WO2019039659A1 - Emotion-based user management method and apparatuses performing same - Google Patents

Abstract

An emotion-based user management method and apparatuses for performing the same are disclosed. The emotion-based user management method, according to one embodiment, comprises the steps of: estimating a mood state of a user on the basis of learning data for an image tagged with emotion values, a sentiment lexicon, and content generated by the user; and managing the user by determining a state of the user on the basis of the mood state.

Description

Sensibility based user management method and apparatuses performing the same

The following embodiments relate to a method of managing a user based on emotion and apparatuses for performing the same.

The emotion of the user is represented not only by the voice but also through various contents. The user's emotional information can be obtained in various ways. For example, the user emotion information varies depending on various applications.

Emotion recognition technology can extract basic emotions of humans through complex signals obtained from human and environment in which human beings live, and can give ability to intelligently respond to human emotions.

In recent years, emotion recognition technology has been utilized in the field of human interface technology.

Embodiments can provide a content analysis technique with improved performance through not only image but also text segmentation, sentence structure, and morphology and emotion estimation for emoticons.

In addition, embodiments may provide a technique that allows a user to monitor the status of a user by providing the user with information about the status of the user.

Embodiments can also provide techniques for diagnosing and managing depression by providing customized user management for the user who created the content.

The emotion-based user management method according to an exemplary embodiment estimates the emotional state of the user based on learning data for a tagged image, sentiment lexicon, and contents generated by a user, And managing the user by determining the state of the user based on the emotional state.

The content may include at least one of the image, text, and emoticon of the user.

Estimating a first emotion state corresponding to the image using the learning data; estimating a second emotion state corresponding to the text and the emoticon using the emotion dictionary; Generating the integrated emotion state corresponding to the emotion state of the user by integrating the first emotion state, the first emotion state, and the second emotion state.

Wherein the step of estimating the first emotion state comprises: generating the learning data based on an emotion value for an arbitrary image; and selecting one of the deep learning and the probabilistic latent semantic analysis learning And estimating the first emotion state by analyzing a correlation between the characteristic of emotion for the image and the learning data using the first emotion state.

Wherein the generating step comprises the steps of: acquiring the arbitrary image; determining win / lose of the arbitrary image with respect to a specific emotion based on a response of an arbitrary user; And generating the learning data by tagging the normalized emotion value to the arbitrary image by normalizing the emotion value.

Wherein the step of estimating the second emotional state includes the steps of generating the emotional dictionary based on a feature of emotion of any text, a structure of sentences, and a feature of emotion of a form and an emoticon of an arbitrary emoticon, And estimating the second emotional state by analyzing a feature of emotion of the text and a feature of emotion of the emoticon using a dictionary.

Wherein the managing comprises: determining the state of the user based on the integrated emotional state and the social networking service activity propensity of the user; and determining a state of the user based on the state of the user Grouping the user into a group, and managing the user by providing information about the user.

Wherein the determining comprises: determining a first state of the user based on the integrated emotional state; determining a second state of the user based on the SNS activity propensity; And determining a state of the user by integrating the second state.

Determining the first state may include monitoring the integrated emotional state and determining a first state of the user by calculating a statistical value for the integrated emotional state based on the monitoring result have.

The group may be any one of a normal group and a depressed group.

The information may include at least one of a graph of the emotional state of the user and a feature value of the group.

The feature value of the group may include at least one of a word, a sentence, and a color of an image frequently used in the group.

The emotion-based user management apparatus according to an exemplary embodiment of the present invention includes a transmission / reception unit that receives a content generated by a user, a learning unit that generates learning data for a tagged emotion value, a sentiment lexicon, And a controller for managing the user by estimating a mood state of the user and determining the state of the user based on the emotion state.

The content may include at least one of the image, text, and emoticon of the user.

Wherein the controller estimates a first emotion state corresponding to the image using the learning data, estimates a second emotion state corresponding to the text and the emoticon using the emotion dictionary, And an estimator for generating an integrated emotion state by integrating the second emotion state.

Wherein the estimator generates the learning data based on the emotion value for an arbitrary image and uses the depth learning and the probabilistic latent semantic analysis learning The first emotion state can be estimated by analyzing the correlation between the feature and the learning data.

The estimation unit acquires the arbitrary image, determines the win or loss of the arbitrary image with respect to the specific emotion based on a response of an arbitrary user, and determines an emotion value for the arbitrary image based on the determination result , The learning data can be generated by normalizing the emotion value and tagging the normalized emotion value to the arbitrary image.

Wherein the estimating unit generates the emotion dictionary based on a feature of a part of text, a structure of a sentence, a feature of emotion for a form, and a feature of emotion for an arbitrary emoticon, And the second emotional state can be estimated by analyzing the characteristics of the emotional state and the characteristics of the emotional state.

Wherein the controller comprises: a determination unit for determining the state of the user based on the integrated emotion state and the social networking service activity tendency of the user; and a group corresponding to the state of the user based on the state of the user And a management unit for classifying the user and managing the user by providing information about the user.

Wherein the determination unit determines the first state of the user based on the integrated emotion state, determines a second state of the user based on the SNS activity propensity, and integrates the first state and the second state The state of the user can be determined.

The determining unit may determine the first state of the user by monitoring the integrated state of emotion and calculating a statistical value for the integrated state of emotion based on the monitoring result.

The group may be any one of a normal group and a depressed group.

The information may include at least one of a graph of the emotional state of the user and a feature value of the group.

The feature value of the group may include at least one of a word, a sentence, and a color of an image frequently used in the group.

1 shows a schematic block diagram of a user management system according to one embodiment.

Fig. 2 shows a schematic block diagram of the user management apparatus shown in Fig. 1. Fig.

Fig. 3 shows an example for explaining learning data.

Fig. 4 shows an example for explaining the correlation of learning data.

Fig. 5 shows an example for explaining the integrated emotion state.

6A and 6B show an example for explaining a graph of the emotional state of the user.

FIG. 7 shows a flowchart for explaining the operation of the user management apparatus shown in FIG. 1. FIG.

It is to be understood that the specific structural or functional descriptions of embodiments of the present invention disclosed herein are presented for the purpose of describing embodiments only in accordance with the concepts of the present invention, May be embodied in various forms and are not limited to the embodiments described herein.

Embodiments in accordance with the concepts of the present invention are capable of various modifications and may take various forms, so that the embodiments are illustrated in the drawings and described in detail herein. However, it is not intended to limit the embodiments according to the concepts of the present invention to the specific disclosure forms, but includes changes, equivalents, or alternatives falling within the spirit and scope of the present invention.

The terms first, second, or the like may be used to describe various elements, but the elements should not be limited by the terms. The terms may be named for the purpose of distinguishing one element from another, for example without departing from the scope of the right according to the concept of the present invention, the first element being referred to as the second element, Similarly, the second component may also be referred to as the first component.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between. Expressions that describe the relationship between components, such as "between" and "between" or "neighboring to" and "directly adjacent to" should be interpreted as well.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. It is to be understood that in the present context, terms such as "include" or "having" are intended to designate the presence of stated features, integers, steps, operations, elements, parts, or combinations thereof, , Steps, operations, components, parts, or combinations thereof, as a matter of principle.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the meaning of the context in the relevant art and, unless explicitly defined herein, are to be interpreted as ideal or overly formal Do not.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, the scope of the patent application is not limited or limited by these embodiments. Like reference symbols in the drawings denote like elements.

1 shows a schematic block diagram of a user management system according to one embodiment.

1, a user managing system 10 includes a user device 100, a user managing apparatus 200, and a content storage 300, .

The user device 100 may be implemented as an electronic device. For example, the electronic device may be implemented as a personal computer (PC), a data server, or a portable electronic device.

The portable electronic device may be a laptop computer, a mobile phone, a smart phone, a tablet PC, a mobile internet device (MID), a personal digital assistant (PDA), an enterprise digital assistant A digital still camera, a digital video camera, a portable multimedia player (PMP), a personal navigation device or a portable navigation device (PND), a handheld game console, an e-book, or a smart device. At this time, the smart device can be implemented as a smart watch or a smart band.

The user device 100 may communicate with at least one of the user management device 200 and the content repository 300.

For example, the user device 100 may be connected to the user management device 200 through various communication networks such as an Internet communication network, an intranet, a local area network (LAN), a wireless LAN, Wi-Fi, LF, Xbee, Zigbee, Blue- And the content repository 300, as shown in FIG.

The user device 100 may transmit the content created by the user to at least one of the user management apparatus 200 and the content repository 300 based on the user response. For example, the user device 100 may be provided with an application program (APP) for registering and managing content created by a user, and may be linked to the user management apparatus 200 and the content repository 300. The application program (APP) may be a program associated with a social networking service (SNS).

The user device 100 can receive and visualize information on the status of the user from either the user management device 200 or the content storage 300. For example, the user device 100 may visualize information about the user transmitted by the user management device 200 to allow the user to monitor the status of the user through an application program (APP).

The user management apparatus 200 can provide a customized user management according to the state of the user by estimating (or recognizing) the emotional state of the user through the content created by the user and determining the state of the user. The user management apparatus 200 may be a user management apparatus based on emotion.

For example, the user management apparatus 200 may continuously monitor the user's image, text, and emoticons to determine the user's state by estimating (or recognizing) the emotional state of the user with respect to the user-generated content. At this time, the user management apparatus 200 can estimate the emotional state of the user through the characteristics of the emotion of the user's image, the parts of the text of the user, the structure of the sentence, and the emotional characteristics of the form.

Accordingly, the user management apparatus 200 can provide customized user management for the user who generated the content by classifying the user according to the status of the user and providing information on the status of the user.

That is, the user management apparatus 200 can provide a content analysis technique with improved performance through not only the image but also the parts of speech, the structure of sentences, and the emotion estimation of the form and the emoticon. For example, improved content analysis techniques can be used to investigate preferences for specific politicians, polls on specific issues, reviews on certain products, and preferences.

In addition, the user management apparatus 200 may provide the user with information on the status of the user, thereby enabling the user to monitor the status of the user.

In addition, the user management apparatus 200 can diagnose and manage depression by providing customized user management to the user who created the content.

The content store 300 may store various contents. For example, the content repository 300 may store content created by the user, content created by any user, and information transmitted by the user management device 200. [

The content storage 300 may include a data base 310, a social network service server 330, and the Internet 350.

For convenience of explanation, the content storage 300 includes the database 310, the SNS server 330, and the Internet 350, but is not limited thereto. For example, the content repository 300 may include various repositories that store various content that the user management device 200 acquires.

Fig. 2 shows a schematic block diagram of the user management apparatus shown in Fig. 1. Fig. 3 shows an example for explaining learning data, Fig. 4 shows an example for explaining the correlation of learning data, FIG. 5 shows an example for explaining the integrated emotional state, and FIGS. 6A and 6B show an example for explaining a graph of the emotional state of the user.

Referring to FIGS. 2 to 6B, the user management apparatus 200 includes a transceiver 210 and a controller 230.

The transceiver 210 may receive the signal (or data) transmitted from the user device 100 and the content storage 300 and may transmit the received signal (or data) to the controller 230. For example, the signal may be user generated content, and any user generated content. The user-generated content may be at least one of a user's image, text, and emoticon. The content created by an arbitrary user may be at least one of video, text, and emoticons of an arbitrary user.

The transceiver 210 may transmit the signal (or data) transmitted from the controller 230 to at least one of the user device 100 and the content repository 300.

The controller 230 can control the overall operation of the user management apparatus 200. [ For example, the controller 230 may control the operation of the transceiver 210.

The controller 230 may estimate (or recognize) the mood state of the user based on the learning data for the tagged image, the sentiment lexicon, and the content generated by the user.

The controller 230 can manage the user by determining the state of the user based on the emotional state.

The controller 230 includes an estimator 231, a determiner 233, and a manager 237.

The estimating unit 231 estimates (or recognizes) the first emotion state corresponding to the user's image using the learning data, estimates the second emotion state corresponding to the user's text and emoticon using the emotion dictionary, (Or recognizes) the first emotion state, and integrate the first emotion state and the second emotion state to generate an integrated emotion state.

In order to estimate (or recognize) the first emotional state, the estimation unit 231 can generate learning data based on emotion values for an arbitrary image. An arbitrary image may be an arbitrary user's image.

In order to generate the learning data, the estimation unit 231 acquires an arbitrary image and can determine the win or loss of an arbitrary image for a specific emotion based on a response of an arbitrary user. At this time, the estimator 231 can determine the win or loss of an arbitrary image by using an online rating game based on crowd sourcing or pair-wise competition.

First, the estimation unit 231 may provide an arbitrary image to an arbitrary user using crow sourcing and an online rating game based on fair-wise competition, so that an arbitrary user can select an arbitrary image. An arbitrary image may be two images randomly selected differently. Any selected image may have a higher relevance to a particular emotion. The estimator 231 can repeat an arbitrary pair of images 100 times.

Thereafter, the estimation unit 231 can determine and estimate (or recognize) the win or loss of an arbitrary image based on the selection result. For example, the estimation unit 231 may determine that an arbitrary image selected is a positive value. The estimation unit 231 can estimate (or recognize) the win / loss of an arbitrary image based on the determined win / loss. The estimation unit 231 can estimate (or recognize) the win / loss of the image A and the image C based on the win / loss of the image A and the image B and the win / loss of the image B and the image C. Image A, image B, and image C may be different arbitrary images.

The estimation unit 231 can determine the emotion value for an arbitrary image based on the determination result. For example, the estimator 231 may determine emotion values for an arbitrary image using visual link analysis, a three-column method, and a page rank. The emotion value may be a ground truth value.

First, the estimation unit 231 may generate a visual-link graph based on the determination result using the visual-link analysis. For example, the estimator 231 may arrange an arbitrary image with a large multiplier in the head of the visual link graph. The estimating unit 231 can arrange an arbitrary image having a large number of eyes on the tail of the visual link graph.

Thereafter, the estimation unit 231 can determine the ranking for an arbitrary image using the visual link graph. For example, the estimator 231 may use a page rank to determine a high ranking for any image disposed at the head of the visual link graph. The estimator 231 may generate an inferred matrix as in the algorithm of FIG. 3 using a three-column method, and determine a ranking for an arbitrary image. The estimation unit 231 can determine an emotion value of an arbitrary image according to the rank of an arbitrary image.

The estimator 231 can generate the learning data by normalizing the emotion value for an arbitrary image and tagging the normalized emotion value to an arbitrary image. For example, the estimation unit 231 can generate learning data using the algorithm of FIG. The normalized emotion value may be a value between -1 and 1. A value close to 1 corresponds to the winner image and a value close to -1 may correspond to the loser image.

The estimating unit 231 compares the correlation between the characteristic of emotion with respect to the user's image and the learning data using any one of deep learning and probabilistic latent semantic analysis learning, Estimation (or recognition).

When PLSA learning is used, the estimating unit 231 can generate correlation of learning data such as GRAPH1 in FIG. 4 through an expectation-maximization (EM) algorithm. For example, GRAPH1 can be a graph showing a correlation of 1170 color combinations with a particular emotion. If there is a high correlation between the color combination and a certain emotion, GRAPH1 may be close to white. Certain emotions are pretty, colorful, dynamic, gorgeous, wild, romantic, natural, graceful, quiet, classical, May be at least one of classic, dandy, majestic, cool, pure, and modern.

The estimator 231 may estimate (or recognize) the first emotion state by extracting visual features of the user's image and comparing the correlation of the learning data. For example, the visual feature may be at least one of color compositions, and type information based on scale invariant feature transform (SIFT).

The estimator 231 can estimate (or recognize) the first emotion state using Equation (1).

는 i번째 사용자의 영상에 j번째 감성이 추정(또는 인식)될 결합 확률이고,

는 i번째 사용자의 영상이 입력될 확률이고,

는 i번째 사용자의 영상에 j번째 감성이 있을 조건부 확률이고,

는 사용자의 영상으로부터 추출이 가능한 k번째 비주얼 특징이고,

는 k번째 비주얼 특징이 j번째 감성에 매핑될 확률이고,

는 i번째 사용자의 영상에 k번째 비주얼 특징이 나타날 확률을 의미한다.here,

Is the joint probability that the jth emotion is estimated (or recognized) in the image of the ith user,

Is the probability that an image of the i < th > user is input,

Is the conditional probability that there is j-th emotion in the image of the i-th user,

Is a k-th visual feature that can be extracted from the user's image,

Is the probability that the kth visual feature is mapped to the jth emotion,

Is the probability that the kth visual feature appears in the image of the ith user.

In order to estimate (or recognize) the second emotional state, the estimator 231 calculates the emotional state of the emotional dictionary 231 based on the feature of emotion of any text, the structure of the sentence, Can be generated. Any text, and any emoticon, can be any user's text, and emoticons.

For example, the estimator 231 may generate an emotion dictionary using visual sentiment ontology (VSO), sentistrength, and wikipedia. The estimator 231 can learn the emotion dictionary using a support vector machine (SVM). The learned emotional dictionaries are composed of emotional vocabularies included in interrogative, negative, exclamation, punctuation, adjectives, nouns, verbs, and adverbs. The correlation between the features of the base and the parts of arbitrary text, the structure of the sentences, and the features of the emotions on the form can be learned emotional dictionaries.

The estimating unit 231 can estimate (or recognize) the second emotional state by analyzing the characteristics of the emotion of the user with respect to the text and the emotional characteristics of the emoticon using the emotional dictionary. For example, the estimating unit 231 can extract characteristics of emotion of a user's emoticon of the part of speech, structure of a sentence, and type of a text divided by a sentence, and emotional characteristics of the emoticon of the user. The estimator 231 can estimate (or recognize) the second emotion state by comparing and analyzing the extracted emotion feature and the emotion dictionary.

In order to integrate the first emotion state and the second emotion state, the estimator 231 may assign a weight to the first emotion state and the second emotion state using a harmonic analysis technique. For example, the estimator 231 may assign a weight to the first emotion state to be higher than a weight to the second emotion state. Harmonic analysis techniques may be at least one of rule-based aggregation, and learning-based aggregation techniques. The optimum weight values of the first emotion state and the second emotion state can be given through experiments.

The determination unit 233 can determine the status of the user based on the integrated emotion state and the social networking service activity tendency of the user.

For example, the determination unit 233 may determine the first state of the user based on the integrated emotion state, determine the second state of the user based on the SNS activity propensity, and integrate the first state and the second state To determine the state of the user. The user's state can be either a negative state, a positive state, and a normal state.

In order to determine the first state, the determination unit 233 may monitor the integrated emotion state. For example, the determination unit 233 may monitor the integrated emotional state of the user's image, text, and emoticon for a predetermined period of time. The period may be various periods such as a day, a week, and a month.

The determination unit 233 can determine the first state of the user by calculating a statistical value for the integrated emotion state based on the monitoring result. For example, the determination unit 233 may calculate a statistical value for the integrated emotion state using the algorithm shown in FIG.

In order to determine the second state, the determination unit 233 may extract either the user's SNS activity propensity and the user's posting habits through the information associated with the user-generated content. The information may include at least one of a content volume, a content length, a content word count, a social engagement in the SNS, and a content upload time.

The determining unit 233 can determine the second state of the user based on any one of the extracted user's SNS activity propensity and the user's posting habits.

The management unit 235 classifies users into groups corresponding to the status of the user based on the status of the user, and provides information on the users to manage the users.

For example, the management unit 235 may classify users into groups corresponding to the status of the user using Bayesian, clustering, and a convolutional neural network (CNN). The group can be either a normal group, or a depressed group. A normal group may be a group that is classified into a normal state, and a user in a positive state. A depressed group can be a grouped group of users who are in a negative state.

The management unit 235 can provide information on the user to the user device 100 and the content repository 300 to manage the user. For example, the management unit 235 may provide information on the user classified into the depression group to the user device 100 and the content repository 300 to manage the user in a negative state.

For example, the information may include at least one of a graph of the emotional state of the user, and a feature value of the group. A graph of the emotional state of the user may be as shown in Figs. 5A and 5B. GRAPH2, GRAPH3, and GRAPH4 in FIG. 6A may be graphs of the emotional state of the user classified in the depression group. GRAPH5, GRAPH6, and GRAPH7 in FIG. 6B may be graphs of the emotional state of the user classified into the normal group. 6A and 6B may be the date, and the Y axis may be the emotion value. The feature value of the group may include at least one of the words, sentences, and color of the image frequently used in the group.

FIG. 7 shows a flowchart for explaining the operation of the user management apparatus shown in FIG. 1. FIG.

Referring to FIG. 7, the transmitter / receiver 210 may receive user-generated content from the user device 100 and the content repository 300 (S710).

The estimator 231 estimates and generates the first emotion state, the second emotion state, and the integrated emotion state of the user on the basis of the learning data for the tagged emotion value, the emotion dictionary, and the contents generated by the user (S730).

For example, the estimation unit 231 can estimate the first emotion state of the user based on the learning data and the user's image. The estimating unit 231 can estimate the second emotional state of the user based on the emotional dictionary and the user's text and emoticons. The estimator 231 can generate the integrated emotion state by integrating the first emotion state and the second emotion state.

The determining unit 233 may determine the state of the user based on the integrated emotion state, the user's SNS activity propensity, and the user's posting habit (S750).

For example, the determination unit 233 can determine the first state of the user based on the statistical value of the integrated emotion state. The determination unit 233 can determine the second state of the user based on the user's SNS activity propensity extracted from the information associated with the user-generated content, and the posting habit of the user. The determination unit 233 may determine the state of the user by integrating the first state and the second state.

The management unit 233 can classify users into groups corresponding to the status of the user based on the status of the user and provide information on the users to the user device 100 and the content storage 300 to manage the users (S770).

For example, the management unit 233 may classify a user who is in a negative state into a depression group, and provide information of the user in a negative state to the user device 100 and the content storage 300 to manage the user.

The apparatus described above may be implemented as a hardware component, a software component, and / or a combination of hardware components and software components. For example, the apparatus and components described in the embodiments may be implemented within a computer system, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA) , A programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For ease of understanding, the processing apparatus may be described as being used singly, but those skilled in the art will recognize that the processing apparatus may have a plurality of processing elements and / As shown in FIG. For example, the processing unit may comprise a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of the foregoing, and may be configured to configure the processing device to operate as desired or to process it collectively or collectively Device can be commanded. The software and / or data may be in the form of any type of machine, component, physical device, virtual equipment, computer storage media, or device , Or may be permanently or temporarily embodied in a transmitted signal wave. The software may be distributed over a networked computer system and stored or executed in a distributed manner. The software and data may be stored on one or more computer readable recording media.

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

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

Claims (20)

Estimating a mood state of the user based on learning data for a tagged sensibility value, a sentiment lexicon, and a user-generated content; And Managing the user by determining the state of the user based on the emotional state Lt; / RTI > The content may be, And at least one of image, text, and emoticon of the user. The method according to claim 1, Wherein the estimating step comprises: Estimating a first emotion state corresponding to the image using the learning data; Estimating a second emotion state corresponding to the text and the emoticon using the emotion dictionary; And Generating an integrated emotion state corresponding to the emotion state of the user by integrating the first emotion state and the second emotion state Based user management method. 3. The method of claim 2, Wherein the step of estimating the first emotion state comprises: Generating the learning data based on an emotion value for an arbitrary image; And Estimating the first emotional state by analyzing a correlation between the characteristic of emotion for the image and the learning data using one of the deep learning and the probabilistic latent semantic analysis learning Based user management method. The method of claim 3, Wherein the generating comprises: Acquiring the arbitrary image; Determining success or failure of the arbitrary image with respect to a specific emotion based on a response of an arbitrary user; Determining an emotion value for the arbitrary image based on the determination result; And Generating the learning data by normalizing the emotion value and tagging the normalized emotion value to the arbitrary image Based user management method. 3. The method of claim 2, The step of estimating the second emotional state includes: Generating the emotional dictionary based on a feature of emotion of a certain emoticon and a feature of emotion of a part of arbitrary text, a structure of a sentence, and a form; And Estimating the second emotional state by analyzing a feature of emotion of the text and a feature of emotion of the emoticon using the emotional dictionary, Based user management method. 3. The method of claim 2, Wherein the managing comprises: Determining the state of the user based on the integrated emotional state and the social networking service activity propensity of the user; Classifying the user into a group corresponding to the status of the user based on the status of the user; And Managing the user by providing information about the user Based user management method. The method according to claim 6, Wherein the determining comprises: Determining a first state of the user based on the integrated emotional state; Determining a second state of the user based on the SNS activity propensity; And Determining a state of the user by integrating the first state and the second state, Based user management method. 8. The method of claim 7, Wherein determining the first state comprises: Monitoring the integrated emotional state; And Determining a first state of the user by calculating a statistical value for the integrated emotional state based on the monitoring result Based user management method. The method according to claim 6, Wherein the group comprises: Normal group, and depressive group, The information includes: A graph of the emotional state of the user, and a feature value of the group. 10. The method of claim 9, The feature value of the group may include, And at least one of a word, a sentence, and a color of an image frequently used in the group. A transmission / reception unit for receiving contents generated by a user; And A sentiment lexicon, and a mood state of the user based on the content, and determining the state of the user based on the emotion state, A controller Lt; / RTI > The content may be, And at least one of image, text, and emoticon of the user. 12. The method of claim 11, The controller comprising: Estimating a first emotion state corresponding to the image using the learning data, estimating a second emotion state corresponding to the text and the emoticon using the emotion dictionary, And a second emotional state is integrated to generate an integrated emotional state, Based user management device. 13. The method of claim 12, Wherein the estimating unit comprises: Generating learning data based on an emotion value for an arbitrary image, and using the deep learning and the probabilistic latent semantic analysis learning (PLSA) And estimating the first emotional state by analyzing the correlation of the learning data. 14. The method of claim 13, Wherein the estimating unit comprises: Determining an emotion value for the arbitrary image on the basis of the determination result, and determining the emotion value of the sensed image based on the determination result, Wherein the learning data is generated by tagging the normalized emotion value to the arbitrary image. 13. The method of claim 12, Wherein the estimating unit comprises: A method of generating emotional dictionaries, the method comprising: generating the emotional dictionaries based on a feature of an arbitrary text part of speech, a structure of a sentence, and a feature of emotion for a form and a feature of emotion for an arbitrary emoticon; And estimating the second emotional state by analyzing characteristics of the emotional state with respect to the emoticon. 13. The method of claim 12, The controller comprising: A determination unit for determining the state of the user based on the integrated emotion state and the social networking service activity tendency of the user; And A management unit for classifying the user into a group corresponding to the status of the user based on the status of the user and managing the user by providing information about the user, Based user management device. 17. The method of claim 16, Wherein, Determine a first state of the user based on the integrated emotional state, determine a second state of the user based on the SNS activity propensity, and integrate the first state and the second state to determine Based user management device that determines the status of the user. 18. The method of claim 17, Wherein, Wherein the first state of the user is determined by monitoring the integrated emotional state and calculating a statistical value for the integrated emotional state based on the monitoring result. 17. The method of claim 16, Wherein the group comprises: Normal group, and depressive group, The information includes: A graph of the emotional state of the user, and a feature value of the group. 20. The method of claim 19, The feature value of the group may include, And at least one of a word, a sentence, and a color of an image frequently used in the group.

Images