KR101045659B1 - System and Method for Emotional Information Service - Google Patents

Abstract

The present invention relates to a system for providing emotional information to a user, comprising: an input module for receiving emotional information scattered on a network such as the Internet, a processing module for processing meaningful emotional information by processing the emotional information input from the input module, and An output module for outputting the collected emotional information to service the user, the input module is a personal emotion search engine for searching the content expressing the personal emotion on the Internet, and the data for inferring the emotion of the objective society And an emotional metadata search engine, wherein the processing module is configured to include an emotional emotion DB in which emotional information of each individual is stored as a DB, and emotional metadata input from an individual emotional information and emotional metadata search engine stored in an individual emotion DB. Includes a group sentiment inference engine that infers and predicts group sentiment using Power modules include emotional information server to provide services based on the sensitivity to the user using the information of individual and collective sensibilities.

Description

Emotional Information Service System and Method {System and Method for Emotional Information Service}

The present invention relates to a system for servicing information on emotion, and more particularly, to an emotional information service system and method for collecting and inferring personal emotion and collective emotion to serve a user.

Emotions or feelings are very personal. Psychological factors that influence human emotions are largely divided into surprise, fear, disgust, anger, joy, happiness, sadness, and so on. However, emotions or feelings, and moreover, emotion-based information, have been information that remains in the tacit knowledge of individuals, and by moving and sharing them in the form of society, intimate and personalized services have not been developed in the context of emotion information. Therefore, specific and unbroadcasting of hidden emotions of individuals to specific groups, to specific individuals, to specific website members, to specific device holders, to emotion groups that share specific emotions, to people located in specific regions, to families, to society, and to nations The service model has been requested.

The emotions of individuals are collectively called emotions, and they can be called images at the social and national levels. The emotions collected through the interaction of individuals are transformed into images in various contexts with social and state institutions. Visualize trust sensitivity values, emotion scales, mood values, national image levels, and individual emotion distribution maps through storytelling-based emotion collection and filtering that individuals can actively participate in can do. In other words, it is possible to induce individual participation, share it on the platform, and open it to other services. As a result, service developers can obtain additional revenues such as advertisements and target marketing according to the emotional scale or emotional index value. .

The present invention has been invented in view of the above points, and an object thereof is to provide an emotional information service system and method for providing emotional information of individual and group specialized to a user.

The present invention for achieving the above object is an input module for receiving sentimental information scattered in a network, such as the Internet, a processing module for processing the sentiment information input from the input module to generate meaningful sentimental data, collected sentiment An output module for outputting information and serving a user, wherein the input module includes a personal emotion search engine for searching contents expressing personal emotions on the Internet, and an emotional metadata search for searching data for inferring an objective social emotion. It includes an engine, the processing module is a group sentiment using the sentiment metadata input from the individual sentiment DB and the sentiment metadata of the individual stored in the individual sentiment DB and the emotional sentiment metadata search engine DB Group emotion inference engine that infers and predicts the output module. Provides an emotional information service system including an emotional information service server that provides emotion-based services to users using information.

In the emotion information service system, the input module preferably further includes a user emotion input platform for the user to directly access the emotion information service system and input his emotions.

In the input module, when searching for personal emotion on the personal Internet site, the personal emotion search engine may input a known ID and password or log on with the OPEN-ID to search for relevant information. Also, in the input module, the emotional metadata search engine can objectively derive metadata by selecting data by keywords such as country, region, weather, time, and issues of major daily newspapers in XML file format and various tags on the Internet.

In the processing module, the unit of network composition for inferring group's emotion in group emotion inference engine is divided into node-hub-cluster, individual is represented as node in network structure, group is represented as hub connected node. , A cluster may be represented by a combination of nodes and hubs. In addition, it is preferable that the group emotion inference engine infers the emotion of the group based on various groups such as a specific region, gender, age, occupation, school, and company. It is desirable to infer the group's emotion by weighting metadata.

In the output module, the emotion information service server provides the emotion information of a specific user to other users in close relationship or broadcasts the emotion information to a group to which the specific user belongs. Alternatively, the emotion information service server provides the group emotion information input from the group emotion inference engine to users belonging to the group. In this case, it is preferable that the emotion information service server provides group emotion information based on the region where the user is located by using a map. The emotion information service server may provide the emotions of the group to the user according to various groups such as gender, age, occupation, school, company, etc. to which the user belongs. The emotion information service server may provide the user with the emotion change predicted for a specific group by using the emotion metadata.

The emotional information service method performed in the emotional information service system according to the present invention includes searching for and inputting personal emotions and emotional metadata, or directly inputting the user's emotions, and personal emotions and users found in the personal emotion search engine. Generating a personal emotion DB based on the user input emotion input on the emotion input platform, and inferring group emotion using the emotion metadata input from the personal emotion and emotion metadata search engine input from the personal emotion DB And, by the emotion information service server, providing a user with an individual emotion or a group emotion.

As described above, according to the present invention, an emotion-oriented platform can be implemented through a method and a technique for extracting and quantifying emotions. Furthermore, by moving emotion-based information from one's tacit information to a form of society and sharing it, it is possible to develop an emotion-based service that is intimate and personalized to customers in context.

Specifically, the storytelling-based emotion collection and filtering, which can extract emotions and actively participate, can provide a reliable emotional index value, emotion scale, mood value, and national image level. By visualizing the distribution of personal emotions, inducing individual participation, sharing it on the platform, and opening it to other services, service developers can earn additional revenues such as advertisements, and develop an emotional scale to become a standard technology scale. You can expect royalties.

Cyworld and other homepages and blogs remain stagnant due to the explosive growth in subscribers. This is due to the lack of a description that can provide the centripetal force to participate in the community. However, each site has a variety of sentimental information such as today's mood. It is possible to establish and revitalize a new community based on emotion by building a site that converges and re-spreads this.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the following embodiments are provided to those skilled in the art to fully understand the present invention, and may be modified in various forms, and the scope of the present invention is limited to the embodiments described below. It doesn't happen.

1 is a block diagram illustrating an emotional information service system according to a preferred embodiment of the present invention.

Referring to FIG. 1, an emotional information service system according to a preferred embodiment of the present invention includes an input module 100, a processing module 110, and an output module 120. The input module 100 includes an individual emotion search engine 102, an emotion metadata search engine 104, and a user emotion input platform 106, and the processing module 110 includes an individual emotion DB 112 and a group emotion inference. It includes an engine 114, the output module 120 includes an emotional information service server 122.

The input module 100 is a module for receiving emotional information scattered on a network such as the Internet. The personal emotion search engine 102 searches for content in which personal emotion is expressed on the Internet. That is, the personal emotion search engine 102 is an engine for searching the subjective individual's emotion, where the subjective individual's emotion or emotion refers to data that the user directly judges and inputs directly into a portal site or a personal site.

For example, the personality search engine 102 may express a feeling of intimacy through a cyworld widget expressing a mood, a Naver blog expressing the mood of the day, and a 'sympathetic view'. Search sentiment information in Daum planet). If these sites are not open for anyone to access, the personality search engine enters a known ID and password or logs on with OPEN-ID to search for relevant information. Alternatively, it may be possible to search only if the user of the site is allowed to collect personal emotion information.

It is preferable that each Internet site supports an open environment with various technologies such as OPEN-API. Each site includes a pre-arranged tag, which is an agreement between the emotional information service system, portals, blogs, cafes, planets, and mini homepages, and expresses and inserts specific emotional tags in the form of structural language (XML, etc.). This can make your search more efficient.

The emotion metadata search engine 104 is an engine for searching data for inferring the emotion of an objective society. That is, the emotion metadata search engine 104 collects information about group emotions related to the whole society, a specific group, and a specific region, not an individual's emotion.

For example, based on the newspaper article of the day, the story of an event focused on a specific region, the weather, and the like, the emotion information can be produced based on general common sense, such as 'the depression' of the entire nation's sentiment index becomes 'depressed'. . Therefore, the emotional metadata search engine 104 obtains information from related sites such as news and organizes it as metadata.

The emotional metadata search engine 104 may objectively derive metadata by selecting data by keywords such as country, region, weather, time, and issues of major daily newspapers in XML file format and various tags on the Internet. Metadata of social emotion calculation can be built into a table to continuously track the dynamic state of emotion.

The user emotion input platform 106 is for a user to directly connect to the emotion information service system and input his emotions. The personal emotion search engine 102 searches for personal emotions scattered on the Internet and provides an emotional information service to the individual. The personal emotional information input through the user emotional input platform 106 is more suitable for a specific individual. Can be provided.

For example, the user may input the emotion using a mobile phone linked with the emotion information service system. The mobile phone number plate is used as an emotional expression interface to give a representative emotional state to each number key. For example, in the joy mode of the phone, the 1 key can be very uplifted, the 2 key can be very joyful, and the 3 key can be a bit of joy.In the cell phone's sadness mode, the 1 key is very sad and the 2 key is Depressed sadness, the third key can mean that the heart is stuffy.

As such, when the user presses the first key in the joy mode of the mobile phone, a very exciting binary bit is wirelessly transmitted to the emotional information service system, and the transmitted binary bit is recorded very excited in the user emotional input platform 106.

Next, the processing module 110 processes the emotional information input from the input module 100 to generate meaningful emotional data. The processing module 110 includes an individual emotion DB 112 and a group emotion inference engine 114.

The personal emotion DB 112 stores the individual emotion information input from the personal emotion search engine 102 and the individual emotion information input through the user emotion input platform 106 into a DB. In this case, after collecting and quantifying data of the mood, intimacy, reputation, status, etc. of individuals distributed in each site, it may be expressed and stored as emotional information.

The group emotion inference engine 114 infers and predicts the group emotion using the collected emotion information of the individual. That is, core emotion index may be extracted by clustering emotions, and emotion metadata may be used as an initial value.

The actor unit of network configuration for inferring group emotion can be divided into node-hub-cluster. Here, individuals are represented as nodes in the network structure, and groups are represented as hubs to which nodes are connected. A cluster can be represented by a combination of nodes and hubs. Nodes, hubs, and clusters can vary dynamically depending on the strength, structure, and manner of connections on the network. In the emotional service, it is assumed that the state of emotion is transferred to the hub, cluster, or node by the interaction between nodes. Therefore, when certain emotions dominate at the node level, which is an individual level, the emotional states of specific hubs and clusters are changed. Alternatively, when the emotional state of a particular hub and cluster changes, the emotion of the node may change. In other words, it is a unit for classifying and analyzing this, and it can express and infer by introducing nodes, hubs, and clusters as network concepts.

For example, based on the region, the emotional state is added based on a range of a specific unit such as a radius of 500m or 1000m where User A is located or a building or street name where User A is located. At this time, the emotional state is information added on the map in the form of a vote of an unspecified person, not the identified emotion of user A. The greater the specific emotions of individuals, the more inferred the emotional state of the group on a particular road, building or area on the map.

As a specific example to infer the emotion of a group by using the emotional metadata, when it snows in a specific area, the corresponding emotional metadata is applied to 'lift' the emotion of the group, which is a set of individuals located in the area. Can be deduced by Or, when a murder of a woman occurs in a specific area, group emotions are inferred by automatically expressing emotions related to articles such as anxiety. When a festival is held at a particular school, it is inferred that the group located near the school or the student group of the school has emotions such as 'excitement' and 'expectation'.

Emotional reasoning of such groups is applicable not only to groups located in specific regions but also to groups such as gender, age, occupation, school, and company. In addition, it is applicable not only to geographic regions but also to groups of people located in specific buildings and on specific roads. In addition, the emotion of the group may be inferred by assigning weights to various metadata input from the emotion metadata search engine 104. For example, high weights may be given to events occurring in a specific area and low weights may be given to weather in the area.

Next, the output module 120 is a module for outputting the collected emotional information and serving the user, and includes an emotional information service server 122.

The emotion information service server 122 is a server that provides emotion-based services to users of the emotion information service system by using emotion information of individuals and groups. That is, the emotional information of the individual may be provided to others closely related to each other, or the emotional state of the group may be divided and sold for each network hub or cluster.

In the case of providing the personal emotional information to others, if the user A expresses the feeling of 'very joy' on his homepage, the personal emotion search engine 102 searches for this and sends it to the personal emotion DB 112. Record it. Alternatively, user A may directly input his feelings 'very joy' into user's emotion input platform 106 using his mobile phone.

Then, the emotional information service server 122 provides the information that the user A is 'very joy' to the user B, the user C, and the like, which the user A has registered as a friend. Alternatively, the user A's feelings of “very joy” can be broadcast to group A (having n members) set by user A. Using this, for example, emotional marketing can be performed on a site such as Cyworld (www.cyworld.com). In other words, by telling a friend of user A that there is a friend today, the friend may purchase content such as minihompy music, background, etc. as a gift box, and present it to user A.

Referring to the service for providing the emotional information of the group to the user, using the group emotional information input from the group emotion inference engine 114 provides the emotional information related to the user belonging to the group.

For example, whenever a user with a mobile terminal such as a mobile phone moves, the emotion map of the surrounding geography may be mashed up based on a meta tag of emotion calculation at time, longitude, and latitude. In other words, if demonstrations are conducted in a specific area, users located in the demonstration space can receive information such as anxiety and tension through their mobile phones. Therefore, if the user accidentally passes the protest site, it is possible to determine whether to pass the road if the representative emotional information of a specific area is obtained before the protest site passes.

The provision of collective emotion information based on these regions can be implemented using a map. That is, it is desirable to provide an emotional topographical map of the place where the person currently resides by using an emotional emoticon, an evaluation scale, a diagrammatic symbol, and an emotional color expression for each area scattered on physical space coordinates. Then, the user can easily visually recognize the emotional information about the area where he or she is staying or the area to be moved.

In particular, emotion metadata may be used to predict changes in emotion over time for a particular group or region. For example, it may be inferred and displayed the state of emotion changed for 24 hours for each road. In other words, during a rush hour for a particular road, heavy emotions may be predicted to predict emotion information, or emotion changes predicted based on event metadata such as a festival previously searched for each space may be displayed on a map.

In this way, in addition to providing services based on the region to the user by using the map, the group's feelings may be provided to the user according to various groups such as gender, age, occupation, school, and company. For example, if a reduction is carried out in a particular company, the sentiment of the group belonging to the company is 'depressed', so that the emotional information can be provided to the individual belonging to the group, and the emotional information service system is based on marketing based on the sentiment information of the group. can do.

Hereinafter, an emotional information service method performed in the emotional information service system will be described with reference to FIG. 2. 2 is a flowchart illustrating a emotion information service method according to an embodiment of the present invention.

First, an individual emotion and emotion metadata is searched and inputted, or a user directly inputs his or her emotion (S200). This step is performed in the personal emotion search engine, the emotion metadata search engine, and the user emotion input platform in the input module of FIG.

Subsequently, a personal emotion DB is generated based on the personal emotion retrieved from the personal emotion search engine and the user input emotion input from the user emotion input platform (S202).

Subsequently, the group emotion is inferred using the emotion sent from the personal emotion DB and the emotion metadata input from the emotion metadata search engine (S204). The unit of network composition to infer such group sentiment can be divided into node-hub-cluster, individual is represented by node in network structure, group is represented by hub connected node, and cluster is composed of node and hub. It can be expressed in combination. In addition, group emotion inference can infer the group's emotions based on various groups such as a specific region, gender, age, occupation, school, company, etc., and infer the group's emotions by weighting the emotion metadata. have.

Subsequently, the emotion information service server services the individual emotion or the group emotion to the user (S206). Such emotional information service can provide emotional information of a specific user to other users who are closely related to each other, or broadcast emotional information to a group to which a specific user belongs, or provide group emotional information to a user belonging to the group. It is possible. For example, the emotion information service server may provide group emotion information based on the region where the user is located by using a map, and the emotion of the group may be adjusted according to various groups such as gender, age, occupation, school, and company of the user. It can be provided to the user.

1 is a block diagram illustrating an emotional information service system according to a preferred embodiment of the present invention.

2 is a flowchart illustrating a emotion information service method according to an embodiment of the present invention.

Claims (13)

In the system for providing emotional information to the user, An input module for receiving sentiment information scattered over a network such as the Internet, a processing module for generating meaningful sentiment data by processing sentiment information input from the input module, and outputting the collected sentiment information and serving the user With modules, The input module includes a personal emotion search engine for searching contents expressing personal emotions on the Internet, and an emotional metadata search engine for searching data for inferring the objective social emotions. The processing module infers the emotion of the group by using the individual emotion DB in which the emotion information of each individual is made into DB and stored, the emotion information of the individual stored in the personal emotion DB, and the emotion metadata input from the emotion metadata search engine. And predictive group sentiment inference engines, The output module includes an emotion information service server that provides an emotion-based service to a user by using emotion information of individuals and groups, In the group emotion inference engine, the unit of network configuration for inferring group emotion is divided into node-hub-cluster, individual is represented by node in network structure, group is represented by hub connected node, and cluster is Represented by a combination of nodes and hubs, The group emotion inference engine infers the emotion of the group based on various groups of a specific region, gender, age, occupation, school or company, The group emotion inference engine infers group emotion by assigning weights to various metadata inputted from the emotion metadata search engine. The emotion information service system according to claim 1, wherein the input module further comprises a user emotion input platform for a user to directly access the emotion information service system and input his emotions. The personal emotion search engine of claim 1, wherein when searching for personal emotion on a personal Internet site, the user inputs a known ID and password, or logs on with an OPEN-ID to search for relevant information. Emotional information service system. The method of claim 1, wherein the emotional metadata search engine selects data by keyword, such as country, region, weather, time, issues of major daily newspapers, etc. in XML file format and various tags on the Internet. Emotional information service system characterized in that. delete delete delete The emotion information service system of claim 1, wherein the emotion information service server provides emotion information of a specific user to another user in close relation or broadcasts emotion information to a group to which the specific user belongs. The emotion information service system of claim 1, wherein the emotion information service server provides group emotion information input from the group emotion inference engine to a user belonging to the group. The emotion information service system of claim 9, wherein the emotion information service server provides group emotion information based on a region where a user is located by using a map. The emotion information service system of claim 9, wherein the emotion information service server provides the emotions of the group to the user according to various groups such as gender, age, occupation, school, company, etc. to which the user belongs. The emotion information service system of claim 10 or 11, wherein the emotion information service server provides a user with a change of emotion predicted for a specific group by using emotion metadata. delete

Images