KR102474974B1 - Method, device and system for analyzing brand based on artificial intelligence - Google Patents

Abstract

According to an embodiment of the present invention, provided is a method for analyzing a brand based on artificial Intelligence (AI), which is performed and provided by a device, comprising: a step of collecting posted data on a first brand among posts uploaded on the SNS accounts of users; a step of, when the posted data are collected for a preset first period, extracting a first keyword information based on a hashtag of the posted data collected for the first period; and a step of applying the first keyword information to a first artificial neural network, and generating a first analysis result showing the recognition of the users on the first brand for the first period based on an output of the first artificial neural network. Therefore, the recognition of users on a brand can be analyzed.

Description

AI-based brand analysis method, device and system {METHOD, DEVICE AND SYSTEM FOR ANALYZING BRAND BASED ON ARTIFICIAL INTELLIGENCE}

The following embodiments relate to a technique for analyzing users' perceptions of a brand based on artificial intelligence.

A brand is a general term for names, symbols, designs, etc. used to identify a specific product, and brand names that can be expressed in words, and symbols, designs, and lettering that cannot be expressed in words are called brand marks.

A company that sells a trademark assigns a brand to its products in order to clearly distinguish them from competitors' products, but it is not irrelevant to the existence of brand loyalty among consumers. Brand loyalty is the favorable attitude consumers have toward a particular brand in brand selection and the consequent tendency to purchase the same brand repeatedly.

Accordingly, if a certain brand has a high reputation, if the same brand is used when introducing a new product related to that brand, sales promotion activities for each product can bring about a synergistic effect under one brand.

Therefore, there is an increasing demand for analyzing users' perceptions of brands, and research on technologies related thereto is required.

Korean Patent Registration No. 10-2322845 Korean Patent Registration No. 10-1838162 Korean Patent Registration No. 10-1799561 Korean Patent Registration No. 10-1679750

According to one embodiment, posting materials for a first brand among posts uploaded on SNS by users are collected, and when posting materials are collected during a preset first period, a hash tag of the posting materials collected during the first period is set. base, first keyword information is extracted, and the first keyword information is applied to the first artificial neural network, so that based on the output of the first artificial neural network, a first representation of users' perception of the first brand during the first period is obtained. Its purpose is to provide an artificial intelligence-based brand analysis method, device, and system that generate analysis results.

The object of the present invention is not limited to the object mentioned above, and other objects not mentioned will be clearly understood from the description below.

According to one embodiment, a method for analyzing a brand based on artificial intelligence, performed by a device, comprising: collecting posting data for a first brand among postings uploaded on SNS of users; extracting first keyword information based on a hash tag of the posted material collected during the first period when the posting material is collected during a first preset period; and applying the first keyword information to a first artificial neural network to generate a first analysis result representing users' perception of the first brand during the first period based on an output of the first artificial neural network. Including, an artificial intelligence-based brand analysis method is provided.

The artificial intelligence-based brand analysis method includes extracting second keyword information based on hash tags of posted materials collected during the second period when the posting material is collected during a second period after the first period. ; applying the second keyword information to the first artificial neural network to generate a second analysis result representing users' perception of the first brand during the second period based on an output of the first artificial neural network; ; and comparing the first analysis result and the second analysis result to generate a third analysis result representing a change in users' perception of the first brand.

The artificial intelligence-based brand analysis method may include obtaining first brand information including information about a direction, target, and value that the first brand seeks to pursue; comparing a first list that is a list of words included in the first analysis result with a second list that is a list of words included in the first brand information; As a result of comparing the first list and the second list, words included in the first list are included in the 1-1 list, which is a list of words included in the second list, and words not included in the second list. Dividing into a first-second list that is a list of words; Calculating a first ratio through a value obtained by dividing the number of words in the 1-1 list by the number of words in the first list; checking whether the first ratio is greater than a preset reference ratio; calculating a first value by applying a preset first weight to the first ratio when it is confirmed that the first ratio is greater than the reference ratio; and setting the first numerical value as an evaluation index of the first brand during the first period.

According to one embodiment, posting materials for a first brand among posts uploaded on SNS by users are collected, and when posting materials are collected during a preset first period, a hash tag of the posting materials collected during the first period is set. base, first keyword information is extracted, and the first keyword information is applied to the first artificial neural network, so that based on the output of the first artificial neural network, a first representation of users' perception of the first brand during the first period is obtained. By generating the analysis result, there is an effect of analyzing users' perception of the brand.

On the other hand, the effects according to the embodiments are not limited to those mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

1 is a diagram schematically showing the configuration of a system according to an embodiment.
2 is a flowchart illustrating a process of analyzing users' perceptions of a brand based on artificial intelligence according to an embodiment.
3 is a flowchart illustrating a process of analyzing changes in users' perceptions of a brand based on artificial intelligence according to an embodiment.
4 is a flowchart illustrating a process of calculating an evaluation index of a brand according to an embodiment.
5 is a flowchart illustrating a process of calculating a brand evaluation index according to another embodiment.
6 is a diagram for explaining learning of a first artificial neural network according to an embodiment.
7 is an exemplary diagram of a configuration of a device according to an embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, since various changes can be made to the embodiments, the scope of the patent application is not limited or limited by these embodiments. It should be understood that all changes, equivalents or substitutes to the embodiments are included within the scope of rights.

Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only, and may be modified and implemented in various forms. Therefore, the embodiments are not limited to the specific disclosed form, and the scope of the present specification includes changes, equivalents, or substitutes included in the technical spirit.

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

It should be understood that when an element is referred to as being “connected” to another element, it may be directly connected or connected to the other element, but other elements may exist in the middle.

Terms used in the examples are used only for descriptive purposes and should not be construed as limiting. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the embodiment belongs. 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 unless explicitly defined in the present application, they should not be interpreted in an ideal or excessively formal meaning. don't

In addition, in the description with reference to the accompanying drawings, the same reference numerals are given to the same components regardless of reference numerals, and overlapping descriptions thereof will be omitted. In describing the embodiment, if it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the embodiment, the detailed description will be omitted.

The embodiments may be implemented in various types of products such as personal computers, laptop computers, tablet computers, smart phones, televisions, smart home appliances, intelligent vehicles, kiosks, and wearable devices.

In an embodiment, an artificial intelligence (AI) system is a computer system that implements human-level intelligence, and unlike existing rule-based smart systems, a machine learns and judges on its own. The more AI systems are used, the higher the recognition rate and the more accurate understanding of user preferences. Conventional rule-based smart systems are gradually being replaced by deep learning-based AI systems.

Artificial intelligence technology consists of machine learning and element technologies using machine learning. Machine learning is an algorithm technology that classifies/learns the characteristics of input data by itself, and element technology is a technology that uses machine learning algorithms such as deep learning to mimic functions such as recognition and judgment of the human brain. It consists of technical fields such as understanding, inference/prediction, knowledge expression, and motion control.

The various fields where artificial intelligence technology is applied are as follows. Linguistic understanding is a technology for recognizing and applying/processing human language/characters, including natural language processing, machine translation, dialogue systems, question and answering, voice recognition/synthesis, and the like. Visual understanding is a technology for recognizing and processing objects like human vision, and includes object recognition, object tracking, image search, person recognition, scene understanding, space understanding, image improvement, and the like. Inference prediction is a technique of reasoning and predicting logically by judging information, and includes knowledge/probability-based reasoning, optimization prediction, preference-based planning, and recommendation. Knowledge expression is a technology that automatically processes human experience information into knowledge data, and includes knowledge construction (data creation/classification) and knowledge management (data utilization). Motion control is a technology for controlling the autonomous driving of a vehicle and the movement of a robot, and includes motion control (navigation, collision, driving), manipulation control (action control), and the like.

In general, in order to apply machine learning algorithms to real life, learning is performed in a trial and error method due to the nature of the basic methodology of machine learning. In particular, deep learning requires hundreds of thousands of iterations. It is impossible to execute this in an actual physical external environment, so instead, the actual physical external environment is virtually implemented on a computer and learning is performed through simulation.

1 is a diagram schematically showing the configuration of a system according to an embodiment.

Referring to FIG. 1 , a system according to an embodiment may include a manager terminal 100 and a device 200 .

First, the manager terminal 100 and the device 200 may be connected through a communication network, and the communication network is configured regardless of its communication mode, such as wired and wireless, so that communication between servers and communication between servers and terminals is performed It can be implemented in various forms.

The manager terminal 100 may be implemented as a computing device having a communication function, and may be implemented as, for example, a mobile phone, a desktop PC, a laptop PC, a tablet PC, a smart phone, etc., but is not limited thereto, and may be implemented as an external device. It may be implemented in various types of communication devices that can be connected to the server.

The administrator terminal 100 uses the device 200 to access a web page operated by a service provider or organization, or an application developed and distributed by a service provider or organization using the device 200. can be installed The manager terminal 100 may interwork with the device 200 through a web page or application.

Each manager terminal 100 may access the device 200 through a web page or application provided by the device 200 .

The device 200 may be a server owned by a person or organization that provides services using the device 200, may be a cloud server, or may be a peer-to-peer (p2p) set of distributed nodes. may be The device 200 may be configured to perform all or some of the arithmetic function, storage/reference function, input/output function, and control function of a normal computer. The device 200 may include one or more artificial neural networks that perform an inference function.

The apparatus 200 may be configured to communicate with the manager terminal 100 by wire or wirelessly, control the overall operation of the manager terminal 100, and control which information is to be displayed on the screen of the manager terminal 100. have.

Meanwhile, for convenience of explanation, only the manager terminal 100 is illustrated in FIG. 1 , but the number of terminals may vary according to embodiments. As long as the processing capacity of the device 200 allows, the number of terminals is not particularly limited.

According to an embodiment, the device 200 may analyze users' perception of a brand based on artificial intelligence. To this end, the device 200 may include a plurality of pre-trained artificial neural networks to perform the machine learning algorithm.

In the present invention, artificial intelligence (AI) means a technology that imitates human learning ability, reasoning ability, perception ability, etc., and implements it with a computer, and concepts such as machine learning and symbolic logic can include Machine learning (ML) is an algorithm technology that classifies or learns the characteristics of input data by itself. Artificial intelligence technology is a machine learning algorithm that analyzes input data, learns the result of the analysis, and can make judgments or predictions based on the result of the learning. In addition, technologies that mimic the functions of the human brain, such as recognition and judgment, using machine learning algorithms, can also be understood as artificial intelligence. For example, technical fields such as linguistic understanding, visual understanding, inference/prediction, knowledge expression, and motion control may be included.

Machine learning may refer to processing that trains a neural network model using experience of processing data. Through machine learning, computer software could mean improving its own data processing capabilities. A neural network model is constructed by modeling a correlation between data, and the correlation may be expressed by a plurality of parameters. The neural network model derives a correlation between data by extracting and analyzing features from given data, and optimizing the parameters of the neural network model by repeating this process can be referred to as machine learning. For example, a neural network model may learn a mapping (correlation) between an input and an output with respect to data given as an input/output pair. Alternatively, even when only input data is given, the neural network model may learn the relationship by deriving a regularity between given data.

An artificial intelligence learning model or a neural network model may be designed to implement a human brain structure on a computer, and may include a plurality of network nodes having weights while simulating neurons of a human neural network. A plurality of network nodes may have a connection relationship between them by simulating synaptic activities of neurons that transmit and receive signals through synapses. In the artificial intelligence learning model, a plurality of network nodes can send and receive data according to a convolutional connection relationship while being located in layers of different depths. The artificial intelligence learning model may be, for example, an artificial neural network model, a convolutional neural network model (CNN), and the like. As an embodiment, the artificial intelligence learning model may be machine-learned according to methods such as supervised learning, unsupervised learning, and reinforcement learning. Machine learning algorithms for performing machine learning include Decision Tree, Bayesian Network, Support Vector Machine, Artificial Neural Network, Ada-boost , Perceptron, Genetic Programming, Clustering, etc. may be used.

Of these, CNNs are a type of multilayer perceptrons designed to use minimal preprocessing. A CNN consists of one or several convolution layers and general artificial neural network layers on top, and additionally utilizes weights and pooling layers. Thanks to this structure, CNN can fully utilize the input data with a two-dimensional structure. Compared to other deep learning structures, CNN shows good performance in both video and audio fields. CNNs can also be trained via standard back-propagation. CNNs are easier to train than other feedforward artificial neural network techniques and have the advantage of using fewer parameters.

Convolutional networks are neural networks that contain sets of nodes with bound parameters. The increasing size of available training data and the availability of computational power, combined with algorithmic advances such as piecewise linear unit and dropout training, have greatly improved many computer vision tasks. With huge data sets, such as those available for many tasks today, overfitting is not critical, and increasing the size of the network improves test accuracy. Optimal use of computing resources becomes a limiting factor. To this end, a distributed, scalable implementation of deep neural networks can be used.

2 is a flowchart illustrating a process of analyzing users' perceptions of a brand based on artificial intelligence according to an embodiment.

Referring to FIG. 2 , first, in step S201 , the device 200 may collect posting data for a first brand from among postings uploaded on users' SNS.

The device 200 may collect posting data on the first brand from among postings uploaded on SNS (Social Network Service) of users through various techniques such as crawling.

Specifically, the device 200 may collect posting data about the first brand from posts uploaded on SNS from a plurality of SNS servers (not shown) connected to the device 200 . In this case, the device 200 may collect posting materials in which the name of the first brand is set as a hashtag as posting materials for the first brand.

In step S202, the device 200 may check whether or not posting data on the first brand is collected during the first period. Here, the first period may be set differently according to embodiments, and for example, the first period may be set in various ways such as month, quarter, year, and the like.

If it is confirmed in step S202 that no posting materials are collected during the first period, the device 200 may continue to collect posting materials for the first brand, returning to step S201.

If it is determined in step S202 that the posted materials are collected during the first period, in step S203, the device 200 may extract first keyword information based on the hash tag of the posted materials collected during the first period.

For example, during the first period, the first post and the second post were collected, the hashtag of the first post was identified as “brand A”, “luxury product”, and the hashtag of the second post was “brand A”, If it is confirmed as “organic product”, the device 200 may extract keywords including “luxury” and “organic” as keyword information of brand A based on the hash tags of the first post and the second post. .

In step S204, the device 200 may apply the first keyword information to the first artificial neural network.

According to an embodiment, the first artificial neural network may be an algorithm that receives keyword information, analyzes users' perceptions of a brand, and outputs an analysis result.

For example, if “luxury” is included most in the first keyword information, the first artificial neural network receives the first keyword information as an input, and determines that users' perception of the first brand is “luxury”. It can be analyzed, and the number '1' indicating “luxury” can be output as an output value.

In addition, when “organic” is included in the first keyword information, the first artificial neural network receives the first keyword information as an input and analyzes that users' perception of the first brand is “eco-friendly” related to “organic” and the number '2' indicating “eco-friendly” can be output as an output value.

In step S205, the device 200 may generate a first analysis result based on the output of the first artificial neural network. Here, the first analysis result may include information representing users' perceptions of the first brand during the first period.

The first artificial neural network may be trained to analyze what users think of a specific brand through keyword information. The first artificial neural network may be learned through a method described below with reference to FIG. 6 .

When the first analysis result is generated, the apparatus 200 may provide the first analysis result to the manager terminal 100 . Here, the manager terminal 100 is in a state in which the manager account of the first company that operates the first brand is logged in.

3 is a flowchart illustrating a process of analyzing a change in users' perception of a brand based on artificial intelligence according to an embodiment.

Referring to FIG. 3 , first, in step S301 , the device 200 may collect posting data for a first brand from posts uploaded on users' SNS.

In step S302, the device 200 may check whether or not posting data on the first brand is collected during the second period. Here, the second period may be set to have the same length as the first period, starting from a time point after the first period ends. For example, when the first period is set to January, the second period is set to February, and when the first period is set to one quarter, the second period is set to two quarters, and the first period is set to two quarters. If the year 2020 is set, the second period may be set to the year 2021.

If it is determined in step S302 that the posted material is not collected during the second period, the device 200 may continue to collect posted material for the first brand in step S301 .

If it is determined in step S302 that the posted materials are collected during the second period, in step S303, the device 200 may extract second keyword information based on the hash tag of the posted materials collected during the second period.

In step S304, the device 200 may apply the second keyword information to the first artificial neural network.

In step S305, the device 200 may generate a second analysis result based on the output of the first artificial neural network. Here, the second analysis result may include information representing users' perceptions of the first brand during the second period.

When the second analysis result is generated, the device 200 may provide the second analysis result to the manager terminal 100 .

In step S306, the apparatus 200 may compare the first analysis result and the second analysis result to generate a third analysis result representing a change in users' perception of the first brand.

For example, as a result of analyzing users' perceptions of the first brand during the first period, the device 200 may generate a first analysis result including a first word and a second word, and during a second period As a result of analyzing users' perceptions of the first brand, a second analysis result including the first word and the third word may be generated. Thereafter, the device 200 compares the first analysis result and the second analysis result, and confirms that the user's perception of the first brand has changed from the second word to the third word, and from the second word to the third word. It is possible to generate a third analysis result including information indicating a change to .

When the third analysis result is generated, the apparatus 200 may provide the third analysis result to the manager terminal 100 .

4 is a flowchart illustrating a process of calculating an evaluation index of a brand according to an embodiment.

Referring to FIG. 4 , first, in step S401 , the apparatus 200 may generate a first analysis result representing users' perception of the first brand based on the output of the first artificial neural network. Here, the first analysis result includes a plurality of words indicating how users perceive the first brand.

In step S402, the device 200 may obtain first brand information including information about the direction, target, and value that the first brand seeks. To this end, the device 200 may receive first brand information from the manager terminal 100 and store the first brand information in a database, and obtain the first brand information through querying the database. Here, the first brand information includes a plurality of words representing the direction, target, and value that the first brand seeks.

In step S403, the device 200 may generate a first list using words included in the first analysis result.

For example, when the first analysis result includes the first word and the second word, the device 200 may generate a first list including the first word and the second word.

In step S404, the device 200 may generate a second list using words included in the first brand information.

For example, when the first brand information includes the first word and the third word, the device 200 may generate a second list including the first word and the third word.

In step S405, the device 200 may compare the first list and the second list. At this time, the device 200 may check whether each word included in the first list is included in the second list.

In step S406, the device 200 may divide the first list into a 1-1 list and a 1-2 list. Here, the 1-1 list is a list of words included in the second list among the words included in the first list, and the 1-2 list is included in the second list among the words included in the first list. It is a list of words that do not exist.

For example, the first list includes the first word, the second word, the third word, and the fourth word, the second list includes the first word, the second word, the fifth word, and the sixth word; If there is, the device 200 compares the first list and the second list, and since the first word and the second word are included in the first list and also included in the second list, the first word and the second word It is possible to generate a 1-1 list including, and since the third word and the fourth word are included only in the first list and not included in the second list, the first list including the third word and the fourth word -2 Can create lists.

In step S407, the device 200 may calculate the first ratio through a value obtained by dividing the number of words in the 1-1 list by the number of words in the first list. That is, the apparatus 200 may calculate the first ratio through a value obtained by dividing the number of words included in the 1-1 list by the number of words included in the first list.

In step S408, the device 200 may check whether the first ratio is greater than the reference ratio. Here, the reference ratio may be set differently according to embodiments.

If it is confirmed in step S408 that the first ratio is greater than the reference ratio, in step S409, the device 200 may calculate a first value by applying a first weight to the first ratio. Here, the first weight may be set differently according to embodiments.

For example, when the first ratio is 80% and the first weight is set to 1.5, the device 200 may calculate 120 as the first value.

In step S410, the device 200 may set the first numerical value as the evaluation index of the first brand for the first period.

When the first numerical value is set as the evaluation index of the first brand, the device 200 may provide information on the first numerical value to the manager terminal 100 .

On the other hand, if it is confirmed in step S408 that the first ratio is smaller than the reference ratio, step S501 may be performed, and a detailed description related to this will be described later with reference to FIG. 5 .

5 is a flowchart illustrating a process of calculating a brand evaluation index according to another embodiment.

Referring to FIG. 5, first, in step S501, if the device 200 determines that the first ratio is smaller than the reference ratio, the 1-2 list is divided into the 1-2-1 list and the 1-2-2 list. can be distinguished by Here, the 1-2-1 list is a list of words having a positive meaning among words included in the 1-2 list, and the 1-2-2 list is a list of words included in the 1-2 list. This is a list of words with negative connotations.

For example, when the 1-2 list includes the first word, the second word, the third word, and the fourth word, the device 200 determines the meaning of each of the words included in the 1-2 list. As a result of the analysis, if the first word and the second word are identified as words having a positive meaning, a 1-2-1 list including the first word and the second word may be generated, and the third word and the second word may be generated. If the 4 words are identified as words having a negative meaning, a 1-2-2 list including the third word and the fourth word may be generated.

In step S502, the device 200 may calculate the second ratio through a value obtained by dividing the number of words in list 1-2-1 by the number of words in list 1-2. That is, the apparatus 200 may calculate the second ratio through a value obtained by dividing the number of words included in the 1-2-1 list by the number of words included in the 1-2 list.

In step S503, the device 200 may calculate the third ratio through a value obtained by dividing the number of words in list 1-2-2 by the number of words in list 1-2. That is, the apparatus 200 may calculate the third ratio through a value obtained by dividing the number of words included in the 1-2-2 list by the number of words included in the 1-2 list.

In step S504, the apparatus 200 may calculate a first value by applying a first weight to the first ratio.

In step S505, the device 200 may calculate a second value by applying a second weight to the second ratio. Here, the second weight may be set to a value smaller than the first weight.

For example, when the second ratio is 40%, the first weight is set to 1.5, and the second weight is set to 1, the device 200 may calculate 40 as the second value.

In step S506, the apparatus 200 may calculate a third value by applying a second weight to the third ratio.

In step S507, the device 200 may calculate a fourth numerical value by summing the first numerical value and the second numerical value and subtracting the third numerical value from the summed value.

In step S508, the device 200 may set the fourth value as an evaluation index of the first brand for the first period.

When the fourth value is set as the evaluation index of the first brand, the device 200 may provide information on the fourth value to the manager terminal 100 .

According to one embodiment, the device 200 may group a plurality of employees working for a company that operates a first brand and classify them into teams for carrying out a team project.

Specifically, the device 200 may group a plurality of employees who access the waiting space by identifying employee accounts that have access to the waiting space for conducting a team project. In this case, the device 200 may group a plurality of employees by using basic information of the employees. Here, the basic information of the employee may include information about the department name, position, age, work area, team project experience, and the like.

For example, the apparatus 200 may determine the age of each of the employees who are connected to the waiting space, and then classify employees having a similar age and designate them as a group. In addition, the device 200 may determine the working area of each of the employees who are connected to the waiting space, and then classify the employees working in the same area and designate them as a group. In addition, the apparatus 200 may determine the team project performance careers of each of the employees who are connected to the waiting space, and then classify employees having similar careers and designate them as a group. In addition, a plurality of employees may be grouped by classifying a plurality of employees who access the waiting space through various conditions.

When the apparatus 200 groups a plurality of employees and assigns the plurality of employees to each group, each group is set as a team for carrying out a team project, and each of the plurality of employees is classified into a team for carrying out a team project. can do.

For example, when a first employee, a second employee, a third employee, and a fourth employee are identified as employee accounts who access a waiting area for team projects, the device 200 provides the first employee and the second employee. as one group, designate the third employee and the fourth employee as one group, classify the group including the first employee and the second employee as the first team, and assign the third employee and the fourth employee to the first team. The group including the group may be classified as a second team.

Thereafter, the device 200 may assign a role to each employee classified as a team. At this time, the device 200 may assign a role to each employee in consideration of the employee's team project performance experience, preferred work, and the like.

For example, a team project is a voyage that requires reaching a destination in order to promote team unity, and 1st, 2nd, 3rd, 4th, 5th, and 6th employees are part of the 1st team. If classified, the device 200 designates the first employee as the captain to give the first employee the captain role, designates the second employee as the navigator to give the second employee the navigator role, and designates the third employee as the carrier , assigning the 3rd employee the role of carrier, designating the 4th employee as the helmsman to give the 4th employee the helmsman role, designating the 5th employee as the boatswain to give the 5th employee the role of the boatswain, 6 employees can be designated as the chef, giving the role of chef to the 6th employee.

After that, team members of the team who want to perform the team project may prepare to perform the team project. To this end, the device 200 may provide a virtual space for team project performance preparation.

For example, when a team project is a voyage task and the first team wants to perform a voyage task, the device 200 may provide a virtual space where items necessary for the voyage can be purchased before starting the voyage. At this time, in the virtual space, stores that sell items such as water, food, repair kits, life jackets, and sonar are placed in the village, so team members of the team who want to carry out the team project can shop at the store according to their assigned roles. By purchasing supplies, you can prepare for team project execution.

Thereafter, the device 200 may check whether the team project is ready to be performed. At this time, the device 200 may confirm that the team project preparation is complete when all team members belonging to the team to perform the team project complete preparation for the team project.

If it is confirmed that the team project is not ready to be carried out, members of the team who want to carry out the team project can continue to prepare for the team project.

When it is confirmed that the team project is ready to be carried out, members of the team who want to carry out the team project can carry out the team project. To this end, the device 200 may provide a virtual space for team project execution.

For example, if the team project is a sailing task and all members of the first team are ready for the sailing task, the device 200 sails the sailing ship of the first team in the virtual space and processes the sailing game to be executed. can do.

Thereafter, the device 200 may check whether a problem situation occurs while the first team is performing the team project.

For example, it is possible to check whether a problem situation occurs due to a storm or tsunami while the team project is a sailing task and the sailing ship of the first team is sailing.

When it is confirmed that the problem situation has occurred, the device 200 may acquire response information selected according to the problem situation for each employee. Here, the reaction information is information indicating a reaction selected by the employee in response to a problem situation.

For example, when a team project is a sailing task and a problem with a risk of sinking occurs due to a storm while the sailing ship of the first team is sailing, the device 200 transmits response information selected from each member of the first team. can be obtained In this case, the apparatus 200 may acquire reaction information for each employee by dividing roles assigned to each member of the first team.

For example, when the first employee's role is a captain and the first employee serving as a captain warns other team members of a dangerous situation as a reaction of the first employee performing the captain's role in case of danger of sinking, the device 200 determines the response of the first employee as the first employee. Obtained from the response information of the staff member, and if the second staff member's role is a navigator and the course of the sailing ship is changed with the response of the second staff member who acts as a navigator in case of danger of sinking, the response of the second staff member is determined by the response information of the second staff member. can be obtained with

Thereafter, the device 200 may check whether the team project performed by the first team is finished.

According to one embodiment, the team project may be terminated when all is performed to the end, and may be terminated when the state is changed to a state in which no further progress can be made while performing the team project. For example, if the team project is a sailing task, the sailing task may end when the destination is reached through sailing, and the sailing task may be terminated when the sailing ship sinks while sailing and becomes unavailable for further navigation. It can be.

If it is confirmed that the team project is not closed, members of the team who wish to carry out the team project can continue to carry out the team project.

When it is confirmed that the team project is finished, the apparatus 200 may evaluate whether the task of the role assigned to each employee has been performed well for each situation based on the response information obtained for each problem situation.

For example, when a team project for navigation work is performed in a first team, a first employee plays a role of a captain in the first team, and a second employee plays a role of a navigator in the first team, the device 200 performs a first team project. When a first problem situation occurs while the team is performing a team project, 1-1 reaction information generated as a response of a first employee serving as a captain is acquired to solve the first problem situation, and the first problem In order to solve the situation, first-second response information generated as a response of a second employee serving as a navigator may be obtained. Thereafter, when a second problem situation occurs while the first team is performing a team project, the device 200 generates a second problem situation generated as a response of a first employee serving as a captain to solve the second problem situation. Response information may be obtained, and 2-2 response information generated as a response of a second employee serving as a navigator may be obtained to solve the second problem situation.

Thereafter, when the team project being performed by the first team is completed, the apparatus 200 performs the task of the first employee who serves as the captain well by situation based on the 1-1 response information and the 2-1 response information. It is possible to evaluate whether or not the task of the second employee performing the role of the navigator was performed well in each situation based on the 1-2 response information and the 2-2 response information.

When evaluating whether the device 200 or the role assigned to each employee has been handled well for each situation, quantitative evaluation can be performed based on how quickly and accurately the employee responds based on a quantitative indicator set in advance according to the type of role.

The specific explanation for evaluating whether the role assigned to each employee was handled well in each situation is as follows.

First, the apparatus 200 may obtain conversation information of the first team input to the chat window while the team project is being performed when the team project being performed by the first team is finished. To this end, a chat window may be provided so that team members can exchange opinions while performing a team project.

Thereafter, the device 200 may extract conversation information of the first employee input by the first employee from conversation information of the first team.

For example, if the first employee and the second employee are a first team, the conversation information of the first team includes a conversation input by the first employee and a conversation input by the second employee, and the device 200 Conversation information of the first employee may be extracted by selecting only the conversation input by the first employee from the conversation information of the first team.

Thereafter, the device 200 may classify words having an active meaning in the conversation of the first employee based on the conversation information of the first employee.

For example, if it is confirmed that there is “let's change the route to the right” in the dialog of the first employee, the device 200 may “change the route”, “to the right”, “let's change” in the dialog of the first employee. Each word can be classified, and among the classified words, “let's change” can be classified as a word having an active meaning.

Thereafter, the device 200 may check whether a word having an active meaning exists in the conversation of the first employee within a reference time from the occurrence of the problem situation. Here, the reference time may be set differently according to embodiments.

For example, when the reference time is set to 10 seconds, the device 200 checks whether there is a word having an active meaning in the conversation of the first employee within 10 seconds after the first problem situation occurs, and 2 Within 10 seconds of the occurrence of the problem situation, it is possible to check whether there are words that have an active meaning in the conversation of the first employee.

Thereafter, the device 200 may calculate the number of problem situations in which a word having an active meaning appears in the conversation of the first employee within a reference time as the first number of times.

For example, while a team project is being performed, a first problem situation, a second problem situation, and a third problem situation occur, and the device 200 actively responds to the first employee's conversation within a standard time after the first problem situation occurs. It is confirmed that a word having an active meaning exists, it is confirmed that a word having an active meaning exists in the conversation of the first employee within a standard time after the occurrence of the second problem situation, and a standard time after the occurrence of the third problem situation If it is confirmed that there is no word with an active meaning in the conversation of the first employee within the first employee's conversation, the word with an active meaning appears only in the first problem situation and the second problem situation within the standard time, so the first number of times is doubled. can be calculated

Meanwhile, the device 200 may distinguish words having a passive meaning in the conversation of the first employee based on the conversation information of the first employee.

For example, if it is confirmed that there is “yes and then” in the conversation of the first employee, the device 200 may classify each word as “yes” and “therefore” in the conversation of the first employee, and the classification Among the words, “then” can be distinguished as a word with a passive meaning.

Thereafter, the device 200 may check whether a word having a passive meaning exists in the conversation of the first employee within a reference time from the occurrence of the problem situation.

For example, when the reference time is set to 10 seconds, the device 200 checks whether there is a word having a passive meaning in the conversation of the first employee within 10 seconds after the first problem situation occurs, and 2 Within 10 seconds of the occurrence of the problem situation, it is possible to check whether there are words with passive meaning in the conversation of the first employee.

Thereafter, the device 200 may calculate the number of problem situations in which a word having a passive meaning appeared within a reference time in the conversation of the first employee as a second number of times.

For example, while a team project is being performed, a first problem situation, a second problem situation, and a third problem situation occur, and the device 200 is passive in the conversation of the first employee within a standard time after the occurrence of the first problem situation. It is confirmed that a word having a negative meaning exists, it is confirmed that a word having a passive meaning exists in the conversation of the first employee within a standard time after the occurrence of the second problem situation, and a standard time after the occurrence of the third problem situation If it is confirmed that a word having a passive meaning exists in the conversation of the first employee within the first problem situation, the word having a passive meaning appears in the first problem situation, the second problem situation, and the third problem situation within the reference time. The number of times can be calculated 3 times.

Meanwhile, the device 200 may classify words having important meaning in resolving a problem situation from the conversation of the first employee based on the conversation information of the first employee. Here, a word having an important meaning in solving a problem situation may be set differently for each problem situation.

For example, if it is confirmed that there is “let's change the route to the right” in the dialog of the first employee, the device 200 may “change the route”, “to the right”, “let's change” in the dialog of the first employee. Each word can be classified, and “to the right” among the classified words can be classified as a word having an important meaning.

Thereafter, the device 200 may check whether a word having an important meaning exists in the conversation of the first employee within a reference time from the occurrence of the problem situation.

For example, when the reference time is set to 10 seconds, the device 200 determines whether there is a word having an important meaning in the first problem situation in the conversation of the first employee within 10 seconds after the first problem situation occurs. and, within 10 seconds after the occurrence of the second problem situation, it is possible to determine whether there is a word having an important meaning in the second problem situation in the conversation of the first employee.

Thereafter, the device 200 may calculate the number of problem situations in which a word having an important meaning in the conversation of the first employee appears within a reference time as a third number of times.

For example, while performing a team project, a first problem situation, a second problem situation, and a third problem situation occur, and the device 200 removes the first employee from the conversation within a standard time after the first problem situation occurs. 1 It is confirmed that a word having an important meaning in the problem situation exists, and it is confirmed that no word having an important meaning in the second problem situation exists in the conversation of the first employee within a standard time after the occurrence of the second problem situation; If it is confirmed that no word having an important meaning in the third problem situation exists in the conversation of the first employee within the standard time after the occurrence of the third problem situation, a word having an important meaning only in the first problem situation appears within the standard time. , the third number of times can be calculated as one.

When the first count, the second count, and the third count are respectively calculated, the device 200 evaluates the leadership of the first employee through the first count and evaluates the follow-up of the first employee through the second count; Through the third number of times, the crisis response ability of the first employee can be evaluated.

For example, when a total of 10 problem situations occur in performing a team project, the device 200 evaluates the leadership of the first employee as 80 points when the first number is 8, and the second number is 6. If the circuit is confirmed, the follow-up of the first employee is evaluated as 60 points, and if the third number is confirmed as 4 times, the crisis response ability of the first employee may be evaluated as 40 points.

Through this, the apparatus 200 may perform qualitative evaluation on leadership, follow-up, and crisis coping ability through natural language processing in employee conversations, quantify them for each item, and provide visualized evaluation results to employees.

6 is a diagram for explaining learning of a first artificial neural network according to an embodiment.

According to an embodiment, the first artificial neural network may be an algorithm that receives keyword information, analyzes users' perceptions of a brand, and outputs an analysis result. The device 200 in which the first artificial neural network is learned may be the same device as the device 200 that analyzes users' perceptions of a brand using the learned first artificial neural network, or may be a separate device. Hereinafter, a process of learning the first artificial neural network will be described.

First, in step S601, the device 200 may generate an input based on keyword information.

Specifically, the device 200 may perform a process of pre-processing keyword information. The preprocessed keyword information may be used as an input to the first artificial neural network, or an input may be generated through a normal process of removing unnecessary information.

In step S602, the device 200 may apply the input to the first artificial neural network. The first artificial neural network may be an artificial neural network trained according to reinforcement learning. The first artificial neural network may be a Q-Network, a Depp Q-Network (DQN), or a relational network (RL) structure suitable for outputting abstract inference through reinforcement learning.

The first artificial neural network learned through reinforcement learning may be updated and optimized by reflecting evaluations on various rewards. For example, the first reward may increase the reward value when a representative word associated with a keyword identified through keyword information is selected, and the second reward is rewarded when a keyword identified the most in keyword information is selected as a representative word. value may increase.

In step S603, the device 200 may obtain an output from the first artificial neural network. The output of the first artificial neural network is an analysis result of analyzing users' perceptions of the brand, and an output value indicating a state of users' perceptions may be output.

That is, the first artificial neural network selects a representative word through keyword information, analyzes the user's perception of the brand using the representative word, and outputs an analysis result representing the user's perception of the brand. .

In step S604, the device 200 may evaluate the output of the first artificial neural network and provide a reward. At this time, the evaluation of the output may be divided into first compensation, second compensation, and the like.

Specifically, the device 200 awards a first reward when a representative word associated with a keyword identified through keyword information is selected, and awards a second reward when a keyword identified the most in the keyword information is selected as a representative word. can do.

For example, if “organic” is included in the keyword information and the first artificial neural network selects “eco-friendly” associated with “organic” as a representative word and outputs it, the device 200 sends the first artificial neural network A first reward may be awarded.

In addition, when the keyword information includes “luxury” the most, the first artificial neural network selects “luxury”, which is the keyword most frequently identified in the keyword information, as a representative word, and outputs it. 1 A second reward may be awarded to the artificial neural network.

In step S605, the device 200 may update the first artificial neural network based on the evaluation.

Specifically, the device 200 is configured to maximize the expectation of the sum of reward values in an environment in which the first artificial neural network analyzes users' perceptions of the brand through keyword information, in a specific state. The first artificial neural network may be updated through a process of optimizing a policy for determining actions to be taken in states.

Meanwhile, the process of optimizing the policy may be performed through a process of estimating the maximum value of the expected value of the sum of rewards or the maximum value of the Q-function, or estimating the minimum value of the loss function of the Q-function. Estimation of the minimum value of the loss function may be performed through stochastic gradient descent (SGD) or the like. The process of optimizing the policy is not limited thereto, and various optimization algorithms used in reinforcement learning may be used.

The apparatus 200 may gradually update the first artificial neural network by repeating the learning process of the first artificial neural network as described above. Through this, the device 200 may train the first artificial neural network that outputs an analysis result of analyzing users' perceptions of the brand through keyword information.

That is, when analyzing users' perception of a brand through keyword information, the device 200 learns the first artificial neural network by adjusting the analysis criteria by reflecting reinforcement learning through the first reward and the second reward. can make it

7 is an exemplary diagram of a configuration of a device according to an embodiment.

Device 200 according to an embodiment includes a processor 210 and a memory 220 . The processor 210 may include at least one device described above with reference to FIGS. 1 to 6 or may perform at least one method described above with reference to FIGS. 1 to 6 . A person or organization using the device 200 may provide services related to some or all of the methods described above with reference to FIGS. 1 to 6 .

The memory 220 may store information related to the methods described above or may store a program in which the methods described below are implemented. Memory 220 may be volatile memory or non-volatile memory.

The processor 210 may execute a program and control the device 200 . Program codes executed by the processor 210 may be stored in the memory 220 . The device 200 may be connected to an external device (eg, a personal computer or network) through an input/output device (not shown) and exchange data through wired/wireless communication.

Apparatus 200 may be used to train an artificial neural network or to use a trained artificial neural network. The memory 220 may include a learning or learned artificial neural network. The processor 210 may learn or execute an artificial neural network algorithm stored in the memory 220 . The device 200 for learning the artificial neural network and the device 200 using the learned artificial neural network may be the same or may be separate.

The embodiments described above may be implemented as hardware components, software components, and/or a combination of hardware components and software components. For example, the devices, 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), programmable logic units (PLUs), microprocessors, or any other device capable of executing and responding to instructions. A processing device may run an operating system (OS) and one or more software applications running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of software. For convenience of understanding, there are cases in which one processing device is used, but those skilled in the art will understand that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that it can include. For example, a processing device may include a plurality of processors or a processor and a controller. Other processing configurations are also possible, such as parallel processors.

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 on a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiment or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. - includes hardware devices specially configured to store and execute program instructions, such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Software may include a computer program, code, instructions, or a combination of one or more of the foregoing, which configures a processing device to operate as desired or processes independently or collectively. The device can be commanded. Software and/or data may be any tangible machine, component, physical device, virtual equipment, computer storage medium or device, intended to be interpreted by or provide instructions or data to a processing device. , or may be permanently or temporarily embodied in a transmitted signal wave. Software may be distributed on networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer readable media.

As described above, although the embodiments have been described with limited drawings, those skilled in the art can apply various technical modifications and variations based on the above. For example, the described techniques may be performed in an order different from the method described, and/or components of the described system, structure, device, circuit, etc. may be combined or combined in a different form than the method described, or other components may be used. Or even if it is replaced or substituted by equivalents, appropriate results can be achieved.

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

Claims (3)

A method for analyzing a brand based on artificial intelligence, performed by a device,
Collecting posting data for a first brand among postings uploaded on users'SNS;
extracting first keyword information based on a hash tag of the posted material collected during the first period when the posting material is collected during a first preset period;
applying the first keyword information to a first artificial neural network to generate a first analysis result representing users' perceptions of the first brand during the first period based on an output of the first artificial neural network;
obtaining first brand information including information about a direction, target, and value that the first brand seeks;
comparing a first list that is a list of words included in the first analysis result with a second list that is a list of words included in the first brand information;
As a result of comparing the first list and the second list, words included in the first list are included in the 1-1 list, which is a list of words included in the second list, and words not included in the second list. Dividing into a first-second list that is a list of words;
Calculating a first ratio through a value obtained by dividing the number of words in the 1-1 list by the number of words in the first list;
checking whether the first ratio is greater than a preset reference ratio;
calculating a first value by applying a preset first weight to the first ratio when it is confirmed that the first ratio is greater than the reference ratio;
setting the first numerical value as an evaluation index of the first brand during the first period;
If it is determined that the first ratio is smaller than the reference ratio, the 1-2 list is divided into the 1-2-1 list, which is a list of words having a positive meaning, and the 1-2 list, which is a list of words having a negative meaning. -2 steps to classify into lists;
A second ratio is calculated by dividing the number of words in the 1-2-1 list by the number of words in the 1-2 list, and the number of words in the 1-2-2 list is calculated as the number of words in the 1-2 list. Calculating a third ratio through a value divided by the number of words in;
A first value is calculated by applying the first weight to the first ratio, a second value is calculated by applying a second weight set to a value smaller than the first weight to the second ratio, and the third ratio Calculating a third value by applying the second weight to ;
calculating a fourth value by subtracting the third value after summing the first value and the second value; and
Setting the fourth value as an evaluation index of the first brand during the first period,
AI-based brand analysis method. According to claim 1,
extracting second keyword information based on a hash tag of the posted material collected during the second period, when the posted material is collected during a second period after the first period;
applying the second keyword information to the first artificial neural network to generate a second analysis result representing users' perception of the first brand during the second period based on an output of the first artificial neural network; ; and
Comparing the first analysis result and the second analysis result to generate a third analysis result indicating a change in users' perception of the first brand,
AI-based brand analysis method. delete

Images