KR102362058B1 - Chat-bot system utilizing multiple artificial intelligence model trained for each behaviors of companion animals - Google Patents

Abstract

Disclosed is a companion animal training guidance method of a chatbot system. The method comprises: a step of identifying an artificial intelligence model corresponding to companion animal behavior selected according to user input among a plurality of artificial intelligence models trained for providing guidance for training the companion animal; and a step of providing guidance for training the companion animal performing selected behavior based on the identified artificial intelligence model.

Description

A chatbot system using multiple AI models trained for each behavior of companion animals { CHAT-BOT SYSTEM UTILIZING MULTIPLE ARTIFICIAL INTELLIGENCE MODEL TRAINED FOR EACH BEHAVIORS OF COMPANION ANIMALS }

The present disclosure relates to a chatbot system, and more particularly, to a chatbot system that guides a discipline process based on discipline guidelines defined for each behavior.

Recently, as the number of guardians with companion animals increases, the demand for discipline experts to correct the wrong behavior or unstable psychological state of companion animals is rapidly increasing.

However, since the types and behaviors of companion animals are very diverse, direct visits or one-on-one coaching by discipline experts cannot meet the demand.

In particular, although there is a history of developing various systems related to the health management of companion animals and a matching platform with experts, it is difficult to find an application-based service that specifically guides the discipline process for companion animal behavior based on AI. situation.

Registered Patent Publication No. 10-20486200000 (dog training-based communication service system)

The present disclosure provides a chatbot system that provides a specialized discipline guide for various behaviors by using an artificial intelligence model trained independently for each behavior of a companion animal.

The present disclosure provides a chatbot system that guides a discipline process that is individually optimized for a user and a companion animal based on a conversation with the user.

Objects of the present disclosure are not limited to the above-mentioned purposes, and other objects and advantages of the present disclosure that are not mentioned may be understood by the following description, and will be more clearly understood by the embodiments of the present disclosure. Moreover, it will be readily apparent that the objects and advantages of the present disclosure may be realized by the means and combinations thereof indicated in the claims.

The companion animal discipline guide method of the chatbot system according to an embodiment of the present disclosure provides an artificial intelligence corresponding to the behavior of a companion animal selected according to a user input from among a plurality of artificial intelligence models trained to provide a guide for the discipline of companion animals. Identifying an intelligent model, and based on the identified artificial intelligence model, comprising the step of providing a guide for the discipline of the companion animal performing the selected action.

The companion animal discipline guide method of the chatbot system may further include acquiring discipline guidelines for each behavior of companion animals, and training the plurality of artificial intelligence models to provide a guide based on the discipline guidelines. .

The step of identifying the artificial intelligence model corresponding to the selected behavior may include, when there are a plurality of artificial intelligence models registered for the selected behavior, the plurality of types according to at least one of a species, age, and gender of a companion animal selected according to a user input. One AI model can be selected from among the dog AI models.

On the other hand, the step of providing the guide may provide a discipline guide of the first stage, and provide a discipline guide of the second stage according to the user's progress report for the provided discipline guide of the first stage.

At this time, in the step of providing the guide, when a user input requesting a supplementary explanation for the discipline guide of the first step is received in a state in which the discipline guide of the first step is provided, at least one demonstration image is provided. may be

On the other hand, the companion animal discipline guide method of the chatbot system includes the steps of: acquiring a plurality of discipline guidelines for a specific behavior of a companion animal; training each of the artificial intelligence models based on each of the obtained plurality of discipline guidelines; Based on the user's evaluation of the guide provided through each of the artificial intelligence models, it may further include calculating a score for each of the plurality of discipline guidelines.

The chatbot system that provides a guide related to the discipline of companion animals according to the present disclosure generates a guide through an artificial intelligence model identified according to the behavior of the companion animal. It has the advantage of being able to provide

In addition, the chatbot system according to the present disclosure has an effect of providing a customized guide in real time through a conversation with a user.

1 is a block diagram illustrating a functional configuration of a chatbot system according to an embodiment of the present disclosure;
2 is a flowchart for explaining a discipline guide method of a chatbot system according to an embodiment of the present disclosure;
3A is a diagram for explaining an operation of receiving a user input for selecting a behavior of a companion animal by a chatbot system according to an embodiment of the present disclosure;
3B is a view for explaining an operation of providing a discipline guide according to a selected action by the chatbot system according to an embodiment of the present disclosure; and
4 is a block diagram illustrating a hardware configuration of a chatbot system according to various embodiments of the present disclosure.

Prior to describing the present disclosure in detail, a description will be given of the description of the present specification and drawings.

First, terms used in the present specification and claims have been selected in consideration of functions in various embodiments of the present disclosure. However, these terms may vary depending on the intention or legal or technical interpretation of a person skilled in the art, and the emergence of new technology. Also, some terms are arbitrarily selected by the applicant. These terms may be interpreted in the meanings defined herein, and in the absence of specific definitions, they may be interpreted based on the general content of the present specification and common technical common sense in the art.

Also, the same reference numerals or reference numerals in each drawing attached to this specification indicate parts or components that perform substantially the same functions. For convenience of description and understanding, the same reference numbers or reference numerals are used in different embodiments. That is, even though all components having the same reference number are illustrated in a plurality of drawings, the plurality of drawings do not mean one embodiment.

In addition, in this specification and claims, terms including an ordinal number such as “first” and “second” may be used to distinguish between elements. This ordinal number is used to distinguish the same or similar elements from each other, and the meaning of the term should not be construed as limited due to the use of the ordinal number. As an example, the use order or arrangement order of components combined with such an ordinal number should not be limited by the number. If necessary, each ordinal number may be used interchangeably.

In this specification, the singular expression includes the plural expression unless the context clearly dictates otherwise. In the present application, terms such as "comprises" or "consisting of" are intended to designate that the features, numbers, steps, operations, components, parts, or combinations thereof described in the specification exist, and are intended to indicate that one or more other It should be understood that this does not preclude the possibility of addition or presence of features or numbers, steps, operations, components, parts, or combinations thereof.

In an embodiment of the present disclosure, terms such as “module”, “unit”, “part”, etc. are terms for designating a component that performs at least one function or operation, and such component is hardware or software. It may be implemented or implemented as a combination of hardware and software. In addition, a plurality of "modules", "units", "parts", etc. are integrated into at least one module or chip, except when each needs to be implemented in individual specific hardware, and thus at least one processor. can be implemented as

In addition, in an embodiment of the present disclosure, when it is said that a certain part is connected to another part, this includes not only a direct connection but also an indirect connection through another medium. In addition, the meaning that a certain part includes a certain component means that other components may be further included, rather than excluding other components, unless otherwise stated.

1 is a block diagram illustrating a functional configuration of a chatbot system according to an embodiment of the present disclosure.

Referring to FIG. 1 , the chatbot system 100 may include at least one user terminal 10 and a server 20 .

A chat-bot is an abbreviation of a chat robot, and refers to a service that provides various information or functions by performing a conversation with a user. In this case, the conversation with the user may be performed based on voice or visual text.

The chatbot system 100 corresponds to a system for providing the aforementioned chatbot service.

The chatbot system 100 may perform a conversation with the user through the user terminal 10 using at least one web page or application.

The user terminal 10 may correspond to various terminal devices such as a smart phone, a tablet PC, a smart speaker, a TV, a notebook PC, and a desktop PC.

Referring to FIG. 1 , the user terminal 10 may include a chatbot interface module 110 for communicating with the user.

The chatbot interface module 110 may communicate with the user through a microphone, a display, a speaker, etc. of the user terminal 10 .

The server 20 may include at least one electronic device for understanding the user's conversation content and generating the chatbot's conversation content.

The server 20 is connected to the user terminal 10 through at least one web page or application, and various conversation contents to be provided through the chatbot interface module 110 of the user terminal 10 (eg, a guide related to discipline) ) can be created.

Referring to FIG. 1 , the server 20 may include an artificial intelligence model management module 120 , a guide generation module 130 , and the like.

The artificial intelligence model management module 120 is a module for managing one or more artificial intelligence models trained to provide a guide for the discipline of companion animals.

Referring to FIG. 1 , the artificial intelligence model management module 120 may manage a plurality of artificial intelligence models 121 , 122 , 123 trained through different discipline guidelines.

Discipline guidelines refer to information about the training process for correcting specific behaviors of companion animals. Discipline guidelines may be prescribed by a companion animal discipline expert or scholar, and may include a step-by-step training process for correcting specific behavior.

Examples of discipline guidelines according to the behavior of companion animals are shown in Table 1 below.

The following discipline guidelines are limited to dogs among companion animals. The training guidelines in Table 1 below are exemplary, and the training guidelines used by the chatbot system according to the present disclosure are not limited to the training guidelines in Table 1 below.

action Discipline Guidelines barking at certain sounds - Identify the cause
1. You may have experienced or traumatized by a sound startled or aroused by an outsider visit after a specific sound.
2. You may be more sensitive due to not being able to walk for a long time or becoming more vigilant at home.

- solution
Through “sound desensitization education”, education is conducted in a way that changes bad memories of sound into good memories through compensation”
@ Prior learning: one or more of mat, house, sit down, and wait
1. If you are at home, your dog may be alert or stressed, so it is recommended to proceed after a light walk or play.
2. Prioritize the 'wait' education using a mat or house where your dog can normally rest comfortably.
3. During 'Wait', listen to the dog's usual reaction, starting with a small sound. Instruct your dog to enter the mat or house if it responds to the sound.
4. If you do well on the mat or house, reward them with compliments and snacks.
5. If successful, increase the volume little by little and repeat.
6. If the training does not proceed well at a certain level of sound, lower the sound level by one or two and proceed again.
Tip) The location of the mat or house should be located close to where the sound is coming from at first, and as the success rate of your dog increases, move it to a more comfortable resting place. Aggression when visiting outsiders - Identify the cause
1. Behavior that shows vigilant barking or aggression by accepting a visit from an outsider as an invasion of the territory.
2. Behavior of barking or aggression due to anxiety caused by bad memories after visiting an outsider.
3. Barking due to increased excitement level due to various stimuli (door opening sound, greeting sound, etc.) occurring when an outsider visits
4. Behavior due to any trauma when reacting to a specific person (man, child, etc.).
5. It may be that people with bad memories do not approach or disappear due to their barking and aggression.
6. Socialization period Not meeting many people or having bad memories of people.
CAUTION: Please be especially careful with the naive behavior of young children towards puppies as they may be perceived as intimidating or rude to the puppies.

- Solutions (stimulus desensitization training)
@ Build awareness that outsiders don't harm you and good things happen when outsiders approach you.
@ Prior Learning: Matt, House
@ If you are reacting to a specific appearance or gender, please proceed with the training with similar conditions.
1. Take a short walk before the start of training to relieve tension.
2. If you meet outsiders outside the house and come into the house together, you may be less alert (even better if you go for a walk together).
3. Have your dog calm down by using a treat, such as a mat or house, by the guardian (please stay still in one place so that outsiders do not irritate the dog during training).
4. If your dog still does not calm down, simply take a walk around the house using a leash.
5. Reward your dog when he is calm. Use praise and skinship together so that the child can remain calm even after the reward.
6. If your dog has calmed down, let outsiders move around the room slowly, and if they respond, give them a mat or house again.
7. Repeat by gradually increasing the moving space of the outsider according to the degree of reaction of the dog.
8. If you eat snacks provided by outsiders, it is recommended that outsiders give the child directly.
tip)
1. Steps 1 and 2 lower the dog’s excitement level and help him to focus more on education. If your dog listens to the guardian’s instructions to some extent, you can proceed from step 3.
2. The reward can be anything your dog likes. (Snacks, compliments, skinship, toys, etc.) Increased level of excitement for caregivers - Identify the cause
@ Caused by increased excitability due to high dependence on caregivers.
@ When the behavior is reinforced by a memory of being excessively affectionate when the caregiver enters the house. In that case, the dog can wait that time more painfully, leading to separation anxiety.

- solution
@ Excessive affection can adversely affect your dog. In this case, on the contrary, an attitude of indifference is required.
1. After going out, ask the dog, “Did you have a good time?” Do not greet or greet your dog, but calm your dog's excitement by indifference, such as changing clothes or tidying up.
2. After that, when the dog's excitement subsides, praise him calmly.
3. If you suddenly become indifferent, some dogs will temporarily behave worse, but it is an educational process, so please do not give up and proceed.
Tip) Children who jump, scratch their paws, or stop barking for a while after returning from the outdoors can also reinforce their behavior, so please greet or compliment your dog without being excited. Borderline, anxiety about specific people - Identify the cause
1. Trauma, vigilance, or anxiety caused by a specific person's harsh behavior (e.g. shouting, hitting, etc.).
2. If excessive skinship is regarded as a harassing behavior
3. Inadvertently instilling unpleasant or frightening memories in the dog (eg, the person giving the bath, unintentionally hurting the dog, etc.)

- solution
@ In order to get rid of the dog's alertness and anxiety, certain people need to increase the amount of time they spend with their dog.
1. A specific guardian is in charge of walking, playing, eating, and snacking.
2. Induce them to your arms naturally through snacks or toys.
3. Please allow plenty of time for your dog to approach you calmly (it may take several days depending on the situation).
4. If your dog approaches or comes into your arms right in front of you, give him skinship and create a better relationship. Excessive affection or possessiveness toward a specific person - Identify the cause
1. Obsessive and possessive behavior resulting from excessive affection for a specific person.
2. When behavior such as barking or showing aggression and avoiding a place by a specific person is taken as attention or praise, or the behavior is reinforced by another family member leaving the place.

- solution
@ In the case of a dog-obsessed family, you need to let them know that the behavior is not good.
@ Prior Learning: Get down, wait
1. The process of familiarity with other family members should take precedence (the family members take turns providing snacks, meals, and walks evenly).
2. Find out when your dog's alert behavior is severe and mild.
3. Begin educating your dog on the behavior at the time when the alert behavior was the weakest (it is recommended that all training be changed little by little from low to strong).
4. For example, if your dog barks when he is on his lap, show your intention of refusal, such as silently standing up and letting him go down naturally or turning his back.
5. Afterwards, give instructions such as “lay down” or “wait” and wait (don’t use the directive too often and give your dog time to think).
6. Reward your dog as he calms down (reward is better if the dog's owner can provide it).
Tip) If you reward your dog as soon as it stops barking, it may repeat the barking behavior, so reward your dog after a certain amount of time has passed. demand barking - Identify the cause
1. “Required barking” is a learning-reinforced behavior that says that when a dog barks, scratches, jumps, etc., the guardian will do the behavior the dog wants.
2. Even if the behavior is not demanding, there are many situations where the behavior may have been reinforced by interest or touch (eg, the behavior of a guardian to stop the dog from barking or hugging him to get out of the seat).

- solution
@ Recognize that barking does not prevent snacks, games, walks, etc.
@ At first, you may be more barking or agitated because of previous experiences. If you are rewarded again at this time, the “required behavior” may become more severe.
@ Prior Learning: Wait
1. When your dog shows demanding behavior and barks, tell him to “wait” and the guardian to stand still and wait.
2. When the barking dog calms down over time, reward it with a treat and praise it.
3. In the early stages of training, there may be cases where your dog becomes agitated and does not understand “wait” and continues to perform demanding behaviors. At this time, please do not give up and wait for the dog to calm down.
4. If your dog calms down and responds with excitement again when he tries to give you a reward, stop doing the reward and wait again.
5. At first, practice waiting for a short time, then gradually increase the time. Set a standard time so that the dog can concentrate more without predicting the time, and give a reward randomly (eg, if the standard is 5 seconds, between 3 and 6 seconds) rewards randomly).
Tip) When your dog does not succeed immediately and does other actions, it is important not to panic or repeat instructions (wait, etc.), but to give the dog enough time to think while the guardian waits. unfamiliar pets - Identify the cause
1. Barking caused by an increase in excitement level due to various stimuli (open door, greeting, barking of other dogs, etc.)
2. Boundary barking due to the entry of an unfamiliar dog into a place perceived as one's own space.
3. Behavior of barking or aggression due to anxiety caused by bad memories after visiting an unfamiliar dog.
4. Barking due to fear and alertness of other dogs due to missed socialization period.

- solution
@ I need another dog to help. (A docile dog that has been socialized)
@ This is an education that removes the alertness with unfamiliar dogs and creates good awareness.
< Step 1 >
1. Take your dog outside for a walk and come in with him (this can lower alertness).
2. When you get close to other dogs on a walk, give them snacks, etc., to instill good awareness of other dogs.
3. When you come home together, you may lose your alertness and the behavior of barking may disappear.
4. Make fun with other dogs by playing games and snacks.
5. If you are still barking at an unfamiliar dog at home after a walk, proceed with the following additional training.
< Step 2 >
1. Encourage other dogs to turn their backs with a snack, and focus on the owner to avoid barking at other dogs as much as possible.
2. When your dog eats a snack, give them praise and skinship to lower the excitement level and create a good awareness of other dogs It is recommended to hold it in your hand).
3. At this time, if the guardian's dog is sensitive to the movement of an unfamiliar dog, let the helping dog stay in one place through the lead leash.
4. If you are not sensitive to the movements of other dogs while you are snacking, release the helping dog comfortably.
5. When barking is reduced, create a pleasant time through play, snacks, etc.
Tip) When another dog visits, provide the same training from the beginning, and if the barking has decreased to some extent, proceed from the second step. separation anxiety - Identify the cause
1. When your dog is alone at home, if he continues to bark, destroys everything, or makes a bowel movement mistake, it may be separation anxiety.
2. Even if separation anxiety is not severe, you may be more sensitive to noise than usual due to increased anxiety or alertness due to the absence of a guardian.
3. It is possible that you have a trauma or experience of increased arousal due to a surprise visit by an outsider after a certain sound.
4. You may be more sensitive due to not being able to walk for a long time or becoming more vigilant in the house due to increased stress.
5. Dogs who are usually wary or anxious about external stimuli or sounds may be more sensitive to them because they do not have a guardian.

- solution
@ If the problem is not caused by sound or stimulation, we recommend that you provide education on separation anxiety.
@ Through “sound desensitization education”, we provide an education that changes bad memories of sound into good memories through compensation.
@ Prior learning: one or more of mat, house, sit down, and wait
1. If you are at home, your dog may be alert or stressed, so it is recommended to proceed after a light walk or play.
2. Prioritize the 'wait' education using a mat or house where your dog can normally rest comfortably.
3. During 'Wait', listen to the dog's usual reaction, starting with a small sound. Instruct your dog to enter the mat or house if it responds to the sound.
4. If you do well on the mat or house, reward them with compliments and snacks.
5. If successful, increase the volume little by little and repeat.
6. If the training does not proceed well at a certain level, please lower the sound level by one or two and proceed again.
Tip) Position the mat or house at a place close to where the sound is coming from at first, and as the success rate of your dog increases, move it to a place where you can rest more comfortably. A certain kind of animal, a certain kind of outsider boundary - Identify the cause
1. If you react to an animal or a specific person (man, child, etc.), you often express anxiety due to some trauma or lack of social skills.
2. Socialization period Inexperienced or bad memories of people and animals.
3. When people or animals with bad memories do not approach or disappear due to their barking or aggression behavior, which is naturally learned and reinforced little by little.
※ Caution: Children's innocent behavior toward puppies may be perceived as intimidating or rude, so please be extra careful.

- solution
@ It is a good idea to proceed with an education method that instills the awareness that good things happen when a stranger or a dog approaches without harming you.
@ Prior Learning: Eye Contact
1. Determine where your dog starts barking when a stranger or dog approaches from a distance.
2. When a stranger or dog approaches, call his name before barking and focus on the guardian before the dog responds through eye contact.
3. Give a snack while a stranger or dog passes by (If you see a stranger or dog barking while you are snacking, try again and give the treat away from the stranger) .
4. When you get used to it, gradually increase the interval between consecutive rewards. If the distance is too close, it is also a good idea to have a stranger, rather than a guardian, give you a snack.
5. If you fail while eating a snack, please start over from the beginning.
6. Please keep repeating learning.
Tip) For children who eat snacks quickly and bark right away, grab them a lot and eat them little by little, or hold a chewy snack in your hand so they can focus on the snack until someone passes by. specific object boundary - Identify the cause
1. When reacting to a specific object, it is often alertness or fear.
2. Socialization period There are times when you are afraid because you do not have much experience.
3. A case in which a certain object disappears due to one's own barking or attacking behavior, which is naturally learned and reinforced little by little.
4. When a dog barks while following a moving object due to the dog's chasing instinct.

- solution
@ It is good to proceed with an education method that instills the awareness that good things will happen when an unfamiliar object approaches without harming you (at first, please proceed in a quiet space where there are not many stimuli).
@ Prior Learning: One of Eye Contact, Come.
1. During a natural walk of the dog, the guardian watches for objects approaching.
2. Know the distance your dog reacts to unfamiliar objects (motorcycles, bicycles, etc.).
3. If an unfamiliar object approaches, focus on the guardian before the dog reacts through eye contact.
4. Give a snack while an unfamiliar object is passing by (If you find an unfamiliar object while you are snacking and show barking, repeat the process and give the snack away from the unfamiliar object).
5. Start walking again when an unfamiliar object passes by.
6. Please keep repeating learning.
Tip) For children who eat snacks quickly and bark right away, grab them a lot and eat them little by little, or hold a chewy snack in your hand so they can focus on the snack until someone passes by. Excessive interest in unfamiliar dogs - Identify the cause
1. In the case of a dog that lacks self-control or does not meet his/her activity level, barking due to stress, excessive interest, excitement, or curiosity about a stranger or unfamiliar dog.
2. When the dog approaches the stranger with a cheerful welcome, the behavior is reinforced by the stranger's compliments and touch.

- solution
@ The most important thing is to relieve the child's desires and stress sufficiently through walking, many experiences, and various games on a daily basis.
@ We teach dogs to develop self-control.
@ Prior Learning: Wait, eye contact, come
1. Please wear a leash and harness during training. Dogs who are easily aroused can benefit from wearing an anti-slip harness.
2. If your dog starts barking, do not give any rewards such as attention, skinship, etc., and end the situation your dog likes, such as moving to another place.
(Even if your dog does not want to follow you, move firmly.)
3. After moving from place to place, do not react to the dog's behavior and wait calmly (not using directives at the beginning of training and waiting calmly by the guardian can help reduce confusion for the dog).
4. Give rewards when excitement subsides.
5. Repeat steps 1 to 4 and slowly increase the waiting time.
6. Don't scold or get angry even if your dog is not concentrating well at first (this can make training more difficult by making your dog more excited or trying to run away).
7. Use directives (wait, come, etc.) when your dog begins to follow the behavior of the owner to some extent.
Tip) When a dog barks, giving a reward such as attention or skinship (hugging, etc.) can reinforce the behavior.

For example, the server 20 may obtain training guidelines for each behavior of the companion animal through web crawling, or may receive training guidelines input through a user terminal of at least one expert. And, the server 20 may train each of the plurality of artificial intelligence models (121, 122, 123, ...) to provide a guide for discipline to the user according to each discipline guideline.

Here, each AI model is a deep learning-based neural network model, and may include one or more layers. Each AI model may be trained according to a method such as a Recurrent Neural Network (RNN), a Deep Q-Network (DQN), or the like.

As an example, each artificial intelligence model may be trained to provide a discipline guide for each step, according to the discipline guidelines.

In addition, each artificial intelligence model, when the user's response to the discipline guide provided for each step is input, it can be trained to provide a discipline guide or additional information of the next step matching the response.

Here, the user's response may be a progress report, a question, or the like.

Progress report may be a response from the user informing that the partial discipline process was performed according to the discipline guide and/or notifying the result of performing the partial discipline process.

The question may be a response from a user requesting more additional / specific explanation or related additional information to the provided discipline guide.

That is, each artificial intelligence model, based on the RNN, may be trained to provide a discipline guide for each step, but to generate a conversation content (discipline guide) that follows naturally according to the user's response.

At this time, each artificial intelligence model may be trained using, as a training data set, actual conversation content between an expert who provides a guide for each step according to the discipline guidelines and a guardian who receives the guide, but is not limited thereto.

On the other hand, in the above-described embodiment, each AI model is implemented/defined in an end-to-end manner that generates (outputs) the conversation contents corresponding to the step-by-step discipline guide as the user's response (progress report, question) is input. explained to be.

However, each AI model may be defined/designed in a form including a plurality of unit AI models having different functions.

As an example, each AI model includes a unit model for understanding (converting) natural language (eg, a user's response) in a vector form, a unit model for identifying the next vector following the transformed vector (RNN-based), the identified It can be defined/implemented in a form including a unit model for converting a vector back into a natural language (eg, a training guide for the next stage), respectively.

The guide generation module 130 is a module for providing a guide using at least one artificial intelligence model among a plurality of artificial intelligence models 121, 122, 123, ....

1, it has been described that the chatbot interface module 110 is included in the user terminal 10, and the artificial intelligence model management module 120 and the guide generation module 130 are stored in the server 20. , the location of each module can be changed.

For example, at least one of the artificial intelligence model management module 120 and the guide generation module 130 may be included in the user terminal 10 .

When the artificial intelligence model management module 120 is included in the user terminal 10 , the aforementioned plurality of artificial intelligence models 121 , 122 , 123 , ... may also be stored in the user terminal 10 .

On the other hand, with reference to FIG. 2 below, a discipline guide method of the chatbot system 100 will be described in order.

2 is a flowchart illustrating a discipline guide method of a chatbot system according to an embodiment of the present disclosure.

Referring to FIG. 2 , the chatbot system 100 may identify an artificial intelligence model corresponding to an action selected according to a user input among a plurality of artificial intelligence models 121 , 122 , 123 , ... ( S210 ).

Here, the user input may be received through the user terminal 100, and the behavior of the companion animal to be disciplined may be selected through the user input.

And, the chatbot system 100, a plurality of artificial intelligence models (121, 122, 123, ...) can identify the trained artificial intelligence model according to the discipline guidelines for correcting the selected behavior.

However, it is also possible when there are multiple AI models corresponding to the selected behavior.

For example, even if they are for the same behavior of 'to defecate anywhere', the training guidelines for dogs before 3 months of age and those for dogs after 3 months of age may be different, and therefore the same behavior A plurality of artificial intelligence models may be registered for

Similarly, training guidelines may vary depending on the age and gender of the companion animal, even for the same behavior, and a plurality of different artificial intelligence models may be registered according to age and gender.

* As such, when there are multiple AI models registered for the selected behavior, the chatbot system 100 identifies one AI model according to at least one of the species, age, and gender of the companion animal selected according to the user input. can

In general, information on the species, age, gender, etc. of the companion animal is inputted from the user according to the request of the chatbot system 100 when the application provided by the chatbot system 100 is first executed through the user terminal 10 . may be, but is not limited thereto.

And, the chatbot system 100 may provide a guide for the discipline of the companion animal based on the identified artificial intelligence model (S220).

Specifically, the guide generating module 130 may acquire the conversation content (eg, step-by-step discipline guide) output through the identified artificial intelligence model, and transmit it to the chatbot interface module 110 .

As a result, a guide for discipline may be provided through the chatbot interface module 110 .

In this case, the guide generation module 130 may provide a step-by-step discipline guide (conversation content) output from the (identified) artificial intelligence model to the user step by step through the chatbot interface module 110 .

Here, the guide generation module 130 inputs the user's step-by-step responses (progress reports, questions, etc.) received through the chatbot interface module 110 back into the AI model to obtain the next conversation content of the AI model. As such, it is possible to provide a step-by-step discipline guide to the user.

As a specific example, the chatbot system 100 may provide a first-stage discipline guide, and then provide a second-stage discipline guide according to the user's progress report for the first-stage discipline guide.

As such, the chatbot system 100 according to the present disclosure guides the discipline instructions step-by-step according to the conversation process with the user performed through the artificial intelligence model, so that the individual (real-time) situation of the user and the companion animal and the emotional line are appropriate. It has the advantage of inducing an appropriate discipline process.

That is, the chatbot system 100 according to the present disclosure has an effect that is clearly differentiated from the case where the discipline guidelines for correcting the behavior of companion animals are thrown to the user as simply listed knowledge.

Meanwhile, in the above-described embodiments, only the case in which an artificial intelligence model trained for each behavioral guideline of a companion animal is used has been described.

In this case, the knowledge graph may include information on one or more discipline guides (conversation contents) for each step. And, the knowledge graph, according to the user's response received for each step can be constructed so that the next step of the discipline guide is different.

That is, one or more user responses predicted to follow for each step-by-step discipline guide may be pre-stored, and discipline guides of the next step to follow for each predicted user response may also be preset.

Meanwhile, an example of a service provided by the chatbot system 100 will be described in more detail below with reference to FIGS. 3A to 3B . 3A to 3B show a user terminal 10 and a server 20 implemented as a smart phone.

FIG. 3A is a diagram for describing an operation in which a chatbot system receives a user input for selecting an action of a companion animal according to an embodiment of the present disclosure.

Referring to FIG. 3A , the user terminal 10 may provide a user interface (UI) 310 for selecting the behavior of a companion animal to be disciplined while executing an application of the server 20 .

For example, when a user input of touching the displayed “bark at a specific sound” item is detected, the user terminal 10 selects a “bark at a specific sound” action, and transmits information on the selected action to the server 20 can

Meanwhile, in FIG. 3A , when the 'View More' item is selected, the user terminal 10 may further display at least one item in addition to the currently displayed action items.

And, when the 'search' item is selected, the user terminal 10 may receive a search word from the user and display at least one action including the input search word.

3B is a diagram for explaining an operation of providing a discipline guide according to a selected action by the chatbot system according to an embodiment of the present disclosure.

FIG. 3B is assuming that the action “barking at a specific sound” is selected in FIG. 3A .

Referring to FIG. 3B , the chatbot interface module 110 of the user terminal 10 may provide a UI 320 for a conversation between the chatbot and the user. In FIG. 3B , the chatbot was named 'AI Trainer'.

Meanwhile, the server 20 may obtain a training guide for each step through an artificial intelligence model corresponding to the selected action (barking at a specific sound) and transmit it to the user terminal 10 .

As a result, referring to FIG. 3B , the user terminal 10 may provide a chatting message 321 including a training guide of the first stage.

Here, the chatting message 321 ′ may be obtained according to a user input in response to the chatting message 321 .

At this time, the user terminal 10 may transmit the chatting message 321' to the server 20, and the server 20 inputs the chatting message 321' to the artificial intelligence model, including a training guide for the next step. A chat message 322 may be generated.

As a result, referring to FIG. 3B , a chatting message 322 may be provided through the user terminal 10 .

On the other hand, it is also possible that the user's response to the step-by-step training guide is a request for a question or supplementary explanation.

For example, when a user's response to a question about the training guide included in the chat message 322 is received (eg, what is waiting training?), the chatbot system 100 provides an answer to the question. can be created and provided.

Alternatively, when a response from the user requesting supplementary explanation for the training guide included in the chat message 322 (eg, wait and tell me how to train in detail) is received, the chatbot system 100 includes the supplementary explanation information, or at least one pre-stored demonstration image (eg, a waiting training image of a similar breed) may be provided.

Meanwhile, according to an embodiment, the chatbot system 100 may acquire a plurality of discipline instructions for one specific behavior of the companion animal.

In this case, the chatbot system 100 may acquire a plurality of discipline guidelines for the same behavior in a state where the species, age, or gender of the companion animal are unified.

And, the chatbot system 100 may train a plurality of artificial intelligence models through each of a plurality of discipline guidelines, respectively.

That is, a plurality of artificial intelligence models may be independently trained and prepared for the correction of the same behavior.

In this case, the chatbot system 100 may provide a guide based on one artificial intelligence model arbitrarily selected from among a plurality of artificial intelligence models to each of the users who request correction of the corresponding behavior of the companion animal.

Here, the chatbot system 100 may receive the user's evaluation of the guide provided through each artificial intelligence model, and may calculate a score for each of the plurality of artificial intelligence models based on the input user's evaluation.

This score may be understood as a score for each of a plurality of discipline guidelines used to train each of a plurality of artificial intelligence models.

Through this embodiment, the chatbot system 100 may generate evaluation information or ranking information about which discipline guideline is most effective for a specific action.

Specifically, when provided through the artificial intelligence model used by the chatbot system 100, one indicator indicating which discipline guideline is most effective can be made.

4 is a block diagram illustrating a hardware configuration of a chatbot system according to various embodiments of the present disclosure.

Referring to FIG. 4 , the user terminal 10 may include a memory 11 and a processor 12 .

The memory 11 is a configuration for storing an operating system (OS) for controlling the overall operation of the components of the user terminal 10 and at least one instruction or data related to the components of the user terminal 10 . .

The memory 11 may include a non-volatile memory such as a ROM and a flash memory, and may include a volatile memory such as a DRAM. Also, the memory 11 may include a hard disk, a solid state drive (SSD), or the like.

The processor 12 is a configuration for controlling the user terminal 12 as a whole. Specifically, the processor 12 may perform operations according to various embodiments of the present disclosure by executing at least one instruction stored in the memory 11 while being connected to the memory 11 .

Processor 12 may be comprised of one or more processors. In this case, the one or more processors may include a general-purpose processor such as a CPU, an AP, a digital signal processor (DSP), or the like, a graphics-only processor such as a GPU, a VPU (Vision Processing Unit), or the like, or an artificial intelligence-only processor such as an NPU.

Referring to FIG. 4 , the user terminal 10 may further include at least one of a communication unit 13 , a display 14 , a user input unit 15 , and an audio output unit 16 .

The communication unit 13 is configured to communicate with at least one external device (eg, the server 20 ).

The communication unit 13 may be connected to an external device based on a network implemented through wired communication and/or wireless communication. In this case, the communication unit 13 may be directly connected to an external device, or may be connected to an external electronic device through one or more external servers (eg, Internet Service Providers (ISPs)) that provide a network.

The network may be a personal area network (PAN), a local area network (LAN), a wide area network (WAN), etc. depending on the area or size, and depending on the openness of the network, an intranet, It may be an extranet or the Internet.

Wireless communication includes long-term evolution (LTE), LTE Advance (LTE-A), 5th generation (5G) mobile communication, code division multiple access (CDMA), wideband CDMA (WCDMA), universal mobile telecommunications system (UMTS), WiBro (Wireless Broadband), GSM (Global System for Mobile Communications), DMA (Time Division Multiple Access), WiFi (Wi-Fi), WiFi Direct, Bluetooth, NFC (near field communication), at least one of the communication methods such as Zigbee may include

Wired communication may include at least one of communication methods such as Ethernet, optical network, USB (Universal Serial Bus), and ThunderBolt. Here, the communication unit 150 may include a network interface or a network chip according to the above-described wired/wireless communication method. Meanwhile, the communication method is not limited to the above-described example, and may include a communication method newly appearing according to the development of technology.

The display 14 is a configuration for visually providing various information.

The user terminal 10 may visually provide at least a portion of various web pages or applications constituting the chatbot service of the above-described chatbot system 100 through the display 14 .

The display 14 may be implemented as a liquid crystal display (LCD), a plasma display panel (PDP), an organic light emitting diode (OLED), a transparent OLED (TOLED), a micro LED, etc., but is not limited thereto. It may include various known types of displays.

The display 14 may be implemented in the form of a touch screen capable of sensing a user's touch manipulation, or may be implemented as a flexible display that can be folded or bent.

The user input unit 15 is configured to receive various user inputs or user commands. The user input unit 15 may include various components such as one or more buttons, a touch screen, a microphone, a keypad, and a camera.

The user terminal 10 may receive the behavior of the companion animal to be corrected through the user input unit 15 and may also receive the user's conversation content.

The audio output unit 16 is configured to provide a variety of information aurally, and may include a speaker, an earphone/headphone terminal, and the like.

The user terminal 10 may output the chatbot voice corresponding to the chatbot conversation content (guide) through the audio output unit 16 .

Referring to FIG. 4 , the server 20 includes a memory 21 and a processor 22 . In addition, the server 20 may include a communication unit 23 for performing communication with the user terminal 10 .

The memory 21 may include the various artificial intelligence models described above in addition to the OS and at least one instruction.

According to an embodiment, the processor 22 loads only the artificial intelligence model corresponding to the action selected according to the user input among the plurality of artificial intelligence models 121 , 122 , 123 , ... stored in the memory 21 to correspond to the guide. You can create conversation content.

The server 20 may provide at least a portion of a web page or application for providing a service of the chatbot system 100 to the user terminal 10 through the communication unit 23 .

Meanwhile, the various embodiments described above may be implemented in a recording medium readable by a computer or a similar device using software, hardware, or a combination thereof.

According to the hardware implementation, the embodiments described in the present disclosure are ASICs (Application Specific Integrated Circuits), DSPs (digital signal processors), DSPDs (digital signal processing devices), PLDs (programmable logic devices), FPGAs (field programmable gate arrays) ), processors, controllers, micro-controllers, microprocessors, and other electrical units for performing other functions may be implemented using at least one.

In some cases, the embodiments described herein may be implemented by the processor itself. According to the software implementation, embodiments such as the procedures and functions described in this specification may be implemented as separate software modules. Each of the software modules described above may perform one or more functions and operations described herein.

On the other hand, the computer instructions or computer program for performing the processing operation in the chatbot system 100 according to various embodiments of the present disclosure described above is a non-transitory computer-readable medium. can be stored in When the computer instructions or computer program stored in such a non-transitory computer-readable medium are executed by the processor of the specific device, the specific device performs the processing operation in the chatbot system 100 according to the various embodiments described above.

The non-transitory computer-readable medium refers to a medium that stores data semi-permanently, rather than a medium that stores data for a short moment, such as a register, a cache, a memory, and can be read by a device. Specific examples of the non-transitory computer-readable medium may include a CD, DVD, hard disk, Blu-ray disk, USB, memory card, ROM, and the like.

In the above, preferred embodiments of the present disclosure have been illustrated and described, but the present disclosure is not limited to the specific embodiments described above, and is generally used in the technical field belonging to the present disclosure without departing from the gist of the present disclosure as claimed in the claims. Various modifications may be made by those having the knowledge of

100: chatbot system

Claims (1)

In the companion animal discipline guide method of the chatbot system,
receiving a user input for selecting at least one behavior from among a plurality of behaviors of the companion animal;
Identifying an artificial intelligence model trained based on the discipline guidelines for the selected behavior from among a plurality of first artificial intelligence models trained to provide a guide for the discipline of companion animals; and
Based on the identified artificial intelligence model, providing a guide for the discipline of the companion animal performing the selected action; including,
The step of providing the guide includes:
Based on the output of the identified artificial intelligence model to provide a discipline guide of the first step,
When the user's progress report is received for the provided first-stage discipline guide, based on the output of the identified artificial intelligence model to which the progress report is input, the second-stage discipline guide is provided,
When a user's question for the provided first-step discipline guide is received, an answer to the question is provided based on the output of the identified artificial intelligence model to which the question is input,
When a user input requesting a supplementary explanation is received with respect to the provided first-stage discipline guide, providing at least one demonstration image related to the first-stage discipline guide,
The companion animal discipline guide method of the chatbot system is,
Independently training each of the artificial intelligence models based on each of a plurality of different discipline guidelines for correcting a specific behavior of the companion animal; and
Based on the evaluation of users for the guide provided through each of the artificial intelligence models, calculating a score for each of the plurality of discipline guidelines; includes,
The companion animal discipline guide method of the chatbot system is,
acquiring discipline guidelines for each behavior of companion animals; and
To provide a guide based on the discipline guidelines, training each of the first plurality of artificial intelligence models according to the discipline guidelines for each behavior; further comprising,
The step of identifying an artificial intelligence model corresponding to the selected behavior comprises:
When there are a plurality of artificial intelligence models registered for the selected behavior, a chatbot that selects one artificial intelligence model from among the plurality of artificial intelligence models according to at least one of a species, age, and sex of a companion animal selected according to a user input How the system guides the discipline of companion animals.

Images