed man-machine dialog event. Further, the server 11 may determine whether a second target service item that has not undergone session processing exists in the pre-stored candidate service item set corresponding to the target service scenario. If the second target service item exists, marking the second target service item as a target service item which is adaptive to the requirement of the user; and if the second target service item does not exist, not processing.

For example, in combination with the above example, when the server 11 detects that there is no first target service item that needs to be updated in the service items that the user has subscribed to, the server 11 may determine, according to the above reference historical user dialog events and the above currently processed man-machine dialog events, that the target service scenario that matches the user's requirement is an appointment scenario. Further, the server 11 may determine whether a second target service item that has not undergone session processing exists in the candidate service item set corresponding to the pre-stored appointment scene. Assuming that the candidate service item set corresponding to the pre-stored appointment scenario includes going to a target restaurant to eat and going to a target theater to watch a movie, the server 11 may analyze whether there are user session events for the two service items in the reference historical user session events. If the server 11 analyzes the reference historical user session event and determines that there is a user session event for the service item for the destination restaurant to eat but there is no user session event for the service item for the destination cinema to watch movies, it may be determined that the service item for the destination cinema to watch movies is the second destination service item that is not subjected to the session processing, and the server 11 marks the second destination service item as the destination service item that is adapted to the user's needs. If the server 11 analyzes the reference history user session event to determine that there is a user session event for both the service items of eating at the destination restaurant and watching movie at the destination theater, it may be determined that the second destination service item does not exist and the server 11 does not perform the process.

In an alternative embodiment, the server 11 may create the set of candidate service items according to the event topic subset of the target service scenario. In a specific implementation, the server 11 may perform a candidate service item determination operation on each reference event topic in the event topic subset of the target service scenario to create the candidate service item set. The candidate service item determination operation is exemplarily described below by taking any one of the event topics in the event topic subset of the target service scenario as an example. Specifically, the server 11 may obtain the reference historical user dialog events associated with the reference event topic. Then, the server 11 performs statistical analysis on the reference historical user dialog events associated with the reference event topic, and may determine the reference service items that the user has received in the reference event topic at a high frequency. Further, the server 11 may determine the candidate service item of the reference event topic according to the reference service item and/or the recommendation priority of the merchant belonging to the reference event topic. Then, the server 11 may create a candidate service item set corresponding to the target service scenario according to the candidate service item corresponding to each reference event topic in the event topic subset of the target service scenario.

For example, assuming that the currently processed reference event topic is meal, the server 11 may obtain a reference historical user session event associated with meal, and then perform statistical analysis on the above reference historical user session event associated with meal, and may determine that the reference service items that the user has frequently accepted in the event topic of meal are three reference service items, namely meal at restaurant a, meal at restaurant B and meal at restaurant C. Further, the server 11 can determine candidate service items of the event subject of the meal according to the three reference service items, including meal to restaurant a, meal to restaurant B, and meal to restaurant C.

For another example, assuming that the currently processed reference event topic is meal, the server 11 may obtain a reference historical user session event associated with meal, and then perform statistical analysis on the above reference historical user session event associated with meal, and may determine that the reference service items that the user has received in the event topic frequently include three reference service items, namely meal for restaurant a, meal for restaurant B, and meal for restaurant C. The server 11 can determine that the merchants belonging to the subject matter of the above-mentioned eating event include three restaurants, namely, restaurant D, restaurant E and restaurant F, further, the server 11 can determine the number of times that the above-mentioned three restaurants are recommended by the user based on the big data analysis, so that the recommendation priorities of the above-mentioned three restaurants can be determined to be ranked from high to low as restaurant E, restaurant D and restaurant F, and the server 11 can determine that the candidate service items of the subject matter of the above-mentioned eating event include dining to restaurant a, dining to restaurant B, dining to restaurant C, dining to restaurant D, dining to restaurant and dining to restaurant F.

Further, after the server 11 determines a target service item adapted to the user's requirement according to the reference historical user dialog event and the currently processed man-machine dialog event, the server 11 may generate a target machine output statement according to the target service item, and send the target machine output statement to the terminal device 12.

In specific implementation, the server 11 performs semantic recognition according to the target service item to obtain a first triple. It should be appreciated that the first triplet described above can characterize the user intent.

Illustratively, the expression form of the triplet may be { r (x, y) }, where x refers to an entity at one end of the triplet, y refers to an entity at the other end of the triplet, and r refers to a semantic/syntactic relationship between entities at two ends of the triplet, and the semantic/syntactic relationship is directional, and the triplet { r (x, y) }maybe understood as a semantic/syntactic relationship r between x and y. For example, referring to fig. 4, fig. 4 is a schematic diagram of a triplet provided in the embodiment of the present application, and as shown in fig. 4, semantic recognition is performed on a statement "mingming walking to eat room", and three triplets including { noumenal subject (mingming, walking) }, { dependency (walking, go) }, { direct object (go, eat room) } can be obtained. In the following, a simple description is given by taking a triplet { noumenon (xiaoming, walking) } as an example, where "xiaoming" is an entity x, "walking" is an entity y, "noumenon" is a semantic/grammatical relation r, and the direction is pointed to "xiaoming" by "walking", and the meaning is represented as: the noun subject of "walking" is "Xiaoming".

The semantic/syntactic relations may include, among other things, dependencies in the natural language processing tool Stanford CoreNLP. For example, the verb (give) - - > direct object (dobj), the main component (usually the verb) - - > subject (nsubj) of the sentence, the dependency (depondent, dep), and the dynamic modifier (mmod) are dependent.

Further, the server 11 may determine a target human-machine interaction scenario from a plurality of human-machine interaction scenarios included in a preset human-machine interaction scenario according to the first triple, and determine a machine output content included in the target human-machine interaction scenario as a target machine output statement. It should be noted that each of the multiple human-computer interaction scenarios corresponds to a second triple. And the second triple corresponding to the target man-machine conversation plot is matched with the first triple.

Specifically, the server 11 may determine, according to the first triple, a second triple that matches the first triple from among second triples corresponding to the multiple human-computer interaction scenarios. Further, the server 11 may determine the man-machine interaction scenario corresponding to the second triple as a target man-machine interaction scenario, and determine the machine output content included in the target man-machine interaction scenario as a target machine output statement.

For the convenience of explaining the process of determining the target human-computer interaction scenario by the server 11, please refer to fig. 5, and fig. 5 is a schematic view of another structure of a human-computer interaction system provided in the embodiment of the present application, as shown in fig. 5, the human-computer interaction system includes a processing server for a session in human-computer interaction and a user terminal device. The processing server for the conversation in the man-machine interaction comprises a knowledge base and a man-machine conversation engine, wherein the knowledge base is used for storing a man-machine conversation scenario, the man-machine conversation engine is used for performing man-machine interaction with user terminal equipment according to the man-machine conversation scenario, and the user terminal equipment is used for displaying an interactive conversation interface.

The man-machine conversation scenario comprises a plurality of man-machine conversation scenarios. The specific single human-computer dialog scenario comprises a machine response strategy, a skipping scenario and a scenario skipping identifier, the machine response strategy comprises an output machine statement, the skipping condition comprises a semantic triple representing an input statement of an expected user, the triple comprises a first entity, a second entity and an incidence relation between the first entity and the second entity, the incidence relation comprises a semantic relation or a syntactic relation, and the scenario skipping identifier is used for indicating the human-computer dialog scenario which needs to be skipped to when the current scenario skipping condition is met.

Wherein the first triple is composed of a semantic/syntactic relationship and two knowledge nodes, the knowledge nodes include a first knowledge node and a second knowledge node, and the knowledge nodes are single entities. The second triplet is composed of a semantic/syntactic relationship and two template nodes, the template nodes include a first template node and a second template node, and the template nodes are a set composed of entities having a common reference relationship. The entities refer to words or phrases. Here, the matching of the first triplet with the second triplet corresponding to the plurality of human-computer interaction scenarios means: the semantic/syntactic relationship between the first triplet and the second triplet is the same, and a first knowledge node of the first triplet belongs to the first template node, and a second knowledge node of the first triplet belongs to the second template node.

Specifically, the semantic/syntactic relations are the same, which means that the relation type and the direction of the relation type of the connection relation between the entities at the two ends of the triple are the same. For example, a triplet (1) { noumenon (xiaoming, walking) }, a triplet (2) { noumenon (reddish, swimming) }, where the semantic/grammatical relation r of the triplet (1) and the triplet (2) is "noumenon", and the direction is that the entity x points to the entity y, the semantic/grammatical relation of the triplet (1) and the triplet (2) is determined to be the same.

To assist understanding, the matching mechanism of the first triple and the second triple is described below by a specific example. There is a general self-learning system for semantic recognition in the server, the general self-learning system exists in a knowledge base, the knowledge base has undergone a lot of self-learning and historical training, there exists at least one second triple in the knowledge base, x and y of { r (x, y) } in the second triple are both in the form of sets, and the entities in each set have a common reference relationship. For example, x can be a set of (Xiaoming + Xiaohong + Xiaohua + He + her) entities all having a common reference relationship. Similarly, y can be a set of (walk + walk), then r is the "noun" and the direction is from y to x. At this time, if one of the first triples extracted from the target service item is { noumenon (mingmen, walkie) }, it is determined that the first triplet matches with the second triplet, and the semantic recognition result includes the first triplet { noumenon (mingmen, walkie) }. It is to be understood that the currently processed user input content may include a plurality of entities, a plurality of semantic/syntactic relations and a plurality of first triples, and the above simple example is only used to assist understanding of the scheme disclosed in the embodiment of the present application, and does not constitute any limitation to the present application.

Further, the server 11 transmits the target machine output sentence to the terminal device 12.

And S204, sending a conventional machine reply sentence for the positive reply to the terminal equipment.

In some possible embodiments, if the server 11 queries that there is no reference historical user dialog event in the pre-stored historical user dialog events that has an association relationship with the first event topic, the server 11 sends a conventional machine reply sentence for the positive reply to the terminal device 12. It should be understood that the conventional machine reply sentence described above for the above-described positive reply does not include the service recommendation information. For example, it is assumed here that the first machine query sentence in the man-machine dialog event currently processed by the server 11 is "recommend B movie theater, and select eight-click movie ticket of C movie, the unit price is 298 yuan, and you need to buy it. "the user reply sentence of the user to the above first machine inquiry sentence is a positive reply" buy bar ". And the server 11 inquires that there is no reference historical user dialog event in the pre-stored historical user dialog events which has an association relation with the first event topic according to the first event topic of the current processed human-computer dialog event, the server sends the conventional machine reply sentence aiming at the positive reply sentence to the terminal device 12, that is, "good" and has been purchased successfully. ", the conventional machine reply statement does not include service recommendation information.

S205, generating second service recommendation information aiming at the requirements of the user according to the user response sentence; and generating a second machine inquiry sentence according to the second service recommendation information, and sending the second machine inquiry sentence to the terminal equipment.

In some possible embodiments, after determining that the user reply sentence of the user to the first machine query sentence is a negative reply, the server 11 may generate second service recommendation information for the requirement of the user according to the user reply sentence. Further, the server 11 may generate a second device query statement according to the second service recommendation information, and transmit the second device query statement to the terminal device 12. It should be appreciated that the second machine query is different from the service recommendation information contained in the first machine query statement described above.

For example, referring to fig. 6, fig. 6 is a schematic diagram of another dialog content of a human-machine interaction provided by an embodiment of the present application, and it is assumed that the first machine query statement in the human-machine dialog event currently processed by the server 11 is "need help you reserve dinner of restaurant a", as shown in fig. 6. The dishes in restaurant A are good. "the user's answer sentence to the first machine query sentence is" don't do not do it, eat at home this evening. ". After the server 11 detects that the user reply sentence to the first machine query sentence in the current human-machine dialog event is a negative reply, the server 11 may generate second service recommendation information for the user's requirement according to the user reply sentence. Assuming that the second service recommendation information for the user's needs is to purchase the food in store B, the server 11 may generate a second machine query sentence "need to help you purchase the food in store B" according to the second service recommendation information. And transmits the second machine inquiry sentence to the terminal device 12.

In the embodiment of the application, the server can analyze and predict the current requirements of the user by combining the responses of the user to the service recommendation information pushed by the user, determine new service recommendation information according to the analysis and prediction results, and further give corresponding answers to the requirements of the user, so that the server can more flexibly perform human-computer interaction with the user, and the user experience in the human-computer interaction process is improved.

Referring to fig. 7, fig. 7 is a schematic flowchart of a human-computer interaction service processing method according to an embodiment of the present application. It should be understood that, in the embodiment of the present application, step S206, step S207, and step S208 are performed after step S203 described above. As shown in fig. 7, the service processing method for human-computer interaction may specifically include the steps of:

and S206, determining the number of positive answers of the user to the third machine inquiry sentence according to the reference historical user conversation event, and judging whether the number of positive answers exceeds the preset number.

In some possible embodiments, the server 11 may determine the number of positive responses of the user to the third machine query statement from the reference historical user dialog events. Further, the server 11 may determine whether the number of positive responses exceeds a preset number. It should be noted that the third machine inquiry statement includes third service recommendation information associated with the target service, and it should be understood that the third service recommendation information indicates the operating state information of the target electrical device. For example, assuming that the target service item is to turn off the lighting device, the third machine query statement includes third service recommendation information associated with turning off the lighting device, and may include turning off a desk lamp or turning off a room lamp.

Specifically, it is assumed here that the analysis target service item is to turn off the lighting device. The server 11 may determine the number of times the user has answered in the affirmative for the third machine query statement based on the reference historical user dialog events. Further, the server 11 determines whether the number of positive responses exceeds a preset number. If the server 11 determines that the number of positive responses is equal to or greater than the preset number, a target control instruction for turning off the lighting device may be generated, and the target control instruction may be sent to the terminal device 12; if the server 11 determines that the number of positive responses is less than the preset number, no processing is performed.

S207, if the number of times of the positive answer is equal to or larger than the preset number of times, generating a target control instruction for the target service item; and sending the target control instruction to the terminal equipment.

In some possible embodiments, the server 11 may generate the target control instruction for the target service item if it is determined that the number of times of positive responses of the user to the third machine query statement is equal to or greater than a preset number of times. Further, the server 11 may transmit the target control command to the terminal device 12. Here, it should be noted that the target control command is used for controlling the operating state of the target electrical device.

For example, it is assumed here that the target service item in the currently processed human-machine dialog event that matches the user's requirement is to turn off the lighting device, and specifically, the third machine query statement is "need help you turn off the desk lamp". "the user responds affirmatively to this third machine inquiry statement as" required ". ", assume that the preset number of times is 8. The server 11 may determine, according to the reference historical user dialog event, that the number of times of positive responses of the user to the third machine query sentence is 10 times, and if the number of times exceeds the preset number of times, the server 11 may generate a target control command for turning off the lighting device according to the third machine query sentence, where the target control command is assumed to be turning off the desk lamp. Further, the server 11 sends the target control instruction for turning off the table lamp to the terminal device 12, so as to turn off the table lamp.

And S208, if the judgment result shows that the number of positive responses is less than the preset number, no processing is carried out.

In some possible embodiments, if the server 11 determines that the number of times of positive responses of the user to the third machine query sentence is less than the preset number of times, the server does not perform the processing. For example, it is assumed here that the target service item in the currently processed human-computer conversation event that matches the user's requirement is to turn off the lighting device, and specifically, the third machine query statement is "need help you turn off the desk lamp". "the user responds affirmatively to this third machine inquiry statement as" required ". ", assume that the preset number of times is 8. The server 11 may determine, according to the reference history user dialog event, that the number of times of positive responses of the user to the third machine query sentence is 5 times, which is less than the preset number of times, and the server 11 does not perform processing.

In an alternative embodiment, after determining that the number of positive responses of the user to the third machine query statement is less than the preset number, the server 11 may generate fourth service recommendation information according to the target service item corresponding to the third machine query statement, and further may generate a fourth machine query statement according to the fourth service recommendation information. Then, the server 11 may determine whether the user reply sentence to the above-described fourth machine inquiry sentence by the user is an affirmative reply. If it is determined that the user response sentence to the fourth device inquiry sentence by the user is an affirmative response, a control command for controlling the operating state of the target electrical appliance indicated by the fourth service recommendation information may be generated based on the fourth service recommendation information. It should be understood that the fourth service recommendation information described above indicates the operating state information of the target electrical device.

In the embodiment of the application, the server can determine the number of positive answers of the user to the user answer sentence of the machine inquiry sentence, and further judge whether the number of positive answers exceeds the preset number, and when the number of positive answers exceeds the preset number, the server can generate a target control instruction for a target service item according to the target service item corresponding to the machine inquiry sentence, so that the function of controlling the intelligent electrical equipment is realized, and the user experience is improved.

Please refer to fig. 8, fig. 8 is a schematic structural diagram of a service processing apparatus for human-computer interaction according to an embodiment of the present application, which is applied to a server of a human-computer interaction system, the human-computer interaction system includes the server and a terminal device for human-computer interaction by a user, where the server includes a human-computer dialog engine. As shown in fig. 8, the service processing device for human-computer interaction may include: a calling unit 81 and a human-machine dialog engine unit 82.

In a specific implementation, the invoking unit 81 is configured to invoke the human-machine dialog engine unit 82 to perform the following operations. The human-machine dialog engine unit 82 is configured to detect whether a user reply sentence of a first machine query sentence, which contains first service recommendation information required by the user in the currently processed human-machine dialog event, of the user in the currently processed human-machine dialog event is an affirmative reply. If the answer is positive, whether reference historical user conversation events which have an association relation with the first event theme exist in prestored historical user conversation events or not is inquired according to the first event theme of the currently processed man-machine conversation events, wherein the association relation comprises a time sequence and/or a place adaptation relation, the historical user conversation events comprise historical man-machine conversation events and/or historical call records, and the historical call records refer to call records which are received by the terminal equipment and are subjected to authorized cache and use by the user. If the reference historical user dialogue event exists, determining a target service item which is adapted to the requirement of the user according to the reference historical user dialogue event and the currently processed man-machine dialogue event; generating a target machine output statement according to the target service project; and sending the target machine output statement to the terminal equipment. And if the inquiry shows that the reference history user dialogue event does not exist, sending a conventional machine response sentence aiming at the positive response to the terminal equipment. If the answer is negative, generating second service recommendation information aiming at the requirement of the user according to the user answer sentence; and generating a second device query sentence according to the second service recommendation information, and transmitting the second device query sentence to the terminal device.

In an alternative embodiment, the human-machine conversation engine unit 82 is configured to detect whether there is a first target service item that needs to be updated in the service items that the user has subscribed to, according to the reference historical user conversation event and the currently processed human-machine conversation event. If the first target service item is detected to exist, marking the first target service item as a target service item which is adaptive to the requirement of the user; determining a target service scene adapting to the user's requirement according to the reference historical user dialogue event and the currently processed man-machine dialogue event; and judging whether a second target service item which is not subjected to conversation processing exists in a candidate service item set corresponding to the pre-stored target service scene. And if the second target service item exists, marking the second target service item as a target service item which is adaptive to the requirement of the user. And if the second target service item does not exist, not processing. If the first target service item does not exist, determining a target service scene which is adapted to the requirements of the user according to the reference historical user conversation event and the currently processed man-machine conversation event; and judging whether a second target service item which is not subjected to conversation processing exists in a pre-stored candidate service item set in the target service scene. And if the second target service item exists, marking the second target service item as a target service item which is adaptive to the requirement of the user. And if the second target service item does not exist, not processing.

In an alternative embodiment, the human machine dialog engine unit 82 is adapted to analyze the currently processed human machine dialog events to determine that the user is scheduled to watch a movie at the target theater during the target time period. If the reference historical call records are analyzed to determine that the service items scheduled by the user include the service items going to the company overtime in the target time period, the detection result indicates that a first target service item needing to be updated exists in the scheduled service items, and the target cinema watching movie is determined to be the first target service item needing to be updated. If the reference historical call records are analyzed to determine that the reference historical call records do not include the arrangement of the target time interval, or the arrangement of the target time interval is determined that the target cinema is going to watch the movie, the detection result indicates that the first target service item needing to be updated does not exist in the preset service items.

In an alternative embodiment, the human-machine dialog engine unit 82 is configured to determine a second event topic of the reference historical user dialog event. Screening a service scene simultaneously containing the second event theme and the first event theme from a pre-stored service scene set, and determining the service scene simultaneously containing the second event theme and the first event theme as a target service scene adapted to the user requirement, wherein the service scene set comprises a corresponding relation between the service scene and an event theme subset, and the event theme subset comprises a plurality of reference event themes related to the corresponding service scene.

In an alternative embodiment, the human machine dialog engine unit 82 is configured to create the set of candidate service items according to the event topic subset of the target service scenario.

In an optional embodiment, the human-machine conversation engine unit 82 is configured to perform, for each reference event topic in the event topic subset of the target service scenario, the following candidate service item determination operations, specifically operation contents: and acquiring a reference historical user dialog event which has an association relation with the currently processed reference event topic. And performing statistical analysis on the reference historical user conversation events with the incidence relation of the currently processed reference event topics, and determining the reference service items which are received by the user at high frequency in the currently processed reference event topics. And determining the candidate service items of the currently processed reference event theme according to the reference service items and/or the recommendation priority of the merchant belonging to the reference event theme. And creating the candidate service item set according to the result of executing the candidate service item determination operation on each reference event topic in the event topic subset of the target service scene.

In an alternative embodiment, the human machine conversation engine unit 82 is configured to analyze the target service item as turning off the lighting device. Determining a number of positive responses of the user to a third machine query statement from the reference historical user dialog event, wherein the third machine query statement contains third service recommendation information associated with the turn-off lighting device; and judging whether the number of positive answers exceeds a preset number. If the positive answer times are judged to be equal to or larger than the preset times, generating a target control instruction for turning off the lighting equipment; and sending the target control instruction to the terminal equipment. And if the positive answer times are judged to be less than the preset times, no processing is carried out.

Referring to fig. 9, fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure. The electronic device may be the server in the foregoing embodiment, and may be configured to implement the steps of the service processing method for human-computer interaction performed by the terminal device described in the foregoing embodiment. The electronic device may include: a processor 91, a memory 92 and a bus system 93.

Memory 92 includes, but is not limited to, RAM, ROM, EPROM or CD-ROM, and memory 92 is used to store the relevant instructions and data. The memory 92 stores the following elements, executable modules or data structures, or a subset thereof, or an expanded set thereof:

and (3) operating instructions: including various operational instructions for performing various operations.

Operating the system: including various system programs for implementing various basic services and for handling hardware-based tasks.

Only one memory is shown in fig. 9, but of course, the memory may be provided in plural numbers as needed.

As shown in fig. 9, the electronic device may further include an input/output device 94, and the input/output device 94 may be a communication module or a transceiver circuit. In the embodiment of the present application, the input/output device 94 is used to execute the process of sending and receiving data or signaling, such as the machine query statement and the target machine output statement in the first embodiment.

The processor 91 may be a controller, CPU, general purpose processor, DSP, ASIC, FPGA or other programmable logic device, transistor logic device, hardware component, or any combination thereof. Which may implement or perform the various illustrative logical blocks, modules, and circuits described in connection with the disclosure of the embodiments of the application. The processor 91 may also be a combination of computing functions, e.g., comprising one or more microprocessors in combination, a DSP and a microprocessor in combination, or the like.

In a particular application, the various components of the electronic device are coupled together by a bus system 93, wherein the bus system 93 may include a power bus, a control bus, a status signal bus, etc., in addition to a data bus. For clarity of illustration, however, the various buses are labeled as bus system 93 in fig. 9. For ease of illustration, it is drawn only schematically in fig. 9.

It should be noted that, in practical applications, the processor in the embodiment of the present application may be an integrated circuit chip having signal processing capability. In implementation, the steps of the above method embodiments may be performed by integrated logic circuits of hardware in a processor or by instructions in the form of software. The Processor may be a general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off the shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, or discrete hardware components. The various methods, steps, and logic blocks disclosed in the embodiments of the present application may be implemented or performed.

It will be appreciated that the memory in the embodiments of the subject application can be either volatile memory or nonvolatile memory, or can include both volatile and nonvolatile memory. The non-volatile memory may be a read-only memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable PROM (EEPROM), or a flash memory. Volatile memory can be Random Access Memory (RAM), which acts as external cache memory. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), synchronous Dynamic Random Access Memory (SDRAM), double data rate SDRAM, enhanced SDRAM, SLDRAM, synchronous Link DRAM (SLDRAM), and direct rambus RAM (DR RAM). It should be noted that the memories described in the embodiments of the present application are intended to comprise, without being limited to, these and any other suitable types of memories.

The embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a computer, implements the method or steps performed by the server in the above-mentioned embodiment.

The embodiment of the present application further provides a computer program product, and when being executed by a computer, the computer program product implements the method or the steps executed by the server in the above embodiments.

It should be noted that, for the sake of simplicity, any one of the above-mentioned embodiments of the service processing method based on human-computer interaction is described as a series of action combinations, but those skilled in the art should understand that the present application is not limited by the described action sequence, because some steps may be performed in other sequences or simultaneously according to the present application. Further, those skilled in the art will recognize that the embodiments described in this specification are preferred embodiments and that no acts are necessarily required to achieve the ends of this application.

While the present application has been described in connection with various embodiments, other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed application, from a review of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the word "a" or "an" does not exclude a plurality. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

Those skilled in the art will appreciate that all or part of the steps of the various methods of any of the above-described method embodiments of the human-computer interaction service processing method may be implemented by a program that instructs associated hardware to perform the steps of the method, where the program may be stored in a computer-readable memory, where the memory may include: flash Memory disks, read-Only memories (ROMs), random Access Memories (RAMs), magnetic or optical disks, and the like.

The foregoing embodiments of the present application have been described in detail, and the principles and implementations of a method, an apparatus and related devices for processing services through human-computer interaction according to the present application are described herein with specific examples, where the descriptions of the foregoing embodiments indicate methods and core ideas for assisting understanding of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, the method, the apparatus and the related device for processing a service through human-computer interaction may be changed in the specific implementation and application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.

Those skilled in the art will recognize that in one or more of the examples described above, the functions described herein may be implemented in hardware, software, firmware, or any combination thereof. When implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media may be any available media that can be accessed by a general purpose or special purpose computer.

The above-mentioned embodiments, objects, technical solutions and advantages of the present application are further described in detail, it should be understood that the above-mentioned embodiments are only examples of the present application, and are not intended to limit the scope of the present application, and any modifications, equivalent substitutions, improvements and the like made on the basis of the technical solutions of the present application should be included in the scope of the present application.

Claims

1. The service processing method of human-computer interaction is characterized by being applied to a server of a human-computer interaction system, wherein the human-computer interaction system comprises the server and terminal equipment for human-computer interaction of a user, and the server comprises a human-computer conversation engine; the method comprises the following steps:

invoking the man-machine conversation engine to execute the following operations:

detecting whether a user answer sentence of the user for a first machine inquiry sentence in the currently processed man-machine conversation event is a positive answer, wherein the first machine inquiry sentence contains first service recommendation information aiming at the requirement of the user in the currently processed man-machine conversation event;

if the answer is positive, inquiring whether reference historical user conversation events which have an association relation with the first event theme exist in prestored historical user conversation events or not according to the first event theme of the currently processed man-machine conversation events, wherein the association relation comprises a time sequence and/or a place adaptation relation, the historical user conversation events comprise historical man-machine conversation events and/or historical call records, and the historical call records refer to call records which are answered by the terminal equipment and are authorized by the user to be cached and used;

if the reference historical user dialogue event exists, determining a target service item which is adapted to the requirement of the user according to the reference historical user dialogue event and the currently processed man-machine dialogue event; generating a target machine output statement according to the target service item; and sending the target machine output statement to the terminal device;

if the reference historical user dialogue event does not exist, sending a conventional machine reply sentence aiming at the positive reply to the terminal equipment;

if the answer is negative, generating second service recommendation information aiming at the requirements of the user according to the user answer sentence; and generating a second machine inquiry statement according to the second service recommendation information, and sending the second machine inquiry statement to the terminal equipment.

2. The method of claim 1, wherein determining a target service item that fits the user's needs based on the reference historical user dialog events and the currently processed human-machine dialog events comprises:

detecting whether a first target service item needing to be updated exists in the service items preset by the user or not according to the reference historical user conversation event and the currently processed man-machine conversation event;

if the first target service item is detected to exist, marking the first target service item as a target service item which is adaptive to the requirement of the user; determining a target service scene adapting to the requirements of the user according to the reference historical user conversation event and the currently processed man-machine conversation event; judging whether a second target service item which is not subjected to conversation processing exists in a candidate service item set corresponding to the pre-stored target service scene;

if the second target service item is judged to exist, marking the second target service item as a target service item which is adaptive to the requirement of the user;

if the second target service item does not exist, no processing is carried out;

if the first target service item does not exist, determining a target service scene which is adapted to the requirements of the user according to the reference historical user conversation event and the currently processed man-machine conversation event; judging whether a second target service item which is not subjected to conversation processing exists in a candidate service item set corresponding to the pre-stored target service scene;

if the second target service item exists, marking the second target service item as a target service item which is adaptive to the requirement of the user;

and if the second target service item does not exist, not processing.

3. The method of claim 2, wherein the reference historical user session events comprise reference historical call records; the step of detecting whether a first target service item needing to be updated exists in the service items preset by the user according to the reference historical user conversation event and the currently processed man-machine conversation event comprises the following steps:

analyzing the currently processed man-machine conversation event to determine that the user is scheduled to watch the movie at the target cinema in the target time period;

if the reference historical call records are analyzed to determine that the service items scheduled by the user include going to company overtime in a target time period, detecting that a first target service item needing to be updated exists in the service items scheduled by the user, and determining that the going-to-target cinema watching movies are the first target service item needing to be updated;

and if the reference historical call record is analyzed to determine that the reference historical call record does not contain the arrangement of the target time interval, or the arrangement of the target time interval is determined to be that the target cinema is used for watching the movie, the detection result indicates that the first target service item needing to be updated does not exist in the service items preset by the user.

4. The method of claim 2, wherein said determining a target service scenario that adapts to the user's needs based on the reference historical user dialog events and the currently processed human-machine dialog events comprises:

determining a second event topic for the reference historical user dialog event;

screening a service scene simultaneously containing the second event theme and the first event theme from a pre-stored service scene set, and determining the service scene simultaneously containing the second event theme and the first event theme as a target service scene adapted to the requirement of the user, wherein the service scene set comprises a corresponding relation between the service scene and an event theme subset, and the event theme subset comprises a plurality of reference event themes involved in the corresponding service scene.

5. The method according to claim 4, wherein before the determining whether there is a second target service item that has not been processed by a dialog process in the pre-stored candidate service item set corresponding to the target service scenario, the method further comprises:

and creating the candidate service item set according to the event topic subset of the target service scene.

6. The method of claim 5, wherein creating the set of candidate service items according to the subset of event topics for the target service scenario comprises:

for each reference event topic in the event topic subset of the target service scenario, performing the following candidate service item determination operations:

acquiring a reference historical user dialogue event which has an incidence relation with a currently processed reference event theme;

performing statistical analysis on the reference historical user conversation event which has an incidence relation with the currently processed reference event theme, and determining a reference service item which is received by the user at a high frequency in the currently processed reference event theme;

determining candidate service items of the currently processed reference event topic according to the reference service items and/or the recommendation priority of the merchant belonging to the reference event topic;

creating the set of candidate service items according to a result of performing the candidate service item determination operation on each reference event topic in the subset of event topics of the target service scenario.

7. The method of claim 1, wherein after determining a target service item that fits the user's needs based on the reference historical user dialog events and the currently processed human-machine dialog events, the method further comprises:

analyzing the target service item as turning off a lighting device;

determining a number of positive responses by the user to a third machine query statement from the reference historical user dialog event, wherein the third machine query statement contains third service recommendation information associated with the turn off lighting device; judging whether the number of positive answers exceeds a preset number;

if the number of times of the positive answers is judged to be equal to or greater than the preset number of times, generating a target control instruction for turning off the lighting equipment; and sending the target control instruction to the terminal equipment;

and if the number of times of the positive answers is smaller than the preset number of times, not processing.

8. A service processing device for man-machine interaction is characterized by comprising a calling unit and a man-machine conversation engine unit:

the calling unit is used for calling the man-machine conversation engine unit to execute the following operations:

detecting whether a user answer sentence of a user aiming at a first machine inquiry sentence in a currently processed man-machine conversation event is an affirmative answer, wherein the first machine inquiry sentence contains first service recommendation information aiming at the requirement of the user in the currently processed man-machine conversation event;

if the answer is positive, inquiring whether reference historical user conversation events which have an association relation with the first event theme exist in prestored historical user conversation events or not according to the first event theme of the currently processed man-machine conversation events, wherein the association relation comprises a time sequence and/or a place adaptation relation, the historical user conversation events comprise historical man-machine conversation events and/or historical call records, and the historical call records refer to call records which are answered by the terminal equipment and are subjected to user authorization cache and used;

if the reference historical user dialogue event exists, determining a target service item which is adapted to the requirement of the user according to the reference historical user dialogue event and the currently processed man-machine dialogue event; generating a target machine output statement according to the target service project; and sending the target machine output statement to the terminal device;

if the answer is negative, second service recommendation information aiming at the requirements of the user is generated according to the user answer sentence; and generating a second machine query statement according to the second service recommendation information, and sending the second machine query statement to the terminal device.

9. A computer-readable storage medium for storing a computer program which, when executed by a processor, performs the steps of the method of any one of claims 1 to 7.

10. An electronic device, comprising a memory storing a computer program and a processor implementing the steps of the method of any of claims 1 to 7 when the processor executes the computer program.