KR20110039733A - Apparatus and method for managing personal life - Google Patents

Abstract

PURPOSE: An apparatus and method for managing personal life are provided to easily store memories and experiences of a user to a database by visually expressing metadata. CONSTITUTION: A data collector(110) collects context information of a user through a mobile device. A behavioral pattern inferring unit(120) infers the behavioral pattern of a user by applying a probability model as the context information. A tagging UI unit(130) forms the context information and behavioral pattern information into a visualized interface. The tagging UI unit receives the inputted tagging information from the user. A metadata generator analyzes the context information, the behavioral pattern information, and the tagging information, and generates a metadata with the analyzed information.

Description

Apparatus and Method for Managing Personal Life}

The present invention relates to a device and method for managing an individual's daily life, and more particularly, to collect information related to a user's life through a mobile device, and to apply the collected information to a probability model based on behavior theory. An apparatus and method for inferring a user's behavior or emotion and tagging mobile content (photos, videos, etc.) based on the inferred information.

Mobile devices such as digital cameras and mobile phones can collect various information such as call logs, photo shoots, music file playback, location information, and the like, and can carry the user's daily information effectively because the user always carries it. In addition, since the mobile device is a strong personality equipment, it can be adapted and specialized according to the preferences or preferences of the individual. As such, if the information collected through the mobile device is effectively used, more various services may be provided to users who use the mobile devices as life recorders. As a part of such technology, a technology for collecting information about a user's daily life through a mobile device and modeling a user's behavior pattern based on the collected information has been studied. However, the technology of searching for personalized information to the user according to the user's query based on the information collected through the mobile device is not dealt with in detail.

In addition, due to the recent improvement in mobile device performance and the generalization of mobile phone camera functions, the ability to create content on mobile devices has greatly increased, but the search and management of multimedia content such as photos and videos created on mobile devices has greatly increased. It is not developing. For this purpose, a method of using image analysis technology or recording situation information (GPS location information, Cell ID, EXIF information, etc.) at the time of content creation has been proposed, but currently, only context information is recorded together in a photo. I'm just at the level.

 Therefore, if you infer the user's behavior pattern by using various information stored in the mobile device, not just the picture, and display the content by using the inferred information and the collected situation information (place, time, and other information) It can be used as a useful indicator to manage or search the contents of the program.

Therefore, the technical problem to be achieved by the present invention is to infer the user's behavior pattern from the data collected through the mobile device, and to provide the inferred information in the form of metadata for the management and search of various multimedia content, personal daily life management device And to provide a management method.

In addition, another technical problem to be achieved by the present invention is to provide the information generated by inferring the information and behavior patterns collected from the mobile device in a visualized interface so that the user can tag the semantic information directly.

Another object of the present invention is to provide a computer-readable recording medium having recorded thereon a program for executing an individual's daily life management method on a computer.

However, the technical problem of the present invention is not limited to the above-mentioned matters, and other objects not mentioned will be clearly understood by those skilled in the art from the following description.

In order to achieve the above technical problem, an individual's daily life management apparatus according to the present invention includes a data collector for collecting context information of a user through a mobile device, and a user's behavior by applying a probabilistic model to the context information. A behavior pattern inference unit that infers pattern information, a tagging UI unit that forms context information and the behavior pattern information as a visualized interface, and receives tagging information from a user, the context information, the behavior pattern information, and the tagging It may include a metadata generator for analyzing the information to generate a metadata that is a meaningful form of information.

Preferably, the tagging UI may be configured as a visualized interface by scheduling the context information and the behavior pattern information by a predetermined time unit.

Also preferably, the probabilistic model may be in the form of a modular Bayesian network. In addition, the modular Bayesian network may be a hierarchical structure having level information for each node that is a variable affecting the behavior pattern. More preferably, each module constituting the probabilistic model may include a virtual node that reflects probabilistic evidence.

In addition, the interface may include at least one of a pull-down menu, an icon menu, a life log window, a photo tab, a life log tab, and an action button.

In order to achieve the above technical problem, the individual's daily life management method according to the present invention comprises the steps of: (a) collecting the user's context information through a mobile device, and (b) applying a probability model to the context information to the user Inferring the behavior pattern information of (c), (c) forming the context information and the behavior pattern information into a visualized interface, and (d) input tagging information from the user using the interface formed in the step (c). Receiving step may include.

 Further, after step (d), it may be further preferable to further include (e) analyzing the context information, the behavior pattern information, and the tagging information to generate metadata that is meaningful information.

Also preferably, the probabilistic model may be in the form of a modular Bayesian network. Each module constituting the probabilistic model may include virtual nodes that reflect probabilistic evidence.

In addition, the modular Bayesian network may be a hierarchical structure having level information for each node that is a variable affecting the behavior pattern.

In addition, in order to achieve the above technical problem, the present invention may provide a computer-readable recording medium that records a program for executing an individual's daily life management method on a computer.

As described above, according to the individual's daily life management apparatus and management method according to the present invention, it is possible to generate metadata reflecting the user's behavior pattern in the multimedia content with relatively little effort and time. By visually expressing to the user, users can easily store their experiences and memories in a database and use them for later retrieval and management.

Hereinafter, in the following description of the present invention with reference to the accompanying drawings, it should be noted that the same reference numerals for the same components have the same reference numerals as much as possible even if displayed on different drawings. At this time, the configuration and operation of the present invention shown in the drawings and described by it will be described as at least one embodiment, by which the technical spirit of the present invention and its core configuration and operation is not limited.

As used herein, the term 'unit' refers to a hardware component such as software or a field programmable gate array (FPGA) or an application specific integrated circuit (ASIC), and performs certain roles. But wealth is not limited to software or hardware. The 'unit' may be configured to be in an addressable storage medium and may be configured to execute one or more processors. By way of example, a 'part' is a component, such as software components, object-oriented software components, class components, and task components, and processes, functions, properties, procedures, subroutines. , Segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. In addition, the functionality provided in each component may be combined into a smaller number of components or further separated into additional components.

The context information used in the present invention refers to context information and state information related to the user of the mobile device, and means various data collected by the mobile device in the present invention. For example, it refers to the web, schedule information, or information obtained by a GPS device, and simultaneously uses multimedia content generated by a mobile device. More specifically, personal information such as age, gender, occupation, symbols, habits, addresses, anniversaries, information about calls, SMS, image capture, Beams such as music file playback, location information collected using a GPS device, and the like may be used.

In addition, the individual's daily life management device may be implemented as a digital device. A digital device is a device having a digital circuit that processes digital data, and includes a computer, a digital camera, a digital home appliance, a digital telephone, a digital project, a home server, a digital video recorder, a digital satellite broadcast receiver, a set-top box, and a digital television broadcast receiver. Etc. can be mentioned. In the following embodiment, a case where the proposed method is implemented based on information collected from a mobile phone will be described as an example.

1 is a block diagram showing an apparatus for managing a daily life of an individual according to an embodiment of the present invention, and FIG. 2 is a block diagram showing a behavior pattern inference unit in an embodiment of the present invention.

As shown in FIG. 1, the apparatus 100 for managing everyday life of an individual according to an embodiment of the present invention includes a data collector 110, a behavior pattern inference unit 120, a tagging UI unit 130, and metadata. The generation unit 150 is included.

The data collector 110 collects context information related to the daily life of the user through the mobile device. For example, the data collector 110 collects user location information. To this end, the data collection unit 110 may include a Global Positioning System (GPS). The satellite navigation apparatus receives a coordinate value indicating the position of the user. Or it can collect information about the visit place using the GSM Cell ID.

 In addition, the data collection unit 110 may collect information such as weather, temperature, wind speed, and news through the web. In addition, the data collector 110 may collect usage information of the mobile device 10, that is, log information related to call history, short message service (SMS) transmission and reception, image capturing, and multimedia content playback. Specifically, when the user transmits the short message, the data collection unit 110 collects data such as the content of the short message, the short message receiver, and the time when the short message is transmitted and received. In the case of the call history, the data collection unit 110 collects data on, for example, a call partner, a call counterpart number and group, a call frequency, a talk time, a call volume, and the like. In addition, when a music file (DMB, video file, etc.) of multimedia content is played, the data collection unit 110, for example, the genre, title, artist name (actor name) of the played music file (DMB, video file, etc.) ), Information about the number of times of play, time of play, etc. is collected. In addition, the data collector 110 may collect charging state information of the mobile device 10. The type of information collected by the data collector 110 may vary depending on a function or a sensor provided by the mobile device 10.

A database may be used to store metadata generated according to the present invention. As a specific embodiment, a database may be implemented as the DB server 140 as shown in FIG. 1. The DB server 140 converts and stores the context information collected by the data collector 110 into an appropriate form. In this case, a short term storage method may be used for simply collecting information, and the information already stored in the DB server 140 may be stored in a large capacity server connected to the mobile device 10 through a wireless or wired network. After moving to (not shown) it can be stored using a long term storage method. As will be described later, the DB server 140 stores metadata generated by the metadata generator 120 in addition to the context information collected from the mobile device 10, and also, at least a part of the generated metadata and context information. It includes the metadata associated with and stores the probabilistic model that is composed by summarizing the user's behavior patterns.

The behavior pattern inference unit 120 probably infers a behavior pattern of the user by using the information provided by the data collection unit 110.

 The metadata generation unit 150 may include a metadata generation unit 150 including a rule analysis unit 151, a statistical analysis unit 152, and a probability analysis unit 153.

The rule analysis unit 151 analyzes the context information of the user collected from the mobile device 10 to generate metadata satisfying a specific rule. For example, when the collected context information is a coordinate value representing a user's location, that is, GPS location information, metadata about a place corresponding to the coordinate value represented by latitude and longitude is generated. Or if the collected context information is a call number from a telephone number, for example, 010-XXX-XXXX, such as 'call from a friend' or 'call from a stranger' according to a predefined rule for the call list. Generate the same metadata.

The statistical analyzer 152 analyzes the context information collected from the mobile device 10 to generate metadata representing statistical meanings. For example, if the collected context information is a call volume, it analyzes it and generates metadata such as 'high call frequency' or 'very high call frequency' when the call volume is higher than the usual value in a specific time period.

The probability analyzer 153 generates metadata based on the context information collected from the mobile device 10 or metadata generated by the rule unit 151 or the statistical analyzer 152. In this case, as a probability model to be used, a probabilistic model previously constructed as an initial model is used, and when the probability of the metadata obtained through the probability model exceeds or exceeds a predetermined reference value, the metadata indicates a user's situation or behavior. You can think of it as data.

In addition, the probabilistic model may use a probabilistic model based on a behavioral theory to infer behavioral pattern information. The probabilistic model based on the theory of behavior herein refers to a model that can calculate the probability of given information through the conditional probability association between the information. In this embodiment, a probability model in the form of a Bayesian network may be used as the probability model.

A Bayesian network is a type of directed acyclic graph (DAG) that expresses the connection relationship of nodes. The Bayesian network can efficiently express and calculate many probability relationships by the Conditional Probability Table (CPT) defined by this structure. It can be a model. In Bayesian networks, the relationship between variables inherent in data is represented by nodes and arcs, and the conditional independence assumption is used for probability relationships not represented by arcs. In a Bayesian network, each node refers to a variable that affects a behavior pattern, and an arc represents an association between each node. A detailed description of the Bayesian network probabilistic model is given in [K. B. Korb, and A. E. Nicholson, Bayesian Artificial Intelligence, Chapman & Hall / CRC, 2003].

 In the present invention, a probabilistic model hierarchy based on a behavioral theory is used as part of a design method for producing a probabilistic model inferring a behavioral pattern.

3 is a diagram illustrating a connection relationship between a context node and an evidence node for a specific action according to an embodiment of the present invention.

3 shows that the hierarchical structure of the probabilistic model for inferring a single behavior called 'study' is classified by each category of the behavior theory. Nodes in a hierarchical structure can be classified into input nodes and output nodes. Here, the input node means a node that affects a specific output node, and the output node means a node that is affected by at least one input node. In the present invention, in probabilistic modeling, three layers of a target node, a context node, and an evidence node are basically constructed. The target node may be one of actions defined as user actions such as 'studying', 'classes' or 'playing video games' as an action to be recognized. The context node represents a category of contexts. Basically, one context node is created for each of the five context categories that exist in the theory of behavior. The evidence node is the node that affects the context node, which is the parent node, and corresponds to the input node for the context node. Here, it is broken down as shown in [Table 1] . In other words, there is space and time for any action. For example, schooling is appropriate for taking classes, and morning or afternoon of weekdays are appropriate for time. The user's condition should be sitting or moving less, and the surrounding sounds should be quiet. And in most cases, you will take classes with others. In conclusion, it is possible to make a complete guess on the behavior in consideration of all these spaces / times / states / circumstances / social roles. In order to increase the accuracy of this behavioral pattern inference, a method of collecting such information from information through a mobile device, classifying each information into types, and dividing each information into context nodes and assigning reliability to each context is used.

TABLE 1

Type of context node detailed description Spatio-Context Evidence based on spatial location appropriate for the inference Temporal context Include temporal information that corresponds to the behavior to infer Personal context Display the user's status for the reasoning action Environmental Context Used to model surrounding objects or sounds Social context Use the user's social role or friendship as evidence

3 shows connecting context nodes and evidence nodes for specific behaviors. The appropriate spatial context for the behavior of studying could be school, library, or home. Of course, each place actually has a different degree of suitability for study, and the influence of this spatial location on the behavior of study also varies. Here, the relationship between the Evidence node and the context node is used to model how well a place such as 'school', 'library', etc. is suitable for study, and again through the relationship between the context node and the 'study' action. Model how relevant is. This structure of the present invention enables minimizing the change of the overall probability model by limiting the change of the probability modeling to a specific context node when the evidence variable is added by the improvement of the device performance or the addition of the sensor.

4 illustrates an implementation of a Bayesian network (BN) designed according to one embodiment of the invention. In FIG. 4, the nodes shown in the bold solid boxes represent context nodes, and the detailed behavior Bayesian network is designed for all defined behaviors in order to recognize the user's behavior. However, it is difficult to accurately recognize all of these behaviors from the information collected on mobile devices. This is because it is difficult to pinpoint similar behaviors that contain similar evidence. However, even if it is not possible to identify similar behaviors correctly, it can provide a basis for judging the current behavior if it can determine which category the behavior belongs to. Therefore, the present invention proposes a higher action category including each detailed action, and is designed to infer the higher action by using the probability value for each action as evidence. In other words, we add virtual nodes so that all observed evidence and probabilities are passed to the higher-order node.

Hereinafter, a description will be given of a process of inferring higher-level behavior by adding virtual nodes.

5A is a diagram for explaining a process of inferring a higher behavior by adding a virtual node according to another embodiment of the present invention. As shown in Figure 5, each individual behavior is used to infer one large behavior category. In other words, behaviors such as “Sports” and “Dance” are used as a result of inferring a large category of behavior, “Active Leisure”. This approach can increase the precision of the category of behavior even if the accuracy of the individual behavior is low.

In addition, in this embodiment, the Bayesian network probability model is preferably a modular structure. 5B is a conceptual diagram of a modular Bayesian network according to another embodiment of the present invention. As shown in Fig. 5B, the Bayesian network is modularized according to the divided domains PN1, PN2, ..., PNn. Here, the Bayesian network may be modularized according to a predetermined criterion, for example, a meal related probability module and a traffic situation probability module. In this case, the probability analysis unit 153 generates metadata such as 'during meal' or 'during out' through a meal related probability module (eg, PN1), and a traffic state probability module (eg, PN2). Generates metadata related to traffic or traffic congestion currently in use by the user.

In this process, the probability calculation method used in the representative Bayesian network probability model can be used. Furthermore, when the probability model is divided into several probability modules, it is efficient to perform probability calculation in consideration of the association between the modules. For this purpose, as shown in FIG. 5B, a virtual concatenation technique may be used to store metadata generated by each probability module and its probability value and input the data as a probability evidence. In the virtual concatenation technique, a virtual node is added to reflect probabilistic evidence, and a probability of evidence is applied through a conditional probability value (CPV) of the added node. This virtual connection scheme is described in [K.-S. Hwang, S.-B. Cho and Jong-Ho Lea, "A Bayesian inference model for landmarks detection on mobile devices," Journal of Korea Information Science Society: Computing Practices, vol. 13, no. 1, pp. 35-45, 2007. 02.] detailed description thereof will be omitted.

When the user's behavior pattern is inferred as described above, the tagging UI 130 (see FIG. 1) uses the probability of the user's behavior pattern inferred by the behavior pattern inference unit 120 to match the current behavior. Pattern information is first formed and provided as a visualized interface. In the present invention, tagging is a keyword that is a search tag that can represent the contents of the content, which is summarized in several words by the person who created and updated the data. It means tagging.

 The tagging UI unit 130 is configured to receive tagging information from the user, and the user may select the behavior pattern information displayed on the interface and store additional information. The tagging information may be stored in the DB server 140. It can be stored and later used to retrieve, sort and manage the data.

In addition, the tagging UI unit 130 may form the behavior pattern information inferred with the highest probability among the behavior pattern information as the first visualized interface. For example, the tagging UI unit 130 may form the visualized interface as the high probability value. The size of the behavior button to be described later corresponding to the behavior pattern information inferred to be greater than the size of the behavior button corresponding to the behavior pattern information inferred with a lower probability value, or the emphasis or emphasis in visualizing the behavior pattern information on the interface. Means to mark using the mark. In addition, the behavior pattern information to be visualized preferentially may be achieved by reducing the size of the action button in the order of high probability or adjusting the color or line thickness of the behavior pattern information to be highlighted.

6 is a view showing an interface that a user can directly tag according to an embodiment of the present invention, Figure 7 shows a variety of action buttons that can be directly tagged by the user according to an embodiment of the present invention according to the probability of occurrence It is a figure which illustrates the visualization.

As illustrated in FIG. 6, the tagging UI unit of the present invention includes an interface that a user can tag directly, and the interface includes a pull-down menu, an icon menu, a lifelog window, a photo tab, a lifelog tab, and an upper action button and a lower one. Contains an action button (no reference number). In addition, the tagging UI may schedule the context information and the behavior pattern information in a predetermined time unit to form a visualized interface. For example, the interface provided in the tagging UI 130 may provide the user with the status information corresponding to the selected date in the form of a schedule of 30 minutes. At this time, the information existing in each line has start time and end time, and it is indicated by numbers or letters according to the characteristics of each information. For example, if you listen to music (mp3) or send and receive text messages (sms), the number of occurrences is displayed as a number to indicate the number of times, while Busy is used to indicate the amount of activity of the current user. The status of the current user's activity, such as, Idle, etc.

The interface selects a user's name and date, or selects a file to tag, and the log flow for 30 minutes can be viewed in the lifelog window located in the center of the screen. In addition, when the user clicks on a time to be tagged, the user may be recommended to the higher level context in the order of the highest probability using the upper action button and the lower action button.

In order to effectively visualize the interface, important information such as place, state, emotion, and high-level and low-level actions were pointed out in prominent colors, and action buttons were provided with appropriate images to enhance the intuitiveness, learning, and design of the interface.

As shown in FIG. 7, the interface of the present invention expresses an action inferred through the behavior pattern inference unit 130 as an image in which the size and color of the action button are different according to the probability of occurrence. It is divided into main category and sub category.

The action defined in the main category is the size of the action button according to the probability of occurrence. The larger the action button, the higher the probability of occurrence.

An action defined in a sub category represents a sub action related to the selected main category, and as in the main category, represents a difference in occurrence probability as the size of the action button.

The user may select an action button recommended as the main category, and may select an action button of a sub category that is changed accordingly. Referring to FIG. 7, an N mark is displayed in an upper right area of each category, which is provided to search for other action buttons not represented by action buttons according to the behavior pattern information recommended for each category. In FIG. 7, three action buttons are formed for each category, but as an example, the number may be varied.

Next, a description will be given of a daily life management method of an individual according to an embodiment of the present invention.

8 is a flowchart illustrating a method for managing an individual's daily life according to an embodiment of the present invention, and FIG. 9 is a flowchart illustrating a behavior pattern inference method according to an embodiment of the present invention.

Referring to FIG. 8, the method of managing daily life of an individual may include collecting context information of the user (S100), generating metadata by analyzing context information (S200), and inferring a behavior pattern using a probabilistic model. In step S300, expressing the context information and the behavior pattern information in a visualized interface (S400), the user directly tagging his information through the interface (S500). Thereafter, the method may further include storing the metadata in a database so as to convert the collected context information, behavior pattern information, and information tagged by the user into metadata, and use the same.

In step S100, the context information about the user is collected through the mobile device, and the information is collected by the data collecting unit. Since the information related to the information collection in the data collection unit has been described above, it will be omitted here.

In step S200, the metadata is generated by analyzing the context information collected in the data collection unit of step S100, and is implemented in the metadata generator provided in the behavior pattern inference unit. A method of generating metadata, which is meaningful information, from the context information will be described later.

Step S300 is a step of generating behavior pattern information of the user by applying a probability model based on the generated metadata, which is implemented in the behavior pattern inference unit, and in the present invention, a probability model in the form of a Bayesian network is used. Bayesian networks are probabilistic models based on behavioral theory, and it is desirable to infer behavioral patterns using hierarchical structures. More preferably, the probabilistic model may use a method of inferring higher-order behavior using a virtual node, and modularizes a Bayesian network probabilistic model to form a divided domain to store metadata generated from each probabilistic model and its probability value. A virtual concatenation technique can be used to input and reflect probability evidence. Since a detailed description thereof has been described above, it will be omitted here.

In step S400, the context information of the user and the inferred behavior pattern information are formed in a visualized interface. The present invention includes a tagging UI unit and provides the information to the interface. Detailed descriptions of the interface are immediately described above.

Step S500 is a step in which a user tags information using an interface. For example, when a user clicks on a time to be tagged, a sequence corresponding to the time using a higher action button and a lower action button has a higher probability. The upper level context is recommended. If there is photo information, the photo is provided in the photo tab. In addition, important points for the effective visualization of the interface, such as places, states, emotions, and high-level and low-level actions, are given points with prominent colors, and action buttons are provided with appropriate images to enhance the intuitiveness, learning, and design of the interface. .

The step S600 is a step of converting the information tagged by the user into a metadata form and storing it in a database. This step is to store the tagging information in a database so that the multimedia data can be retrieved later, classified and managed. In step S500, the user can modify, supplement, or delete the tagging information. The information is again converted into metadata form through the behavior pattern inference unit and stored in the database.

Hereinafter, a method of generating metadata, which is meaningful information, from the context information of step S200 will be described with reference to FIG. 9.

9 is a flowchart illustrating a process of performing metadata in a metadata generator of the present invention. Generating metadata may include generating metadata that satisfies a specific rule (S210), generating metadata representing statistical meanings (S220), and generating metadata through a probability model (S230). Include.

Step S210 is implemented in the rule analyzer of the metadata generator and analyzes the context information of the user collected from the mobile device to generate metadata that satisfies a specific rule.

Step S220 is implemented in the statistical analysis unit of the metadata generator, and analyzes the context information collected from the mobile device to generate metadata representing the statistical meaning.

Step S230 is implemented in the probability analysis unit of the metadata generation unit, and generates metadata through the probability model based on context information collected from the mobile device or metadata generated by the rule analysis unit or the statistical analysis unit.

The daily life management method of an individual according to the present invention includes a computer readable medium including program instructions for performing operations implemented by various computers. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. The media may be program instructions that are specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. Examples of computer readable recording media include magnetic media such as hard disks, floppy disks and magnetic tape, optical media such as CD-ROMs and DVDs, and magnetic disks 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 not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like.

As described above, the present invention has been described by way of limited embodiments and drawings, but the present invention is not limited to the above-described embodiments, which can be variously modified and modified by those skilled in the art to which the present invention pertains. Modifications are possible. Accordingly, it is intended that the scope of the invention be defined solely by the claims appended hereto, and that all equivalents or equivalent variations thereof fall within the spirit and scope of the invention.

1 is a block diagram showing an apparatus for managing a daily life of an individual according to an embodiment of the present invention;

2 is a block diagram showing a behavior pattern inference unit in an embodiment of the present invention;

3 is a view showing a connection between a context node and an evidence node for a specific action according to an embodiment of the present invention;

4 illustrates an implementation of a Bayesian network (BN) designed according to one embodiment of the invention,

FIG. 5A is a view for explaining a process of inferring a higher behavior by adding a virtual node according to another embodiment of the present invention; FIG.

5B is a conceptual diagram of a modular Bayesian network according to another embodiment of the present invention;

6 is a view showing an interface that a user can directly tag according to an embodiment of the present invention;

FIG. 7 is a diagram illustrating various visualizations of action buttons that can be directly tagged by a user according to an embodiment of the present invention according to a probability of occurrence; FIG.

8 is a flowchart illustrating a method for managing an individual's daily life according to an embodiment of the present invention;

9 is a flowchart illustrating a process of generating metadata in a metadata generator of the present invention.

Claims (13)

In the daily life management device of the individual based on the information collected through the mobile device, A data collector configured to collect context information of the user through the mobile device; A behavior pattern inference unit inferring behavior pattern information of a user by applying a probabilistic model to the context information; A tagging UI unit which forms the context information and the behavior pattern information as a visualized interface and receives tagging information from a user; And And a metadata generator for analyzing the context information, the behavior pattern information, and the tagging information and generating the metadata as metadata which is meaningful information. The method of claim 1, The tagging UI unit is configured to schedule the context information and the behavior pattern information by a predetermined time unit (scheduling) to form a visualized interface of the individual, characterized in that the individual. The method of claim 1, The probability model is a daily life management device of the individual, characterized in that the modular Bayesian Network (Bayesian Network) form. The method of claim 3, The modular Bayesian network has a hierarchical structure having a level information for each node that is a variable affecting the behavior pattern. The method of claim 4, wherein Each module constituting the probability model includes a virtual node reflecting probabilistic evidence. The method of claim 1, The interface includes at least one of a pull-down menu, an icon menu, a life log window, a photo tab, a life log tab, and an action button. The method of claim 6, The action button includes an image corresponding to the action, and the daily life management apparatus of the individual, characterized in that the size and color of the action button according to the probability of occurrence of the inferred behavior pattern. In the daily life management method of the individual based on the information collected through the mobile device, (a) collecting context information of the user via the mobile device; (b) inferring behavior pattern information of a user by applying a probabilistic model to the context information; (c) forming the context information and the behavior pattern information into a visualized interface; And (d) receiving the tagging information from the user by using the interface formed in step (c). The method of claim 8, wherein after step (d), and (e) analyzing the context information, the behavior pattern information, and the tagging information to generate metadata which is meaningful information. The method of claim 8, In step (b), the probability model is a daily life management method of the individual, characterized in that the modular Bayesian network form. The method of claim 10, Each module constituting the probabilistic model includes a virtual node reflecting probabilistic evidence. The method of claim 11, The modular Bayesian network has a hierarchical structure having a level information for each node that is a variable affecting the behavior pattern. A computer-readable recording medium having recorded thereon a program for executing the individual's daily life management method according to any one of claims 8 to 12.

Images