KR20220088276A - Cashe-based competition induction method and device for multi-chatbot system - Google Patents

Abstract

Disclosed are a method and apparatus for inducing competition based on a cache for improving satisfaction of a multi-chatbot system. According to various embodiments of the present disclosure, a method of operating an intermediary device for mediating a multi-chatbot service includes: receiving a user's question from a user terminal; checking whether a question matching the user's question exists in the Q&A sets stored in the cache memory; checking whether information on the user's question is in a competition board based on a result of checking the cache memory; and transmitting a response from one of the Q&A sets or a response received from the chatbot to the user terminal based on at least one of the check result for the cache memory and the check result for the competition board may include

Description

CASHE-BASED COMPETITION INDUCTION METHOD AND DEVICE FOR MULTI-CHATBOT SYSTEM

The following disclosure discloses a method and apparatus for inducing competition between cache-based chatbots for improving satisfaction of a multi-chatbot system.

With the development of artificial intelligence (AI) and social network service (SNS), chatbot services that can acquire information as if talking through an existing messenger are being actively introduced.

A chatbot is a conversational messenger that provides an appropriate answer by analyzing the content of a question when a user inputs a question as if they were chatting through an existing messenger. Recently, leading domestic and foreign companies have introduced chatbots to customer support services to reduce the number of manpower required to respond to customers and to process responses 24 hours a day to improve the quality of customer support services. These chatbot services can be used in various fields, such as providing public information, ordering food, and providing map information.

On the other hand, chatbots are separately created according to each provided service, and each chatbot can provide only a response to a query related to the corresponding service. In order to use the chatbot more efficiently, it may be required to implement a chatbot that integrally provides responses to a plurality of services. Implementation of such a chatbot requires a large amount of storage space and network resources, and practical implementation may be difficult.

The background art described above is possessed or acquired by the inventor in the process of deriving the disclosure of the present application, and cannot necessarily be said to be a known technology disclosed to the general public prior to the present application.

Various embodiments may provide a cache-based competition induction technology for improving satisfaction of a multi-chatbot system.

However, the technical tasks are not limited to the above-described technical tasks, and other technical tasks may exist.

According to various embodiments of the present disclosure, a method of operating an intermediary device for mediating a multi-chatbot service includes: receiving a user's question from a user terminal; checking whether a question matching the user's question exists in the Q&A sets stored in the cache memory; checking whether information on the user's question is in a competition board based on a result of checking the cache memory; and transmitting a response from one of the Q&A sets or a response received from the chatbot to the user terminal based on at least one of the check result of the cache memory and the check result of the competition board can do.

The transmitting may include: when a question matching the user's question is in the cache memory, transmitting a response of a Q&A set of the matching question from among the Q&A sets to the user terminal; and transmitting a response received from one of a plurality of chatbots when a question matching the user's question is not in the cache memory.

The operation of transmitting the response received from any chatbot among the plurality of chatbots may include, when information on the user's question is in the competition board, responding to the user's question recorded on the competition board. requesting a response to the user's question from the chatbot; and when the information on the user's question is not in the competition board, requesting an arbitrary chatbot for a response to the user's question.

The chatbot that responded to the user's question recorded on the competition board has the accumulated likeness display obtained for the user's question among one or more chatbots that responded to the user's question recorded on the competition board It may be the tallest chatbot.

The operation of requesting the arbitrary chatbot may include an operation of requesting a response by dispatching the plurality of chatbots to all of the plurality of chatbots in a round-robin manner to the user's question a plurality of times.

The method includes: acquiring Q&A sets related to the user's question from any of a plurality of chatbots; and updating the cache memory by using Q&A sets related to the user's query based on a caching rule.

The certain chatbot may be the chatbot with the highest accumulated crush display obtained for the user's question.

According to various embodiments, an intermediary device for mediating a multi-chatbot service includes: a cache memory for storing Q&A sets; and receiving the user's question from the user terminal, checking whether there is a question matching the user's question in the Q&A sets stored in the cache memory, and information on the user's question based on the check result of the cache memory is in the competition board, and based on at least one of the check result for the cache memory and the check result for the competition board, the response of one of the Q&A sets or the response received from the chatbot It may include a processor for transmitting to the user terminal.

the processor, when the question matching the user's question is in the cache memory, sends a response of the Q&A set of the matching question from among the Q&A sets to the user terminal, the question matching the user's question If this cache is not in the cache memory, a response received from any chatbot among a plurality of chatbots may be transmitted.

the processor, when the information on the user's question is in the competition board, requests a chatbot that has responded to the user's question recorded on the competition board for a response to the user's question; When the information on the user's question is not in the competition board, a response to the user's question may be requested from an arbitrary chatbot.

The chatbot that responded to the user's question recorded on the competition board has the accumulated likeness display obtained for the user's question among one or more chatbots that responded to the user's question recorded on the competition board It may be the tallest chatbot.

The processor may request a response by dispatching the plurality of chatbots to all of the plurality of chatbots in a round-robin manner to the user's question a plurality of times.

The processor may obtain Q&A sets related to the user's question from any of a plurality of chatbots, and update the cache memory by using the Q&A sets related to the user's question based on a caching rule.

The certain chatbot may be the chatbot with the highest accumulated crush display obtained for the user's question.

1 illustrates an example of a multi-chatbot system according to various embodiments.
2 shows an example of a cache memory and a competition board used in the intermediary device.
3 shows an example of a flowchart for explaining the operation of the mediation device.
4 is a diagram for explaining Case 1 of Table 1. Referring to FIG.
5 is a diagram for explaining Case 2 of Table 1. Referring to FIG.
6 is a view for explaining Case 3 of Table 1. Referring to FIG.
7 is a diagram for explaining a dispatching rule.
8 is a diagram for explaining a caching rule rule.
9 is a schematic block diagram illustrating an example of an intermediary device.

Specific structural or functional descriptions of the embodiments are disclosed for purposes of illustration only, and may be changed and implemented in various forms. Accordingly, the actual implementation form is not limited to the specific embodiments disclosed, and the scope of the present specification includes changes, equivalents, or substitutes included in the technical spirit described in the embodiments.

Although terms such as first or second may be used to describe various elements, these terms should be interpreted only for the purpose of distinguishing one element from another. For example, a first component may be termed a second component, and similarly, a second component may also be termed a first component.

When a component is referred to as being “connected to” another component, it may be directly connected or connected to the other component, but it should be understood that another component may exist in between.

The singular expression includes the plural expression unless the context clearly dictates otherwise. In this specification, terms such as "comprise" or "have" are intended to designate that the described feature, number, step, operation, component, part, or combination thereof exists, and includes one or more other features or numbers, It should be understood that the possibility of the presence or addition of steps, operations, components, parts or combinations thereof is not precluded in advance.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and should not be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present specification. does not

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. In the description with reference to the accompanying drawings, the same components are assigned the same reference numerals regardless of the reference numerals, and overlapping descriptions thereof will be omitted.

1 shows an example of a multi-chatbot system according to various embodiments, and FIG. 2 shows an example of a cache memory and a competition board used in an intermediary device.

According to various embodiments, the multi-chatbot system 10 may perform a cache-based competition induction method for improving system satisfaction. The satisfaction level of the multi-chatbot system 100 may mean the cumulative value of the likeness (eg, 'Like', Like it, Like, etc.) that users give to the response (eg, answer) provided by the chatbot. In the multi-chatbot system 10, it is possible to induce an increase in the accumulation of favorable impressions in the multi-chatbot system 10 by operating a ranking of the chatbots and creating a system in which each chatbot competes to receive a lot of likes. .

According to various embodiments, in the multi-chatbot system 10, the chatbots may utilize a cache (eg, a cache memory) as a means of competing to receive a lot of likes from users. The cache memory may store a response with a high crush mark value. Once the response is stored in the cache memory, it can be used as an answer to the user's question coming from that point, so it can be an opportunity to increase the ranking by receiving more likes. From the point of view of chatbot developers, in order to raise the favorable ranking of their chatbot, they will compete with each other to provide better answers so that their answers are put into cache memory. This may increase the cumulative value of the display of liking of the multi-chatbot system 10 as a whole to induce improvement in satisfaction.

According to various embodiments, the multi-chatbot system 10 may include a user terminal 100 , an intermediary device 200 (eg, a meta-chatbot), and a plurality of chatbots. The multi-chatbot system 10 includes a terminal (eg, a first layer) of the user terminal 100, an edge (or cache) (eg, a second layer) of the intermediary device 200, and a chatbot of a plurality of chatbots (eg: The third layer) can be divided into three layers.

In the first layer, the user terminal 100 may include a chatbot interface (eg, mukacha) that is a client program. The chatbot interface serves as a middleware between the user terminal 100 and other components (eg, the intermediary device 200) of the multi-chatbot system 10, and communicates with the user (eg, between the user terminal 100 and the chatbot). incoming conversations). The user terminal 100 may access the intermediary device 200 through a chatbot interface installed in the user terminal 100 . The user may transmit a question (eg, query, query) through the chatbot interface of the user terminal 100 .

The user terminal 100 may be implemented as a personal computer (PC), a data server, or a portable device. Portable electronic devices include laptop computers, mobile phones, smart phones, tablet PCs, mobile internet devices (MIDs), personal digital assistants (PDAs), and enterprise digital assistants (EDAs). ), digital still camera, digital video camera, PMP (portable multimedia player), PND (personal navigation device or portable navigation device), handheld game console, e-book (e-book), may be implemented as a smart device. In this case, the smart device may be implemented as a smart watch or a smart band.

When the multi-chatbot system 10 (eg, the intermediary device 200, the chatbot) returns a response to the query, the user can transmit a liking for the response to the intermediary device 220 through the user terminal 100. have. This may be a task for improving the satisfaction of the multi-chatbot system 10 (eg, the intermediary device 200, the chatbot).

In the second layer, the intermediary device 200 may perform an intermediary role between the user terminal 100 and a plurality of chatbots. The intermediary device 200 may select one or more chatbots (eg, a chatbot most appropriate to a question) from among a plurality of chatbots according to the content of the user's question of the user terminal 100 , and may provide a question to the selected chatbot. The intermediary device 200 may receive a response to the question from the selected chatbot and provide the received response to the user.

The mediation device 200 may include a cache memory and a competition board. The intermediary device 200 may provide a response corresponding to the user's question by using the cache memory and the competition board. The cache memory may store Q&A sets related to the user's question (eg, Q&A sets frequently asked by a chatbot, Q&A sets with a high probability of being asked a question), and information about the user's question in the competition board can be recorded. As shown in FIG. 2 , information on a user question (eg, Q1 to Qm in FIG. 2 ) is the accumulated value of display of liking for each chatbot (eg, C1 to Cm in FIG. 2 ) for the user question (eg, in FIG. 2 ) L1 to L3) and information on the response frequency of each chatbot to the user's question (eg, Fq1 to Fq3 in FIG. 2 ).

The intermediary device 200 may check whether the user's question received from the user terminal 100 is in the cache memory. The intermediary device 200 may be caching (eg, prefetching) Q&A sets (eg, Q&A sets frequently asked by the chatbot) related to the user's question in advance. When receiving the user's question, the intermediary device 200 responds to the user's question based on the related Q&A sets cached in the cache memory before forwarding the user's question to one or more of the plurality of chatbots. It may be provided to the user terminal 100 .

Also, when the intermediary device 200 is unable to provide an answer to the user's question from the relevant Q&A sets cached in the cache memory (eg, the user's query is not in the cache memory), the intermediary 200 may display the user's question in the competition board. You can check the number of favorable impressions (eg, cumulative value) that multiple chatbots are acquiring. The intermediary device 200 may request a response to a question from the chatbot with the highest accumulated value among the plurality of chatbots 310 , and provide the response received from the chatbot to the user terminal 100 .

When considering the roles of the cache memory and the competition board, there may be three cases as shown in [Table 1] between the information of the cache memory and the competition board with respect to the user's question received by the intermediary device 200 . The description of the three cases in Table 1 will be described in detail with reference to FIGS. 4 to 6 .

[Table 1]

In the third layer, the chatbot can analyze the question data for the user's question (eg, query, query) to provide an appropriate response to the question. The chatbot stores a question and a Q&A set, which is a pair of responses corresponding to a question, in the database, and whenever a question is received from a user (eg, the user terminal 100), the chatbot searches the database for a Q&A set corresponding to the question, and from the searched Q&A set To provide a response to the user terminal (100). Each chatbot may provide responses for all topics, but may exist by topic to provide responses only to queries on the corresponding topic. A plurality of chatbots may exist in one server, but is not limited thereto, and may exist interspersed in a plurality of servers.

When the chatbot provides an answer to the current question, it can provide both the answer to the current question and Q&A sets related to the current question. The Q&A sets provided in this way are stored together when the current question is stored in the cache memory of the intermediary device 200, so that the hit rate of the cache memory can be increased later.

3 shows an example of a flowchart for explaining the operation of the mediation device.

The intermediary device 200 may receive the user's question Q from the user terminal 100 .

The intermediary device 200 may check whether the user's question Q is stored in the cache memory, and check whether information about the user's question Q is recorded on the competition board.

When the user's question is stored in the cache memory, the intermediary device 200 may directly provide a response to the user's question Q to the user terminal 100 based on the related Q&A sets cached in the cache memory. If the user's question Q is not stored in the cache memory, but information about the user's question Q is recorded in the competition board, the intermediary device 200 sends the user's question Q to the chatbot with the highest accumulated favoritism. A response to the question Q may be requested, and the response received from the corresponding chatbot may be provided to the user terminal 100 . If the user's question Q is not stored in the cache memory and information on the user's question Q is not recorded in the competition board, the intermediary device 200 randomly sends a response to the question Q among the plurality of chatbots. may be requested, and the response received from the corresponding chatbot may be provided to the user terminal 100 .

The intermediary device 200 may request the user to input a good feeling for this while transmitting a response to the user terminal 100 , and may receive information about the good feeling inputted by the user.

The intermediary device 200 may check the display of liking input by the user, record the user question Q on the competition board, and record or update information about the user question Q. The information on the user question Q may include an indication of a liking received from the user for the response and a response frequency (eg, response frequency information) for the user question Q.

4 is a diagram for explaining Case 1 of Table 1. Referring to FIG.

Case 1 may be that the user's question Q currently received by the intermediary device 200 is not stored (eg, registered) in the cache memory, and information about the user's question Q is not recorded on the competition board.

The user terminal 100 may transmit the user's question Q to the intermediary device 200 through a chatbot interface (eg, a user interface) ( 410 ).

The intermediary device 200 may check whether the user's question Q is stored in the cache memory ( 420 ), and check whether information about the user's question Q is recorded in the competition board ( 430 ). Since the user's question Q is not stored in the cache memory and the information about the user's question Q is not recorded on the competition board, the intermediary device 200 randomly selects one chatbot from among the plurality of chatbots and asks the user's question. Q may be transmitted (440).

The chatbot receiving the user's question Q may return a response A to the corresponding question to the intermediary device 200 ( 450 ).

The intermediary device 200 may transmit the received response A to the user terminal 100 ( 460 ).

When the intermediary device 200 randomly selects one chatbot, the likeness display value for one chatbot that initially responded becomes a valid value, so that an opportunity may be continuously provided only to the corresponding chatbot. The intermediary device 200 can improve this irrationality by giving a fair chance by dispatching all chatbots in a round-robin manner for the first N (eg, N is a natural number greater than or equal to 1) questions for the first N times. This operation may be referred to as a boot-up operation. In bootup behavior, caching may not occur because there is insufficient evidence to determine which chatbot's response is of good quality.

When the boot-up operation is finished, the intermediary device 200 may perform a cache-on operation in which the cache function starts to be applied. When entering the cash-on operation, it is possible to request the chatbot that has obtained the highest accumulation of favorable impressions up to that point to transmit n sets of Q&A (eg, n is a natural number greater than or equal to 1) related to the question. The intermediary device 200 may determine whether a total of n+1 Q&A sets, including n, sent by the corresponding chatbot, will remain in the cache memory according to the corresponding request. The intermediary device 200 performs this determination based on a caching rule, and when K is the size of the cache or the maximum number that the cache can cache for a query, it meets the caching rule. Only K Q&A sets (eg, K is a natural number greater than or equal to 1) can be stored in the cache memory in order of highest degree of precision.

When the user receives the response A through the user terminal 100, the user may input On/Off or yes/no for displaying a good feeling for the response A (470). The intermediary device 200 may record the question Q on the competition board based on the crush information input by the user, and record or update the crush indication and frequency information returned for the response A ( 480 ).

In this way, the user gets the answer A to the question Q, the intermediary 200 decides whether to cache a set of Q&A that is highly relevant to it, updates the cache memory, and the case 1 ends by updating the record to the competition board. can be

5 is a diagram for explaining Case 2 of Table 1. Referring to FIG.

Case 2 may be that the user's question Q currently received by the mediation center 200 is stored (eg, registered) in the cache memory, and information about the user's question Q is recorded on the competition board.

The user terminal 100 may transmit the user's question Q to the intermediary device 200 through a chatbot interface (eg, a user interface) ( 510 ).

The intermediary device 200 checks whether the user's question Q is stored in the cache memory, and the user's question Q is stored in the cache memory, so that the answer A to the user's question Q may be immediately returned ( 520 ).

When the user receives the response A through the user terminal 100, the user may input On/Off or yes/no for the display of liking for it (eg, 'Like') ( 530 ). The intermediary device 200 may update information on the display of liking for the corresponding question (eg, this question) on the competition board based on the information input by the user ( 540 ).

In this way, the user gets the answer A to the question Q, and since the question is already stored in the cache memory of the intermediary device 200, only information about the crush mark in the competition board (eg, crush accumulation value, frequency information) It can be updated and case 2 can be closed.

If the intermediary device 200 returns a response only with the contents in the cache memory (eg, stored Q&A sets), the opportunity may not return to other chatbots. The intermediary device 200 does not use the Q&A sets stored in the cache memory, but allows the opportunity to return to other chatbots with a probability as small as a certain amount (eg, e), which opens an opportunity for the ranking to be reversed. can be performed. Here, e is an abbreviation of epsilon, and this policy may also be called an epsilon-greedy policy.

An operation in which the intermediary device 200 grants an opportunity to another chatbot may be referred to as a call-up operation. However, the call-up response of the chatbot may not be stored in the cache memory of the intermediary device 200 . In the cache memory, only the content of the chatbot with the highest accumulated likes can always be recorded. The call-up action can be seen as an action to give other chatbots that are not currently cached a chance to receive a like and to enter the cache memory.

When the user's question Q is received from the user terminal 100, the intermediary device 200 immediately returns a response A to the user's question Q, and then immediately reviews the like accumulation value of the competition board to obtain the highest You can check to see if a reversal has occurred in the number, and if so, you can update the cache memory.

For the update, the intermediary device 200 requests the chatbot with the highest inverted like accumulation value to transmit not only the response A of the currently input user question Q, but also n sets of Q&A related to the question. can That is, the mediation device 200 may request to transmit a total of n+1 Q&A sets including the user question Q. The intermediary device 200 may determine whether to remain in the cache memory for n+1 Q&A sets sent by the chatbot according to the corresponding request. The intermediary device 200 performs this determination based on a caching rule, and when K is the size of the cache or the maximum number that the cache can cache for a query, it meets the caching rule. Only K Q&A sets in the order of severity can be stored in the cache memory.

6 is a view for explaining Case 3 of Table 1. Referring to FIG.

Case 3 may be that the user's question Q received by the intermediary device 200 is not stored (eg, registered) in the cache memory, and information about the user's question Q is recorded on the competition board.

Even if the intermediary device 200 enters the cache-on operation, there may be no guarantee that the answer to the corresponding question and the set of related Q&A will be stored in the cache memory. What is stored in the cache memory may be determined by a caching rule. Case 3 may correspond to a response method for a case in which the entered question is not in the cache memory even though the cache-on operation has been entered.

As described in Case 2, there may be a case where the currently cached content is not the response of the chatbot with the highest accumulated crush display. That is, a case in which a reversal occurs in the cumulative value ranking may occur. In this case, since it is processed as if there is no content in the cache memory, it can be regarded as case 3.

The user terminal 100 may transmit the user's question Q to the intermediary device 200 through a chatbot interface (eg, a user interface) ( 610 ).

The intermediary device 200 does not store the user's question Q in the cache memory, but may check information about the user's question Q on the competition board ( 620 ). The intermediary device 200 may request a response A from the chatbot having the largest accumulated likes ( 630 ), and receive the response A ( 640 ) and transmit it to the user terminal 100 ( 650 ).

When the user receives the response A through the user terminal 100, the user may input On/Off or yes/no for displaying a good feeling for the response A (660). The intermediary device 200 may update the information on the corresponding question Q on the competition board based on the good impression information input by the user ( 670 ).

After completing the above-described process, the intermediary device 200 may request the chatbot that has just sent a response to transmit n relevant Q&A sets. The intermediary device 200 may determine whether to remain in the cache memory for a total of n+1 Q&A sets including n Q&A sets sent by the chatbot according to the corresponding request. The intermediary device 200 performs this determination based on a caching rule, and when the size of the cache or the maximum number that the cache can cache for a query is K, the caching rule that meets the Only K Q&A sets in the order of severity can be stored in the cache memory.

In this way, the user gets an answer to the question Q, the intermediary device 200 decides whether to cache a set of highly relevant Q&A, and updates the cache memory, and by updating the record on the competition board, case 3 can be closed. .

7 is a diagram for explaining a dispatching rule.

In the competition board, as shown in Fig. 3, the number of times (eg, response frequency information) that chatbots that have had a chance to respond to the user's questions that have been input so far got a chance to respond to each response content (eg, response frequency information) and the favorable feelings obtained from users A cumulative display value may be recorded.

Here, the ratio of the cumulative value of the likeness display to the number of responses (eg, response frequency information) is referred to as LRF (Like Frequency Ratio) and may be defined as in Equation 1 below.

[Equation 1]

Here, Accumulated_Like may mean an accumulated value of display of liking, and Answer_Frequency may mean the number of responses (eg, response frequency information).

It is assumed that the user's question Q is input to the intermediary device 200 . If the corresponding question is entered for the first time, the LFR value is not valid because there is no information on the competition board. : Case 1). In this case, only the LFR of the chatbot that is randomly selected and responded initially becomes a valid value, so that only the chatbot can continue to be provided with an opportunity. Accordingly, the intermediary device 200 can improve this irrationality by performing a boot-up operation in which the first N questions are dispatched to all chatbots in a round-robin manner to give a fair opportunity. Cache may not occur in boot-up behavior because there is insufficient evidence to determine which chatbot response is of good quality.

After the bootup operation is finished, only the chatbot with the highest LFR value until then is given many opportunities. Accordingly, the intermediary device 200 may perform a call-up operation that opens an opportunity for the ranking to be reversed by allowing the other chatbot to return an opportunity with a small talk rate as much as a certain amount of probability (eg, e).

On the other hand, the probability value e may be fixed as a constant, but in some cases, it is set as a variable and if there is little deviation in the liking display value between chatbots, the e value is increased to promote competition. If there is a large deviation between the accumulated number of likes and the like, a variable method of reducing the e value may be adopted, considering that the case of reversal is small.

After the boot-up operation, the intermediary device 200 may perform a cache-on operation in which the cache function starts to be applied. The cache-on operation may be to cache the responses and associated Q&A sets with the highest crush marks accumulation. However, it is not always stored in the cache memory, and whether or not to be cached in the cache memory may follow a caching rule.

If the response to the entered question is not stored in the cache memory, the intermediary device 200 may give the chatbot having the highest LFR for the question an opportunity to respond (eg, case 3). Upon completion of the response, the intermediary device 200 may request the chatbot that has just sent a response to send relevant n Q&A sets. The intermediary device 200 may determine whether to remain in the cache for a total of n+1 Q&A sets including n Q&A sets sent by the corresponding chatbot according to this request. The intermediary device 200 performs this determination based on a caching rule, and k (eg, the size of the cache allowed for the corresponding query) Q&A in the order of the highest degree of conformance to the caching rule. Only sets can be stored in cache memory.

When the response to the input question is in the cache memory, the intermediary device 200 may simply return the response (eg, case 2). Thereafter, the intermediary device 200 may immediately review the accumulated likes of the competition board to determine whether a reversal has occurred at the highest number, and if the reversal has occurred, the cache memory may be updated. To this end, the intermediary device 200 provides n Q&A sets that are related to the question as well as the response of the question currently input to the chatbot whose accumulated like accumulation is the highest, that is, a total of n+1 Q&A sets. You can request to send The intermediary device 200 may determine whether to remain in the cache memory for n+1 Q&A sets sent by the chatbot according to this request. The intermediary device 200 performs this determination based on a caching rule, and may store only k Q&A sets in the cache memory in the order of the degree of conformance to the caching rule (Caching Rule).

8 is a diagram for explaining a caching rule rule.

From the standpoint of the intermediary device 200, it may be ideal to store the response of the chatbot with the highest accumulated user's liking for the answers to the questions entered so far in the cache memory in the cache memory. . This can be problematic in two ways. First, the appropriate number of Q&A sets related to the question must also be cached. Second, if the size of the cache memory is limited, it is impossible to cache all such related Q&A sets. In consideration of this problem, a method for implementing selective caching is required.

If the Q&A set for the currently entered question has not been cached in the cache memory (eg, case 3), the intermediary device 200 sends n pairs of Q&A highly relevant to the question to the chatbot that is currently providing the answer. You can request it to be sent. Then, the intermediary device 200 may determine whether the thus obtained n+1 Q&A sets can be stored in the cache memory.

When the Q&A set for the entered question is in the cache memory (eg, case 2), the intermediary device 200 may simply return the answer. Thereafter, the intermediary device 200 may immediately review the accumulated value of the display of liking of the competition board to determine whether a reversal has occurred at the highest value, and if the reversal has occurred, the cache memory may be updated. To this end, the intermediary device 200 sends n Q&A sets related to the question as well as the response of the currently input question to the chatbot whose accumulated like value is the highest, that is, a total of n+1 Q&A sets. You can ask for it to be sent. The intermediary device 200 may determine whether to remain in the cache memory for n+1 Q&A sets sent by the chatbot according to this request.

The mediation device 200 may determine whether to cache the Q&A set of the caching target obtained through the above process. For this, the most representative caching rule used by the intermediary device 200 may be based on the frequency of queries. The frequency of questions may be defined as the sum of the number of responses (eg, response frequency information) of each chatbot to the corresponding question when referring to FIG. 2 . This is called QOF (Question Occurrence Frequency), and the QOF of a question may be equal to the sum of the number of times the chatbot responded in the corresponding question row on the competition board.

The mediation device 200 may calculate the QOF of each question from the n+1 questions just obtained and the questions already in the cache, and sort them based on this. If the size of the cache is K, K Q&A sets may be left in the cache memory in the order of the highest frequency.

9 is a schematic block diagram illustrating an example of an intermediary device.

The intermediary device 200 may include a memory 210 , a processor 230 , and a cache memory. The cache memory may be implemented in the processor 230 .

The memory 210 may store instructions (eg, a program) executable by the processor 230 . For example, the instructions may include instructions for executing an operation of the processor 230 and/or an operation of each component of the processor 230 .

The cache memory may store the Q&A sets described with reference to FIGS. 1 to 8 . Q&A sets may be prefetched and stored in cache memory.

The processor 230 may process data stored in the memory 210 . The processor 230 may execute computer readable code (eg, software) stored in the memory 210 and instructions induced by the processor 230 .

The processor 230 may be a hardware-implemented data processing device having a circuit having a physical structure for executing desired operations. For example, desired operations may include code or instructions included in a program.

For example, a data processing device implemented as hardware includes a microprocessor, a central processing unit, a processor core, a multi-core processor, and a multiprocessor. , an Application-Specific Integrated Circuit (ASIC), and a Field Programmable Gate Array (FPGA).

The operation performed by the processor 230 may be substantially the same as the operation of the intermediary apparatus 200 described with reference to FIGS. 1 to 8 . Accordingly, detailed description will be omitted.

The embodiments described above may be implemented by a hardware component, a software component, and/or a combination of a hardware component and a software component. For example, the apparatus, methods, and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate (FPGA) array), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions, may be implemented using a general purpose computer or special purpose computer. The processing device may execute an operating system (OS) and a software application running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For convenience of understanding, although one processing device is sometimes described as being used, one of ordinary skill in the art will recognize that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that can include For example, the processing device may include a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as parallel processors.

Software may comprise a computer program, code, instructions, or a combination of one or more thereof, which configures a processing device to operate as desired or is independently or collectively processed You can command the device. The software and/or data may be any kind of machine, component, physical device, virtual equipment, computer storage medium or apparatus, to be interpreted by or to provide instructions or data to the processing device. , or may be permanently or temporarily embody in a transmitted signal wave. The software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored in a computer-readable recording medium.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer readable medium may store program instructions, data files, data structures, etc. alone or in combination, and the program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and available to those skilled in the art of computer software. have. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like.

The hardware devices described above may be configured to operate as one or a plurality of software modules to perform the operations of the embodiments, and vice versa.

As described above, although the embodiments have been described with reference to the limited drawings, those of ordinary skill in the art may apply various technical modifications and variations based thereon. For example, the described techniques are performed in an order different from the described method, and/or the described components of the system, structure, apparatus, circuit, etc. are combined or combined in a different form than the described method, or other components Or substituted or substituted by equivalents may achieve an appropriate result.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (15)

In the operating method of an intermediary device that mediates a multi-chatbot service,
receiving a user's question from the user terminal;
checking whether a question matching the user's question exists in the Q&A sets stored in the cache memory;
checking whether information on the user's question is in a competition board based on a result of checking the cache memory; and
Transmitting a response of one of the Q&A sets or a response received from the chatbot to the user terminal based on at least one of the verification result of the cache memory and the verification result of the competition board
A method comprising
According to claim 1,
The transmitting operation is
transmitting, to the user terminal, a response of a Q&A set of the matching question from among the Q&A sets when a question matching the user's question is in the cache memory; and
Transmitting a response received from a chatbot among a plurality of chatbots when a question matching the user's question is not in the cache memory
A method comprising
3. The method of claim 2,
The operation of transmitting a response received from one of the plurality of chatbots includes:
requesting a response to the user's question from a chatbot that responded to the user's question recorded on the competition board when the information about the user's question is in the competition board; and
When the information on the user's question is not in the competition board, requesting a response to the user's question to an arbitrary chatbot
A method comprising
4. The method of claim 3,
The chatbot that responded to the user's question recorded on the competition board,
The method of claim 1, wherein the chatbot has the highest accumulated crush display obtained for the user's question among one or more chatbots that responded to the user's question recorded on the competition board.
4. The method of claim 3,
The operation of requesting the arbitrary chatbot is,
An operation of requesting a response by dispatching to all of the plurality of chatbots in a round-robin manner to the user's question a plurality of times
A method comprising
According to claim 1,
acquiring Q&A sets related to the user's question from one of a plurality of chatbots; and
updating the cache memory by using Q&A sets related to the user's query based on caching rules
A method further comprising:
7. The method of claim 6,
Any of the above chatbots,
The method of claim 1, wherein the chatbot has the highest accumulated liking for the user's question.
A computer program stored in a computer-readable recording medium in combination with hardware to execute the method of any one of claims 1 to 7.
In the intermediary device for mediating multi-chatbot service,
a cache memory for storing Q&A sets; and
Receives a user's question from the user terminal, checks whether there is a question matching the user's question in the Q&A sets stored in the cache memory, and provides information on the user's question based on the result of checking the cache memory check whether it is in the competition board, and display the response of one of the Q&A sets or the response received from the chatbot based on at least one of the verification result of the cache memory and the verification result of the competition board A processor that transmits to a user terminal
Including, intermediary device.
10. The method of claim 9,
The processor is
when a question matching the user's question is in the cache memory, sending a response of a Q&A set of the matching question among the Q&A sets to the user terminal;
and, when a question matching the user's question is not in the cache memory, transmitting a response received from any chatbot among a plurality of chatbots.
11. The method of claim 10,
The processor is
when the information about the user's question is in the competition board, requesting a response to the user's question from the chatbot that responded to the user's question recorded on the competition board;
When the information on the user's question is not in the competition board, the intermediary device for requesting a response to the user's question to an arbitrary chatbot.
12. The method of claim 11,
The chatbot that responded to the user's question recorded on the competition board,
The method of claim 1, wherein the chatbot has the highest accumulated crush display obtained for the user's question among one or more chatbots that responded to the user's question recorded on the competition board.
12. The method of claim 11,
The processor is
Dispatch to all of the plurality of chatbots in a round robin manner to request a response to the user's query multiple times.
10. The method of claim 9,
The processor is
obtaining Q&A sets related to the user's question from any one of a plurality of chatbots;
and updating the cache memory by using Q&A sets related to the user's query based on caching rules.
15. The method of claim 14,
Any of the above chatbots,
An intermediary device, wherein the chatbot has the highest accumulated liking for the user's question.

Images