JP2006134308A - Operation history usage system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an operation history usage system performing a timely service and a service reflecting characteristics in a life of each person. <P>SOLUTION: A user agent part 202 is provided with a pattern extraction part 404 extracting a history frequent operation pattern satisfying a predetermined condition from an operation history data registered in an operation history database 402, a pattern database 405 storing the frequent operation pattern, a pattern monitoring part 406 monitoring a pattern matching the frequent operation pattern registered in the pattern database 405 from operation history data to be newly registered, and a pattern classification part 601 specifying an operator of the frequent operation pattern stored in the pattern database 405. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an operation history utilization system for effectively using operation histories of a plurality of devices and providing an appropriate information service to a user.

  As a technique for predicting user behavior, a method for using marketing history and information recommendation using a movement history has been proposed. For example, a user's movement history is recorded with an information terminal provided with a PHS (Personal Handyphone System) or a mobile phone function, the user's action is predicted from the movement history, and the action predicted via the PHS network or the mobile phone network There is a method for transmitting information corresponding to the information terminal to the information terminal (see, for example, Japanese Patent Laid-Open Nos. 2000-293540 and 2000-324246).

  However, the conventional technology is for a single device, such as collecting a mobile phone movement history between base stations. In the future, networking of devices is expected to continue to progress, and various historical information on multiple devices owned by users will be integrated to conduct marketing surveys and content recommendation such as analysis of user preferences via the network. Need to be available for other services.

  It also uses context-aware services that provide services tailored to the current situation using information that analyzes user preferences and conditions based on various historical information, and individual preferences and interests. And services that better reflect interests and usage (hereinafter collectively referred to as characteristics in daily life).

  Therefore, the present invention has been made in view of the above circumstances, and provides an operation history utilization system capable of providing a service according to the situation at that time and a service more reflecting the characteristics of each individual's life. For the purpose.

  In order to achieve the above object, an operation history utilization system according to the present invention is based on an operation history management device that manages operation history data transmitted from at least one device, and operation history data managed by the operation history management device. The operation history utilization system is connected to a service providing device that controls the operation of the device according to the user, and the operation history management device transmits an operation history registered in the operation history database. Sort the data by operation date and time, classify operation history data whose interval between adjacent data is less than or equal to a predetermined time as the same group, extract operation sequences that are the same or partially match between groups, and use them as frequent operation patterns The pattern extraction means to be registered in the pattern database and the operator with certainty about the definitive operation in advance The frequent operation pattern including the definitive operation is extracted from the frequent operation patterns registered in the pattern database, and the operator's confidence level is determined as a user corresponding to the decisive operation. The frequent operation patterns that occur after the frequent operation patterns having the operator's certainty are registered in the pattern database, and the operator's certainty is registered in the pattern database according to the frequency of occurrence. Until the frequent operation pattern disappears, the frequent operation pattern generated following the frequent operation pattern having the certainty of the operator is extracted, and the certainty of the operator is registered in the pattern database according to the frequency of occurrence. A pattern to register the certainty of the corresponding operator in the pattern database by performing recursively And a pattern monitoring unit that searches for a partial match with the frequent operation pattern in the pattern database for the operation sequence of operation history data newly transmitted from the device, and the service providing apparatus includes: When an operation sequence of the operation history data newly transmitted from the device is found by the pattern monitoring unit, when the part that matches the frequent operation pattern is found, the frequent operation pattern and the operator of the frequent operation pattern A service decision for determining whether to perform operation control of the device according to an operation subsequent to the matched operation sequence included in the frequent operation pattern or to output information according to a user who is an operator based on Means are provided.

  In addition, the operation history utilization system according to the present invention provides an operation history management device that manages operation history data transmitted from at least one device and a user based on the operation history data managed by the operation history management device. An operation history utilization system connected to a service providing device that controls operation of the device, wherein the operation history management device stores operation history data transmitted from the device and registered in an operation history database according to an operation date and time. Sorting and classifying operation history data whose adjacent data is less than or equal to a predetermined time as the same group, extracting operation sequences that are identical or partially matching between groups, and registering them in the pattern database as frequent operation patterns About the operation sequence of the operation history data newly transmitted from the pattern extraction means and the device, Pattern monitoring means for searching for a part that matches the frequent operation pattern of the turn database, and operation history data registered in the operation history database in advance by defining the operator in association with the certainty for the definitive operation. The operation history data newly extracted from the device by extracting a frequent operation pattern including the decisive operation from the device, registering the certainty of the operator as the user corresponding to the decisive operation in the pattern database The frequent operation pattern having the certainty of the operator is extracted from, and the frequent operation pattern that appears without sandwiching other operations following the frequent operation pattern, or a part of operations overlaps, is extracted. The operator's confidence is registered in the pattern database according to the degree of adjacency that is the degree of overlap, and Until the frequent operation pattern does not appear, recursively, the frequent operation pattern that appears without intervening other operations following the frequent operation pattern, or a part of operations overlaps is extracted, and the degree of overlap is extracted. Pattern classification means for registering the operator's certainty factor in the pattern database by registering the operator's certainty factor in the pattern database in accordance with a certain degree of adjacency, and the service providing device by the pattern monitoring unit When an operation sequence of operation history data newly transmitted from the device is found that partially matches the frequent operation pattern, based on the frequent operation pattern and the operator of the frequent operation pattern, The operation of the device according to the operation after the matched operation sequence included in the frequent operation pattern Service determining means is provided for determining whether to perform control or to output information according to a user who is an operator.

  The operation history utilization system according to the present invention is based on at least one device, an operation history management device that manages operation history data transmitted from the device, and operation history data managed by the operation history management device. An operation history utilization system to which a service providing apparatus for controlling the operation of the device according to a user is connected, wherein the device associates operation details for operating the device with information on the operation date and time. The operation history management device sorts operation history data transmitted from the device and registered in the operation history database by operation date and time, and records operation history data whose interval between adjacent data is equal to or less than a predetermined time. Categorize as the same group, extract operation sequences that are identical or partially match between groups, and pattern as frequent operation patterns A pattern extracting means to be registered in the database, an operator is defined in association with a certainty in advance with certainty, and a frequent operation pattern including the deterministic operation is extracted from the operation history data registered in the operation history database. When the frequent operation pattern includes a plurality of decisive operations of different operators, pattern classification means for registering the certainty of each operator in the pattern database as being a user corresponding to the decisive operation, Pattern monitoring means for searching for an operation sequence of operation history data newly transmitted from the device that partially matches the frequent operation pattern of the pattern database, and the service providing apparatus includes the pattern monitoring means To the operation sequence of the operation history data newly transmitted from the device. If a part that matches the frequent operation pattern is found, based on the frequent operation pattern and the operator of the frequent operation pattern, the subsequent operation sequence included in the frequent operation pattern Service control means is provided for determining whether to control the operation of the device according to an operation or to output information according to a user who is an operator.

  An operation history utilization system according to an embodiment of the present invention provides an operation history management device that manages operation history data transmitted from at least one device, and a user based on operation history data managed by the operation history management device. The operation history utilization system is connected to a service providing device that controls the operation of the device according to the operation history management device, and the operation history management device performs an operation that satisfies a predetermined temporal condition from the operation history data transmitted from the device. A pattern extracting means for extracting a frequent operation pattern that is a series of the first frequent operation pattern, and a first frequent operation pattern that is a frequent operation pattern including a definitive operation associated with a predefined operator among the frequent operation patterns From the operation history data transmitted in the past from the device, the operator of the second frequent operation pattern that is a frequent operation pattern other than A pattern classifying means for calculating a statistical occurrence chain frequency for the first frequent operation pattern to be estimated as an operator of the frequent operation pattern, and an operation sequence of the operation history data newly transmitted from the device Pattern monitoring means for monitoring whether or not it matches a frequent operation pattern, and the service providing apparatus has an operation sequence of operation history data newly transmitted from the device as a result of monitoring by the pattern monitoring means. Service decision means for determining the content of operation control of the device according to the user who is an operator based on the frequent operation pattern and the operator of the frequent operation pattern when it matches at least a part of the frequent operation pattern It is characterized by providing.

  Thus, operation history data of individual devices is collected, the operation history data is centrally managed, and frequent operations are performed without providing a means for directly identifying a user such as an authentication device. It can be estimated by which user the pattern is operated. Further, by estimating the user in this way, it is possible to provide a detailed service corresponding to the user, such as controlling the operation of the device according to the user.

  The operation history utilization system according to the embodiment of the present invention is based on an operation history management device that manages operation history data transmitted from at least one device, and operation history data managed by the operation history management device. An operation history utilization system connected to a service providing device that controls operation of the device according to a user, wherein the operation history management device sets a predetermined time condition from operation history data transmitted from the device. Pattern extraction means for extracting a frequent operation pattern that is a series of operations to be satisfied, pattern monitoring means for monitoring whether or not an operation sequence of operation history data newly transmitted from the device matches the frequent operation pattern, Corresponds to a predefined operator in the frequent operation patterns included in the operation history data newly transmitted from the device The operator of the second frequent operation pattern that is a frequent operation pattern other than the first frequent operation pattern that is the frequent operation pattern including the determined decisive operation is designated as the first frequent operation pattern and the first frequent operation. A pattern classifying unit that calculates a degree of adjacency with the second frequent operation pattern generated following the operation pattern and estimates the operator as an operator of the frequent operation pattern, and the service providing apparatus is monitored by the pattern monitoring unit As a result, when the operation sequence of the operation history data newly transmitted from the device matches at least a part of the frequent operation pattern, the operator operates based on the frequent operation pattern and the operator of the frequent operation pattern. Service determining means for determining the content of operation control of the device according to a certain user is provided.

  As a result, it is possible to estimate which user has operated the frequent operation pattern that appears frequently without providing a means for directly identifying the user, such as an authentication device, as described above. Further, by estimating the user in this way, it is possible to provide a detailed service corresponding to the user, such as controlling the operation of the device according to the user.

  The operation history utilization system according to the embodiment of the present invention includes at least one device, an operation history management device that manages operation history data transmitted from the device, and an operation history managed by the operation history management device. An operation history utilization system connected to a service providing apparatus that performs operation control of the device according to a user based on data, wherein the operation history management device includes an operation content for the device to operate the device, and Pattern extraction means for transmitting operation history data associated with information related to operation date and time, and extracting frequent operation patterns that are sequences of operations that satisfy a predetermined temporal condition from the operation history data transmitted from the device; In the frequent operation patterns extracted by the extraction means, the deterministic operation associated with the operator is displayed together with the certainty of the operator. It is confirmed whether or not there is a first frequent operation pattern that is a frequently-defined operation pattern, and there are a plurality of the first frequent operation patterns, and operators associated with each decisive operation are different. The pattern classification means for determining the operator of the frequent operation pattern according to the certainty of the operator, and whether the operation sequence of the operation history data newly transmitted from the device matches the frequent operation pattern. And the service providing apparatus, as a result of monitoring by the pattern monitoring means, the operation sequence of the operation history data newly transmitted from the device matches at least a part of the frequent operation pattern. In this case, based on the frequent operation pattern and the operator of the frequent operation pattern, the content of the operation control of the device according to the user who is the operator Characterized in that it comprises a service determination unit determined.

  As a result, it is possible to estimate which user has operated the frequent operation pattern that appears frequently without providing a means for directly identifying the user, such as an authentication device, as described above. Further, by estimating the user in this way, it is possible to provide a detailed service corresponding to the user, such as controlling the operation of the device according to the user.

  The operation history utilization system according to the embodiment of the present invention includes at least one device, an operation history management device that manages operation history data transmitted from the device, and an operation history managed by the operation history management device. An operation history utilization system in which a service corresponding to a user action is determined based on data and a service providing apparatus that provides the determined service to a device is connected, wherein the operation history management device is connected to the device Pattern extracting means for extracting frequent operation patterns from the transmitted operation history data, and pattern monitoring means for monitoring whether or not an operation sequence of operation history data newly transmitted from the device matches the frequent operation patterns; The service providing apparatus includes a user based on a monitoring result by the pattern monitoring means and the frequent operation pattern. Determining whether there is a plurality of services determined by the service determination means and the services determined by the service determination means, and whether the plurality of services can be provided simultaneously; As a result, it is characterized by comprising service selection means for selecting a service to be provided from among the plurality of services when it is impossible to provide the plurality of services simultaneously.

  Accordingly, even when there are a plurality of service candidates to be provided and it is impossible to provide them simultaneously, it is possible to select and provide an appropriate service.

  Here, the service selection means may select a service based on a user priority order defined in advance.

  The service selecting means may select a service based on a priority order of operations defined in advance.

  The service selection means may select a service based on a user distribution probability defined in advance.

  Further, the service selection means may select a service based on a predefined operation allocation probability.

  The operation history utilization system according to the embodiment of the present invention includes at least one device, an operation history management device that manages operation history data transmitted from the device, and an operation history managed by the operation history management device. An operation history utilization system in which a service corresponding to a user action is determined based on data and a service providing apparatus that provides the determined service to a device is connected, wherein the operation history management device is connected to the device Pattern extracting means for extracting frequent operation patterns from the transmitted operation history data, and pattern monitoring means for monitoring whether or not an operation sequence of operation history data newly transmitted from the device matches the frequent operation patterns; Based on the monitoring result by the pattern monitoring unit, the user's unintended operation and the expected operation are confirmed or corrected. Characterized in that it comprises an operation confirmation means for performing.

  As a result, operations that partially deviate from the frequent operation pattern are detected, and the operator is warned about the operation, the operation is interrupted, or the alternative operation expected from the frequent operation pattern is executed. Alternatively, by making corrections, it is possible to prevent an erroneous operation or an unintended operation by the operator.

  Here, the operation confirmation unit is configured to display the operation history data when the operation result of the operation history data newly transmitted from the device partially matches the frequent operation pattern as a result of monitoring by the pattern monitoring unit. The operation sequence that does not match the partially matched frequent operation pattern may be predicted as an operation not intended by the user.

  In addition, the service providing apparatus according to the embodiment of the present invention includes a service determining unit that determines a service according to a user action based on information received from an operation history management apparatus that manages operation history data of at least one device. It is determined whether there are a plurality of services determined by the service determination means and the plurality of services can be provided simultaneously. As a result of the determination, it is not possible to provide the plurality of services simultaneously. Service selection means for selecting a service to be provided from the plurality of services when possible, and service output means for providing a service selected by the service selection means to a device.

  Accordingly, even when there are a plurality of service candidates to be provided and it is impossible to provide them simultaneously, it is possible to select and provide an appropriate service.

  An operation history management apparatus according to an embodiment of the present invention is an operation history management apparatus that manages operation history data transmitted from at least one device, and frequently operates from operation history data transmitted from the device. Pattern extraction means for extracting a pattern, pattern monitoring means for monitoring whether an operation sequence of operation history data newly transmitted from the device matches the frequent operation pattern, and monitoring results by the pattern monitoring means An operation confirmation unit that confirms or corrects an operation unintended by the user and an expected operation is provided.

  As a result, operations that partially deviate from the frequent operation pattern are detected, and the operator is warned about the operation, the operation is interrupted, or the alternative operation expected from the frequent operation pattern is executed. Alternatively, by making corrections, it is possible to prevent an erroneous operation or an unintended operation by the operator.

  The present invention can be realized not only as such an operation history utilization system, an operation history management apparatus, and a service providing apparatus, but also provided by such an operation history utilization system, operation history management apparatus, or service provision apparatus. It can also be realized as an operation history utilization method, an operation history management method, or a service providing method using characteristic means provided as steps, or as a program for causing a computer to execute these steps. Needless to say, such a program can be distributed via a recording medium such as a CD-ROM or a transmission medium such as the Internet.

  Hereinafter, first to sixth embodiments of the present invention will be described with reference to the drawings.

(First embodiment)
FIG. 1 is a block diagram showing a configuration of an operation history utilization system according to the first embodiment of the present invention.

  The operation history utilization system receives operation history data of devices (not shown) such as home appliances connected via a network, and controls the devices and provides information based on the operation history data. As shown in FIG. 1, the system includes a user agent unit 202, a service concierge unit 203, and a network IF 410.

  The user agent unit 202 receives user operation history data, extracts and holds a frequent operation pattern, and estimates which user's operation the frequent operation pattern results from. The operation history reception unit 401, the operation history A database 402, an operation history registration unit 403, a pattern extraction unit 404, a pattern database 405, a pattern monitoring unit 406, and a pattern classification unit 601 are provided.

  The operation history receiving unit 401 receives operation history data such as operations performed on devices connected via a network. The operation performed on the device may be performed on the device main body, or may be operated on a device other than the device main body using, for example, a remote controller. The operation history registration unit 403 registers the operation history data received by the operation history reception unit 401 in the operation history database 402. The operation history database 402 stores operation history data. The pattern extraction unit 404 extracts a frequent operation pattern of a history satisfying a predetermined condition from the operation history data registered in the operation history database 402. The pattern database 405 stores frequent operation patterns of the operation history extracted by the pattern extraction unit 404. The pattern monitoring unit 406 matches the frequent operation patterns registered in the pattern database 405 from the operation history data registered in the operation history database 402 or the operation history data newly registered by the operation history registration unit 403. Monitor. The pattern classification unit 601 identifies the operator of the frequent operation pattern in the operation history stored in the pattern database 405.

  On the other hand, the service concierge unit 203 provides services such as device control and information provision based on information related to the user stored in the user agent unit 202, specifically, user operation history data and frequent operation patterns. The service determination unit 501 and the service output unit 502 are provided.

  The service determination unit 501 estimates service contents to be provided such as device control and information provision to devices connected via a network. The service output unit 502 outputs the service estimated by the service determination unit 501 to a device corresponding to this service.

Next, the operation of the operation history utilization system configured as described above will be described.
In this embodiment, the frequent operation pattern is specified by the user, but the frequent operation pattern that frequently appears is extracted first. FIG. 2 is a flowchart showing an operation flow when registering operation history data. FIG. 3 is a diagram showing an example of an operation type identifier defined for each operation type used for operation history data. FIG. It is a figure which shows the example of data recorded on the history database.

  The operation history receiving unit 401 confirms whether or not operation history data has been received (step S101). Here, when the operation history data is not received, the operation history receiving unit 401 continues to check whether or not the operation history data has been received at a predetermined cycle. On the other hand, when the operation history data is received, the operation history receiving unit 401 notifies the operation history registration unit 403 that the operation history data has been received (step S102). Here, the operation history data received by the operation history receiving unit 401 is, for example, an operation type identifier defined for each operation type as shown in FIG. 3 and an operation date and time. It takes a form with arguments to indicate.

  The operation history registration unit 403 takes out the operation history data received by the operation history reception unit 401 and registers it in the operation history database 402 (step S103). In the operation history database 402, for example, as shown in FIG. 4, received operation history data is stored in the order of operation date and time.

  Next, frequent operation patterns are extracted from the operation history data registered in the operation history database 402 in this way. FIG. 5 is a flowchart showing the flow of operations when extracting frequent operation patterns.

  The pattern extraction unit 404 confirms whether or not there is an instruction for pattern extraction (step S201). As the pattern extraction instruction, for example, an operation history pattern extraction / update command periodically generated inside the system, an inquiry reception from the outside of the system, or the like can be used. As a result of the confirmation, if there is no pattern extraction instruction, the pattern extraction unit 404 continues to confirm whether or not there is a pattern extraction instruction at a predetermined cycle.

  On the other hand, when there is an instruction for pattern extraction, the pattern extraction unit 404 refers to the operation history data registered in the operation history database 402 and classifies the operation history data into groups under predetermined conditions regarding the operation date and time. Data as shown in FIG. 4 is registered in the operation history database 402 as described above. This registered data is sorted by operation date and time, and data whose interval between adjacent data is a predetermined time or less is classified as the same group. For example, assuming that the interval between adjacent data is 5 minutes or less, in the example shown in FIG. 4, it can be classified into five groups as shown in groups 1001 to 1005.

  Each group classified in this way includes an operation sequence as shown in FIG. Such an operation sequence group belonging to one group is called an episode. In addition, although it does not need to consider the generation | occurrence | production order of the operation series in one episode, it demonstrates as what is not considered here. Further, here, the case where the time interval between adjacent data is equal to or less than a predetermined time has been described for group classification. However, as a simple method, a method of classifying into groups at a predetermined time interval may be used. For example, the time interval is 10 minutes, and the grouping is performed every hour from 10 o'clock to 10 minutes, from 10 minutes to 20 minutes, from 20 minutes to 30 minutes (the same applies hereinafter).

  Next, the pattern extraction unit 404 checks whether there is an operation sequence that co-occurs between the episodes classified as described above (step S202). That is, the pattern extraction unit 404 confirms whether or not there is an operation sequence that is identical or partially matching between episodes (step S203). Here, when there is no operation sequence that is identical or partially identical, the pattern extraction unit 404 ends the process.

  On the other hand, if there is an operation sequence that is identical or partially identical, the pattern extraction unit 404 extracts the operation sequence as a frequent operation pattern (step S204). For example, in the example shown in FIG. 6, the episode 1001 and the episode 1004 are both “PowerOn, Opr1, Opr4”, and the episodes completely match. Also, the episode 1003 and the episode 1005 have the same “PowerOff, Opr2” portion. Therefore, two matching episodes or a part of episodes are candidates to be registered in the pattern database 405 as patterns of operation sequences that appear frequently. Here, as a condition for determining a match or partial match between episodes, it is assumed that at least a predetermined number of at least two elements constituting the episode match.

  Next, the pattern extraction unit 404 checks whether or not the extracted frequent operation pattern is a pattern already registered in the pattern database 405 (step S205). If registered, the pattern extraction unit 404 ends the process. On the other hand, if not registered, the pattern extraction unit 404 registers the extracted frequent operation pattern in the pattern database 405 (step S206). In this manner, the frequent operation patterns as shown in FIG. 7 are registered in the pattern database 405, for example.

  For example, as shown in FIG. The data of No. 1 indicates that the action “perform operation 1 and operation 4 when the device is turned on” is a stylized pattern. The data 2 indicates a pattern of “turn off the power after the operation 2”. No. 3 to No. 5 is extracted in the same procedure as described above and registered in the pattern database 405.

  Next, the identification of the user who performed the frequent operation pattern operation in the pattern classification unit 601 will be described. In specifying a user, a definitive operation is defined for each user in advance.

  FIG. 8 is a diagram illustrating an example of the definition of the definitive operation. The deterministic operation means an operation in which the actor is always a specific user for a certain operation, and is an operation that “the user X always performs this operation”. In the example shown in FIG. 8, it is indicated that the user A always performs Opr5, and the user C always turns off the power. Note that this definitive operation may be related to a user who is executing with high probability, not a user who is always executing for a certain operation. In this case, for example, as shown in FIG. 11, it is desirable to manage the certainty factor according to the probability.

  FIG. 9 is a flowchart showing a flow of an operation when specifying which user has operated the frequent operation pattern. Here, it is assumed that a definitive operation is defined as shown in FIG.

  The pattern classification unit 601 checks whether or not there is a frequent operation pattern including the decisive operation shown in FIG. 8 among the frequent operation patterns registered in the pattern database 405 (step S301). Here, if there is no frequent operation pattern including a definitive operation, the pattern classification unit 601 ends the process. On the other hand, if there is a frequent operation pattern including a deterministic operation, the pattern classification unit 601 defines the frequent operation pattern as an operation by the user corresponding to the corresponding deterministic operation, and registers it in the pattern database 405 (step S1). S302).

  For example, when a frequent operation pattern as shown in FIG. 7 is registered in the pattern database 405, referring to the definitive operation shown in FIG. Since the frequent operation pattern 2 includes “PowerOff”, it is defined as an operation of the user C from FIG. Similarly, no. Since the frequent operation pattern 1 includes “Opr1”, the user D, No. 1 Since the frequent operation pattern of No. 3 includes “Opr3”, user B, No. 3 Since the frequent operation pattern 6 includes “Opr5”, it is defined as a user A operation. These frequent operation pattern Nos. 1, 2, 3, and 6 include decisive operations themselves and are called level 1 frequent operation patterns. When a plurality of deterministic operations are included in one frequent operation pattern and the users corresponding to the deterministic operations are different, for example, the operation of the frequent operation pattern is performed according to the certainty as shown in FIG. You just need to define

  Next, the pattern classifying unit 601 uses the level 1 frequent operations among the frequent operation patterns registered in the pattern database 405 and not classified as the above level 1 frequent operation patterns. It is confirmed whether there is a frequent operation pattern that follows the pattern (step S303). Here, if there is no frequent operation pattern that occurs following the level 1 frequent operation pattern, the pattern classification unit 601 ends the processing. On the other hand, if there is a frequent operation pattern that occurs following the frequent operation pattern of level 1, the pattern classification unit 601 defines the operator with certainty according to the frequency of occurrence and registers it in the pattern database 405 (step S304). That is, the pattern classifying unit 601 refers to the operation history database 402 and the pattern database 405 to determine which frequent operation pattern of level 1 occurs frequently, whether frequent operation patterns other than level 1 or extraction The occurrence frequency such as whether it occurs following the operation sequence other than the frequent operation pattern is searched. Then, the pattern classification unit 601 defines the operator with certainty according to the frequency that occurs following the frequent operation pattern of level 1.

  For example, no. No. 4 frequent operation pattern is No. 4. No. 1 frequently occurring operation pattern (level 1) occurs 60%. 3% frequent operation patterns (level 1) followed by 15%, and frequent operation patterns other than level 1 or 25% occurred following operation sequences other than the extracted frequent operation patterns. To do. In this case, no. No. 1 frequent operation patterns are user C, No. Since the frequent operation pattern No. 3 is defined as user B, the frequent operation pattern No. The operator No. 4 is defined such that the certainty factor of the user C is “0.6” and the certainty factor of the user B is “0.15” according to the frequency of occurrence from the frequent operation pattern of the level 1. This frequent operation pattern No. 4 is called a level 2 frequent operation pattern.

  Next, the pattern classifying unit 601 selects level 2 out of the frequent operation patterns registered in the pattern database 405, among the remaining frequent operation patterns not classified into the level 1 and level 2 frequent operation patterns. It is confirmed whether there is a frequent operation pattern that occurs following the frequent operation pattern (step S305). Here, if there is no frequent operation pattern that occurs following the level 1 frequent operation pattern, the pattern classification unit 601 ends the processing. On the other hand, if there is a frequent operation pattern that occurs following the frequent operation pattern of level 1, the pattern classification unit 601 defines the operator with certainty according to the frequency of occurrence and registers it in the pattern database 405 (step S306).

  For example, by referring to the operation history database 402 and the pattern database 405, No. No. 5 frequent operation pattern is No.5. 50% of cases occur following the frequent operation pattern 4 (level 2), and 50% of cases occur after frequent operation patterns other than level 2 and operation sequences other than the extracted frequent operation patterns. To do. In this case, no. The frequent operation pattern No. 4 is defined such that the certainty factor of the user C is “0.6” and the certainty factor of the user B is “0.15”. The operator of 5 multiplies the certainty of the frequent operation pattern of level 2 and the frequency of occurrence from the frequent operation pattern of level 2, so that the certainty of user C is “0.3” and the certainty of user B is “ 0.075 ". This frequent operation pattern No. 4 is called a level 2 frequent operation pattern.

  Similarly, the same processing as level 3 and level 4 is recursively repeated until the corresponding frequent operation pattern does not exist. In this way, the frequent operation patterns are registered in the pattern database 405 by estimating the operator as shown in FIG.

  By using the frequent operation patterns registered in the pattern database 405 and the operator's information in this way, it is possible to provide a service according to the frequent operation patterns of operations performed by the user. As a service corresponding to the frequent operation pattern, a service that partially matches the pattern database 405 is searched from the received operation sequence, the next operation to be received is predicted, and the predicted operation is performed in advance to control the device. For example, various signals may be considered, such as outputting a signal for the user, estimating a user who is operating from the received operation sequence, and outputting information corresponding to the user.

  For example, the user agent unit 202 and the service concierge unit 203 are connected to a digital TV (hereinafter simply referred to as TV or TV) 101a, a video 101b, an air conditioner 101c, a lighting fixture 101d, an MD via a network IF 410 and a network as shown in FIG. In an operation history utilization system to which a terminal 101 such as a component 101e and a PC (Personal Computer) 101f is connected, the service concierge unit 203 can provide the following service.

  For example, when a frequent operation pattern “listen to music with MD when the TV is turned off” is registered, a service such as automatically turning on the power of the MD component when detecting that the TV is turned off by the user Can be provided.

  Also, for example, an operator with a frequent operation pattern of “listen to music with MD when the TV is turned off and the lighting is slightly dark” is frequently operated by user A, “listen to music with MD when the TV is turned off and the light is turned on”. Assume that the operator of the pattern is estimated as user B and registered in the pattern database 405. In this case, when it is detected that the TV is turned off by the user and the lighting is slightly dimmed, the MD component is automatically turned on, and a pre-registered song that suits user A's preference is selected. Services such as playback can be provided. On the other hand, when the user detects that the TV is turned off and the light is turned on, the MD component is automatically turned on, and a pre-registered song that suits user B's preference is selected and played. And other services can be provided.

  Further, for example, an operator with a frequent operation pattern of “turn on the air conditioner when turning on the TV” or “set the temperature to 27 degrees when turning on the air conditioner” is user A, “turn on the air conditioner when the MD component is turned on”. Suppose that an operator with a frequent operation pattern of “when the air conditioner is turned on, the temperature is set to 24 degrees” is estimated as the user B and registered in the pattern database 405. In this case, when it is detected that the TV is turned on by the user, it is possible to provide a service such as automatically turning on the air conditioner and setting the temperature to 27 degrees. On the other hand, when it is detected that the MD component is attached by the user, it is possible to provide a service such as automatically turning on the air conditioner and setting the temperature to 24 degrees.

  As described above, in the present embodiment, frequent operation patterns that frequently appear are extracted from the received operation history data and retained, and based on the operation history data and definitive operation information, which frequent operation patterns are determined by which user It can be estimated whether it has been manipulated.

  Therefore, without providing a means for directly identifying a user such as an authentication device, which user has a history that is indispensable for extracting information for use in services from various history information. Can be identified. In particular, even in a home where a plurality of users still use the same device without using an authentication device, it is possible to identify which user the history belongs to, which is useful.

  Further, by specifying a user in this way, it is possible to provide a detailed service corresponding to the user such as outputting information according to the user.

  In this embodiment, the user agent unit 202 and the service concierge unit 203 are provided. However, the service concierge unit 203 is not necessarily required and exists outside the system, and the user agent unit 202 is a service concierge unit. 203 may be configured to output information related to a frequent operation pattern of a user necessary for providing a service. For example, as shown in FIG. 13, in addition to the configuration of the user agent unit 202 shown in FIG. 1, an inquiry receiving unit 407 that receives an inquiry about user operation information from the service concierge unit 203, and a response to the received inquiry are created. The response creation unit 408 and the response output unit 409 that outputs the created response to the service concierge unit 203 may be used.

  In this embodiment, the system configuration is such that the operation history data of the device is received and the service is output via the network. However, the network IF 410 is not an essential component and can be omitted. For example, when the network IF 410 is omitted, the operation history receiving unit 401 receives a remote control operation signal of the device or receives a direct operation from the operation panel, and the service output unit 502 is a display panel or an interface unit provided separately. To the output.

(Second Embodiment)
In the present embodiment, a case will be described in which a frequent operation pattern is specified by a user by a method different from that of the first embodiment.

  FIG. 14 is a block diagram showing a configuration of an operation history utilization system according to the second embodiment of the present invention. This operation history utilization system is the same as the configuration of the first embodiment, except that the pattern classification unit 601 outputs to the service determination unit 501. In addition, the same code | symbol is attached | subjected about the part similar to 1st Embodiment, and detailed description is abbreviate | omitted. In the present embodiment, the method of extracting and registering frequent operation patterns from the received operation history data is the same as the method described in the first embodiment, so the description is omitted here.

  The identification of the user who performed the frequent operation pattern operation in the pattern classification unit 601 will be described. In specifying the user, a decisive operation is defined for each user in advance as in the first embodiment. Here, it is assumed that the definitive operation is defined as shown in FIG. 8 as in the first embodiment.

  FIG. 16 is a flowchart showing a flow of an operation when specifying which user has operated the frequent operation pattern.

  The pattern classifying unit 601 is a frequent operation pattern registered in the pattern database 405 in the operation sequence of the operation history data received by the operation history receiving unit 401, and includes the decisive operation shown in FIG. It is confirmed whether or not a frequent operation pattern exists (step S401). Here, if the corresponding frequent operation pattern does not exist, the pattern classification unit 601 ends the process. On the other hand, when there is a corresponding frequent operation pattern, the pattern classification unit 601 defines a received part of the operation sequence as a user operation corresponding to the corresponding deterministic operation, and registers it in the pattern database 405 ( Step S402).

  FIG. 15 is a diagram illustrating an example of an operation sequence received by the operation history receiving unit 401. Here, it is assumed that the operation history receiving unit 401 has received the operation sequence shown in FIG. In this case, No. 1 shown in FIG. The portion “PowerOn, Opr1, Opr4” described as “1” is the frequent operation pattern No. 1 shown in FIG. Corresponds to 1. This No. Since one frequent operation pattern includes “Opr1”, it is defined as an operation of the user D. This is called a level 1 frequent operation pattern, and it is estimated that the operation is performed by the user D when the certainty factor = 1. In addition, although the frequent operation pattern including the definitive operation is the operation by the corresponding user with the certainty factor 1, it may be an arbitrary certainty factor defined in advance.

  Next, the pattern classification unit 601 indicates whether a frequent operation pattern registered in the pattern database 405 appears adjacent to the above-described level 1 frequent operation pattern in the operation sequence received by the operation history reception unit 401. It is confirmed whether or not (step S403). Here, if the frequent operation pattern does not appear adjacently, the pattern classification unit 601 ends the processing. On the other hand, when a frequent operation pattern appears adjacently, the pattern classification unit 601 defines the operator with certainty according to the appearance location of the frequent operation pattern that appears adjacently, and registers the operator in the pattern database 405. (Step S404). Here, “adjacent” means that adjacent patterns do not sandwich other operations, or adjacent patterns overlap. That is, the pattern classification unit 601 obtains the certainty factor Conf (n) using the following formula (A), and the operator corresponding to the frequent operation pattern of level 1 operates with the obtained certainty factor Conf (n). Define as a thing.

ここで、ｄ（ｉ）は隣接するレベルｉの頻出操作パターンとレベルｉ−１の頻出操作パターンの重なりで、重なっている操作の数を示す。図１５の例ではｄ（２）＝１である。隣接する頻出操作パターンの間に重なりがなくぴったり隣接している場合はｄ（ｉ）＝０、隣接する頻出操作パターンの間に他の操作が挟まれている場合は、挟まれている操作数に−１を乗じたものがｄ（ｉ）と定義する。また、ｆ（ｄ（ｉ））は０≦ｆ（ｄ（ｉ））＜１であり、０以上のｄ（ｉ）の値に対して単調増加する。また０未満のｄ（ｉ）に対してはｆ＝０である。そしてαは定数である。またＣｏｎｆ（１）は最初のレベル１の頻出操作パターンの確信度であり、Ｃｏｎｆ（１）＝１である。

Here, d (i) represents the number of overlapping operations by overlapping the frequent operation pattern of level i and the frequent operation pattern of level i-1. In the example of FIG. 15, d (2) = 1. D (i) = 0 if there is no overlap between adjacent frequent operation patterns, and the number of operations sandwiched when other operations are sandwiched between adjacent frequent operation patterns. Multiplying -1 by -1 is defined as d (i). Further, f (d (i)) is 0 ≦ f (d (i)) <1, and monotonously increases with respect to a value of d (i) of 0 or more. For d (i) less than 0, f = 0. Α is a constant. Conf (1) is the certainty of the first level 1 frequent operation pattern, and Conf (1) = 1.

  For example, in the operation sequence shown in FIG. No. 1 adjacent to the frequent operation pattern 1. 5 frequent operation patterns appear. Here, no. No. 1 frequent operation pattern and No. 1 With the frequent operation pattern of 5, the operation “Opr4” is overlapped on both. This No. The frequent operation pattern of 5 is called a level 2 frequent operation pattern. At this time, the operator of the frequent operation pattern of level 2 is estimated to be the user D who is the operator of the frequent operation pattern of level 1 with the certainty factor Conf (2) obtained by the above formula (A).

  Next, the pattern classification unit 601 indicates whether a frequent operation pattern registered in the pattern database 405 appears adjacent to the above-described level 2 frequent operation pattern in the operation sequence received by the operation history reception unit 401. It is confirmed whether or not (step S405). Here, if the frequent operation pattern does not appear adjacently, the pattern classification unit 601 ends the processing. On the other hand, when a frequent operation pattern appears adjacently, the pattern classification unit 601 performs an operation according to the appearance location of the frequent operation pattern that appears adjacently, as in the case of the frequent operation pattern of level 2 described above. A person is defined with a certainty factor and registered in the pattern database 405 (step S406).

  For example, in the operation sequence shown in FIG. No. 5 adjacent to the frequent operation pattern No. Four frequent operation patterns appear. At this time, d (i) = 0 in the above formula (A). The frequent operation pattern 4 is referred to as a level 3 frequent operation pattern. At this time, the operator of the level 3 frequent operation pattern is estimated that the user D is the operator with the certainty factor Conf (3).

  Thereafter, similarly, whether or not there is a frequent operation pattern of level x is sequentially confirmed, and the process is repeated until no corresponding frequent operation pattern is found.

  The certainty of the operator obtained by the above formula (A) decreases as the level of the frequent operation pattern increases, but the characteristic that the rate of decrease decreases as the overlap between adjacent patterns increases. Show.

  As described above, in the present embodiment, by using the frequent operation patterns registered in the pattern database 405, it is possible to dynamically determine which user has the operation sequence of the operation history data being received and It is possible to respond appropriately to inquiries from.

  Therefore, as in the first embodiment, it is indispensable to extract information for use in services from various history information without providing a means for directly identifying a user such as an authentication device. It is possible to specify which user the history belongs to. In particular, even in a home where a plurality of users still use the same device without using an authentication device, it is possible to identify which user the history belongs to, which is useful.

  Further, by specifying a user in this way, it is possible to provide a detailed service corresponding to the user such as outputting information according to the user.

  Note that this embodiment is an example, and it is estimated as an operator of a frequent operation pattern including a definitive operation, which user is the operation sequence, and the certainty level is a decisive operation. What is necessary is just to be determined according to the overlapping degree with the frequent operation pattern chained from the frequent operation pattern including or the distance from the frequent operation pattern.

  Further, in the present embodiment, a case has been described in which it is estimated which user the operation sequence of the operation history data being received is from. However, the operation sequence of the operation history data registered in the operation history database 402 is described. However, it can be estimated which user is the same.

(Third embodiment)
FIG. 17 is a block diagram showing a configuration of an operation history utilization system according to the third embodiment of the present invention. In this operation history utilization system, the service concierge unit 203 includes a service priority determination unit 602 in addition to the configuration of the first embodiment, and the other configurations are the same as those of the first embodiment. In the present embodiment, frequent operation patterns are extracted from the received user operation history data and stored in the pattern database 405. Further, by using the stored frequent operation pattern, the user's next operation is predicted from the operation sequence of the operation history data currently received, and services such as device control and information provision are provided accordingly. Since the above-described frequent operation pattern extraction method and service provision method are the same as those in the first and second embodiments, description thereof is omitted here.

  The service priority determination unit 602 selects a service with a high priority when it is impossible to provide all the services provided by the service determination unit 501 at the same time. In the service priority determination unit 602, priorities for users and operations are defined in advance. FIG. 18 is a diagram showing an example of defined priorities, (a) a priority for each user, and (b) a priority for each operation.

  Next, an operation after the service determination unit 501 estimates a service to be provided to the user in the operation history utilization system configured as described above will be described.

  FIG. 19 is a flowchart showing an operation flow when providing the estimated service.

  The service determination unit 501 uses the frequent operation patterns stored in the pattern database 405 to predict the next operation of the user from the currently received operation sequence, and provides device control, information provision, and the like accordingly. A service to be performed is estimated (step S501). Next, the service priority determination unit 602 confirms how many services estimated by the service determination unit 501 exist, and determines whether or not these services can be provided simultaneously (step). S502). Here, when the services estimated by the service determination unit 501 can be provided simultaneously, the service priority determination unit 602 notifies the service output unit 502 to that effect. Then, the service output unit 502 provides the service estimated by the service determination unit 501 (step S503).

  On the other hand, when the services estimated by the service determination unit 501 cannot be simultaneously provided, the service priority determination unit 602 is based on the predefined priority order as shown in FIG. A service having a high priority is selected from the services estimated by the service determination unit 501, and the selected service is notified to the service output unit 502. The service output unit 502 provides the service selected by the service priority determination unit 602 (step S504).

  For example, it is assumed that the relationship between the frequent operation pattern and the operator as shown in FIG. 10 is extracted. Here, when the operation history reception unit 401 receives “PowerOn, Opr1”, the service determination unit 501 selects No. 1 among the frequent operation patterns shown in FIG. The pattern monitoring unit 406 notifies that it partially matches one frequent operation pattern. In this case, the service determination unit 501 determines the No. 5 based on the relationship between the frequent operation patterns registered in the pattern database 405 and the operator. In one frequent operation pattern, it is recognized that the next operation is “Opr4” and that the operator is the user D. The service determining unit 501 estimates the service that provides the next expected operation of “Opr4”. In this example, since there is one estimated service, the service output determining unit 602 indicates that the service priority determining unit 602 can provide the operation of “Opr4” estimated by the service determining unit 501. To notify. The service output unit 502 outputs device control according to the operation of “Opr4” as a service.

  As another example, when the operation history receiving unit 401 receives “Opr8, Opr2”, for example, the No. shown in FIG. 3 and no. 6 partly matches the frequent operation pattern. The service determination unit 501 is configured as No. In the frequent operation pattern of No. 3, the next operation is “Opr3” and the operator is the user B. In the frequent operation pattern 6, it is recognized that the next operation is “Opr5” and the operator is the user A. Therefore, the next expected operations are the two operations “Opr3” and “Opr5”. Here, the service priority determination unit 602 determines whether the operations “Opr3” and “Opr5” can be provided simultaneously.

  When these operations are operations for different devices, for example, when the operation of “Opr3” is an operation for increasing the volume of the television, and the operation of “Opr5” is an operation for adjusting the air volume of the air conditioner, these controls are performed. Can be carried out simultaneously. Therefore, the service priority determination unit 602 notifies the service output unit 502 that the operations “Opr3” and “Opr5” estimated by the service determination unit 501 can be provided. The service output unit 502 outputs device control corresponding to the operation of “Opr3” and “Opr5” as a service.

  On the other hand, when these “Opr3” and “Opr5” operations are operations on the same device, for example, the “Opr3” operation sets the TV channel to “4”, and the “Opr5” operation sets the TV channel. When the operation is “8”, it is impossible to perform these controls simultaneously. In this case, the operation of “Opr3” is No. 3 is estimated from the frequent operation pattern No. 3, the operator is the user B, and the operation of “Opr5” is the frequent operation pattern No. Therefore, the operator is estimated to be the user A. Here, in the priority order shown in FIG. 18A, user B has higher priority among user A and user B. Therefore, the service priority determination unit 602 determines the service that provides the operation “Opr3” predicted based on the frequent operation pattern of the user B. Then, the service output unit 502 outputs device control corresponding to the operation of “Opr3” as a service.

  As described above, in the present embodiment, even when there are a plurality of service candidates to be provided and it is impossible to provide them simultaneously, it is possible to select and provide an appropriate service. Become.

  Therefore, even in a situation where there are physical restrictions such as the same equipment and the same space as in a home where there are multiple users in the same space, it is all that exists there It is possible to provide the best service possible for the user.

  In the present embodiment, the service to be provided is determined based on the priority order between users as shown in FIG. 18A. For example, priority between operations as shown in FIG. It may be determined based on the ranking.

  In the present embodiment, services that were not selected from among a plurality of service candidates were rejected. However, an alternative operation list as shown in FIG. 18C is prepared in advance and selected by referring to this list. If there is an alternative service that can be provided simultaneously with the service, it may be provided. For example, according to the alternative operation list of FIG. 18C for the operation of “Opr5” that was not selected in the above example, it can be seen that the operation of “Opr13” is an alternative operation. At this time, if the operation of “Opr13” is a combination of services that can be provided simultaneously with the operation of “Opr3”, these two types of services can be provided.

  Further, in the present embodiment, as shown in FIG. 17, the user agent unit 202 and the service concierge unit 203 are provided. However, as a system configuration, the service concierge unit 203 may be provided as a minimum, The functions of the user agent unit 202 such as frequent operation pattern extraction management and operator estimation may exist outside the communicable system.

(Fourth embodiment)
In the present embodiment, a case will be described in which which service is to be provided for a plurality of services that cannot be provided at the same time by a method different from the third embodiment.

  FIG. 20 is a block diagram showing a configuration of an operation history utilization system according to the fourth embodiment of the present invention. This operation history utilization system includes a service distribution determination unit 603 instead of the service priority determination unit 602 having the configuration of the third embodiment, and other configurations are the same as those of the third embodiment. In addition, the same code | symbol is attached | subjected about the part similar to 3rd Embodiment, and detailed description is abbreviate | omitted.

  The service allocation determination unit 603 allocates the services to be provided probabilistically when it is impossible to provide all the services provided by the service determination unit 501 at the same time. In addition, the service distribution determination unit 603 defines distribution probabilities for users and operations in advance. FIG. 21 is a diagram showing an example of defined distribution probabilities, (a) a distribution probability for each user and (b) a distribution probability for each operation.

  Next, an operation after the service determination unit 501 estimates a service to be provided to the user in the operation history utilization system configured as described above will be described.

  FIG. 22 is a flowchart showing an operation flow when providing the estimated service.

  The service determination unit 501 uses the frequent operation patterns stored in the pattern database 405 to predict the next operation of the user from the currently received operation sequence, and provides device control, information provision, and the like accordingly. A service to be performed is estimated (step S601). Next, the service allocation determination unit 603 checks how many services estimated by the service determination unit 501 exist, and determines whether or not these services can be provided simultaneously (step S602). ). Here, when the services estimated by the service determination unit 501 can be simultaneously provided, the service distribution determination unit 603 notifies the service output unit 502 to that effect. Then, the service output unit 502 provides the service estimated by the service determination unit 501 (step S603).

  On the other hand, when the services estimated by the service determining unit 501 cannot be simultaneously provided, the service allocation determining unit 603 determines the service based on the predefined allocation probability as shown in FIG. A service is selected from the services estimated by the determination unit 501, and the selected service is notified to the service output unit 502. Then, the service output unit 502 provides the service selected by the service distribution determination unit 603 (step S604).

  For example, assume that the relationship between the frequent operation pattern and the operator as shown in FIG. 10 is extracted as in the third embodiment. Here, when the operation history receiving unit 401 receives “PowerOn, Opr1”, the service determination unit 501 is estimated as a service that provides the next expected operation of “Opr4” as in the third embodiment. To do. In this example, since there are one estimated service, the service distribution determination unit 603 notifies the service output unit 502 that the operation of “Opr4” estimated by the service determination unit 501 can be provided. Notice. The service output unit 502 outputs device control according to the operation of “Opr4” as a service.

  As another example, for example, when the operation history receiving unit 401 receives “Opr8, Opr2”, similarly to the third embodiment, the operations expected next are operations of “Opr3” and “Opr5”. There are two ways. Here, the service allocation determination unit 603 determines whether the operations “Opr3” and “Opr5” can be provided simultaneously.

  When these operations are operations for different devices, the service distribution determination unit 603 indicates that the service output unit 502 can provide the operations “Opr3” and “Opr5” estimated by the service determination unit 501. To notify. The service output unit 502 outputs device control corresponding to the operation of “Opr3” and “Opr5” as a service.

  On the other hand, when the operations of “Opr3” and “Opr5” are operations for the same device, the operation of “Opr3” is the same as that of the third embodiment. 3 is estimated from the frequent operation pattern No. 3, the operator is the user B, and the operation of “Opr5” is the frequent operation pattern No. Therefore, the operator is estimated to be the user A. Here, in the distribution probability shown in FIG. 21A, the distribution probability of the user A is “0.2”, and the distribution probability of the user B is “0.1”. Accordingly, the service allocation determination unit 603 determines which service is provided at this probability ratio, that is, at a ratio of “2: 1”. In other words, in this case, it is determined as a service that provides the operation of “Opr5” by the user A twice every three times and the operation of the “Opr3” by the user B once every three times. Then, the service output unit 502 outputs device control corresponding to the operation selected by the service distribution determination unit 603 as a service. Note that the operation of determining either at a ratio of “2: 1” in this way is, for example, when a random number (for example, 0.000 to 0.999) is generated and “0.000 to 0.666” is generated. Can be determined as user A, and in the case of “0.667 to 0.999”, user B can be determined.

  As described above, in the present embodiment, even when there are a plurality of service candidates to be provided and it is impossible to provide them simultaneously, it is possible to select and provide an appropriate service. .

  Therefore, as in the third embodiment, it is a situation that is subject to physical restrictions such as the same equipment and the same space as in a home where a situation where a plurality of users exist in the same space is daily. However, it is possible to provide the best possible service for all users existing there.

  In this embodiment, it is determined as a service to be provided based on the ratio of distribution probability between users, but is not limited to this, and is provided based on the ratio of distribution probability between users. You may change the service on the way. For example, control is performed such that an operation of “Opr3” is provided for 2 hours, and then an operation of “Opr5” is provided for 1 hour.

  In this embodiment, the distribution probability is defined for each user and operation in advance. However, the distribution probability is not limited to this. For example, the distribution probability is calculated based on the history of usage records used by the user in the past. It may be determined or the distribution probability may be dynamically changed. Further, the history such as the user's usage record used at this time may be information estimated by the method described in the first and second embodiments, for example, a remote controller with an ID. Information obtained using a device capable of specifying an operator may be used.

  In this embodiment, the service to be provided is determined based on the distribution probability between users as shown in FIG. 21A. For example, the distribution between operations as shown in FIG. You may determine based on a probability.

  In the present embodiment, as shown in FIG. 20, the user agent unit 202 and the service concierge unit 203 are provided. However, the system configuration may include the service concierge unit 203 as a minimum, The functions of the user agent unit 202 such as frequent operation pattern extraction management and operator estimation may exist outside the communicable system.

(Fifth embodiment)
FIG. 23 is a block diagram showing a configuration of an operation history utilization system according to the fifth embodiment of the present invention. This operation history utilization system includes an operation confirmation unit 604 instead of the pattern classification unit 601 of the configuration of the first embodiment, and the other configuration is the same as that of the first embodiment. In the present embodiment, frequent operation patterns are extracted from the received user operation history data and stored in the pattern database 405. Since this frequent operation pattern extraction method is the same as that of the first embodiment, the description thereof is omitted here.

  The operation confirmation unit 604 confirms and corrects operations that are expected to be operations that are not intended by the user.

Next, the operation in the operation history utilization system configured as described above will be described.
FIG. 24 is a flowchart showing an operation flow when the operation history data is received.

  The operation history receiving unit 401 confirms whether or not a sequence of operation history data has been received (step S701). Here, when the operation history data is not received, the operation history receiving unit 401 continues to check whether or not the operation history data has been received at a predetermined cycle. On the other hand, when the operation history data is received, the operation history reception unit 401 notifies the operation history registration unit 403 that the operation history data has been received. The operation history registration unit 403 extracts the operation history data received by the operation history reception unit 401 and registers it in the operation history database 402. At this time, the pattern monitoring unit 406 compares whether the operation sequence of the registered operation history data matches the frequent operation pattern stored in the pattern database 405 (step S702).

  For example, assume that the operation history receiving unit 401 receives operation history data “Opr7, Opr6, Opr9” when the frequent operation pattern shown in FIG. 7 is stored in the pattern database 405. In this case, the received operation history data is No. shown in FIG. Only the frequent operation pattern 4 and the part “Opr7, Opr6” partially match, and the third operation is different. In such a case, the received operation history data partially matches the frequent operation pattern.

  As a result of such comparison, when the received operation history data completely matches the frequent operation pattern or does not match at all, the pattern monitoring unit 406 ends the process. On the other hand, when the received operation history data partially matches the frequent operation pattern, the operation confirming unit 604 selects an operation that does not match among the received operation history data partially matched and the frequent operation pattern. A warning is presented to the user who is the operator and the device on which the operation has been performed (step S703).

  In the above example, the operation “Opr9” in the received operation history data is an operation different from the frequent operation pattern. Therefore, the operation confirming unit 604 confirms that the operation of “Opr9” is performed on the display interface of the device on which the operation has been performed. Is output to display a warning such as In addition, you may output to not only the apparatus with which operation was performed at this time but another apparatus which has display means, such as a television, for example. The warning is not limited to display, and other means such as a warning sound and vibration may be used.

  The above operation is performed from the following two viewpoints. First, since a frequent operation pattern is a regular sequence of operations that are observed many times, an operation that is “different” for an operation sequence that is only partially different from the frequent operation pattern. It is a viewpoint that there is a high possibility that it is an erroneous operation or an unintended operation. Second, in the case where the pattern database 405 is appropriately updated for use in service prediction based on frequent operation patterns, it is necessary to register operations that are not intended by the user as much as possible. is there.

  In the present embodiment, of the received operation history data partially matched and the frequent operation pattern, an operation that does not match, that is, an operation that is expected to be an erroneous operation (an operation of “Opr9” in the above example). Although being executed (FIG. 25, operation 1), this operation may be interrupted (FIG. 25, operation 3). For example, if the operation of “Opr9” is an operation of “increase the volume”, the operation confirming unit 604 outputs a control command that lowers the volume once increased to the original volume level again. If the “increase volume” operation is an operation intended by the user, even if the operation confirmation unit 604 automatically decreases the volume, the operation to increase the volume is executed again. Can be judged more clearly.

  Further, the operation confirmation unit 604 may suspend the operation expected to be an erroneous operation instead of the warning as described above and execute the operation stored in the pattern database 405 (FIG. 25, operation 2). . In this case, for example, in the above example, the operation of “Opr9” is interrupted, and No. 3 shown in FIG. The operation “Opr10” is executed from the frequent operation pattern 4.

  As described above, in the present embodiment, an operation that is partly deviated from the frequent operation pattern is detected, a warning is given to the operator about the operation, the operation is interrupted, or an alternative operation expected from the frequent operation pattern is performed. By executing the operation, it is possible to prevent an operator's erroneous operation or an unintended operation. Therefore, it is possible to improve the reliability of the history information that is important when analyzing the user's preference and state based on the history information and using it for the service.

  In the present embodiment, the user agent unit 202 and the service concierge unit 203 are provided as shown in FIG. 23, but the user agent unit 202 may be provided as a minimum system configuration.

(Sixth embodiment)
The present embodiment is an example in which frequent operation patterns are extracted from operation history data of a plurality of devices connected via a network, and a service based on the extracted operation patterns is provided.

  FIG. 26 is a block diagram showing an overall configuration of an operation history utilization system according to the sixth embodiment of the present invention.

  As shown in FIG. 26, the operation history utilization system includes a digital TV (hereinafter simply referred to as a TV or TV) 101a, a video (Video) 101b, an air conditioner (Air Conditioner) 101c, a lighting fixture (Light) 101d, and an MD (MiniDisc). A component home appliance 101 such as a component 101e and a PC (Personal Computer) 101f, a server home appliance (user server) 102 such as a home server or a home gateway, and an application server 103 permitted to access the server home appliance 102; A LAN (Local Area Network) 104 configured by a wired or wireless network, a WAN (Wide Area Network: for example, the Internet) 105, and a router 106 for connecting the LAN 104 to the WAN 105.

  Here, the server home appliance 102 is always energized, and access from the WAN 105 side is permitted according to a predetermined condition. FIG. 26 shows one of a plurality of application servers 103 that provide information services. In addition, the router 106 has a modem function corresponding to the necessity, for example, when using Asymmetric Digital Subscriber Line (ADSL) or Fiber To The Home (FTTH) to access the WAN 105.

  FIG. 27 is a block diagram showing main functional modules (programs) installed in the terminal home appliance 101, the server home appliance 102, and the application server 103 in FIG.

  Each of the terminal home appliances 101 includes a device agent 201 that records device operation history data triggered by a user operation. The server home appliance 102 includes a user agent 202 that receives device operation history data transmitted from the device agent 201 at a predetermined timing and centrally manages the device operation history data. Each of the application servers 103 includes a service concierge 203 that provides a service using device operation history data provided by the user agent 202.

  The server home appliance 102 may include its own device agent 201 in addition to the user agent 202, such as in the case of a home server having a recording function. When there are a plurality of home appliances that are candidates for the server home appliance 102, one device is selected by a predetermined method.

FIG. 28 is a block diagram showing the configuration of the device agent 201.
The device agent 201 includes an operation input unit 301, an operation input interpretation unit 302, a device control unit 303, an operation history storage unit 304, an operation history recording unit 305, an operation history transmission unit 306, and a network IF (network interface) 308. Yes.

  The operation input unit 301 receives an operation from a user input via a remote controller or the like. The operation input interpretation unit 302 interprets the operation input from the operation input unit 301 and determines the operation state of the device. The device control unit 303 controls the device in accordance with the analysis result of the operation input analysis unit 302. The operation history storage unit 304 stores the operation date and time and the operation content as a set for the operation of the device. The operation history recording unit 305 adds the device operation interpreted by the operation input interpretation unit 302 to the operation history storage unit 304. The operation history transmission unit 306 transmits the operation history data stored in the operation history storage unit 304 to the user agent 202 via the network IF 308 at a predetermined timing.

Next, the operation of the device agent 201 configured as described above will be described.
The device agent 201 first interprets the operation received by the operation input unit 301 from the remote controller operated by the user by the operation input interpretation unit 302. For example, it is interpreted that the temperature setting of the air conditioner 101c is 25 degrees and the temperature setting is set to 27 degrees by the user pressing the set temperature up button of the remote controller twice. It is assumed that the input operation from other than the remote controller is similarly processed. As a result, the device control unit 303 executes an operation corresponding to the operation input. At the same time, the operation history transmission unit 306 records the operation history data in the operation history storage unit 304.

  FIG. 29 is a diagram illustrating an example of operation history data to be left in a history defined for each device. For example, in the case of an air conditioner, power-on (PowerOn), power-off (PowerOff), temperature setting, air volume setting, stoppage, reservation, and the like are stored as operation history data.

  FIG. 30 is a diagram illustrating an example of operation history data. The operation history data shown in FIG. 30A is an example of air conditioner operation history data. The operation date and time when the operation was performed, the device type “Air Conditioner”, and the operation type identifier “ “Set” and “27” are described as arguments, indicating that the air conditioner was operated at a temperature of 27 degrees on the operation date. Further, the operation history data shown in FIG. 30B is an example of the operation history data of the television, and indicates that a program whose ID number is uniquely specified by XXXXXX at the operation date / time is viewed. In addition, a history start date storage unit is further provided, the start date / time of operation at 25 degrees so far is stored in advance, and the operation start date / time and operation end date / time are set as the operation end date / time at 25 degrees. It may be recorded in the operation history storage unit 304. In this case, the current date / time is stored in the history start date / time storage unit as the operation start date / time at 27 degrees. At this time, the operation end date / time item may be an elapsed time from the operation start date / time.

  The operation history data recorded in the operation history storage unit 304 as described above is transmitted to the user agent 202 by the operation history transmission unit 306 via the network IF 308 and the network.

  FIG. 31 is a diagram illustrating an example of operation history data received by the user agent 202.

  The operation history data received by the user agent 202 includes, for example, an operation date and time, a device type, an operation type identifier, and an argument as shown in FIG. 31, and various devices and various operation histories are arranged in chronological order. Stored in the history database 402.

  From the operation history data, the episode data shown in FIG. 32 can be generated by the procedure shown in FIG. 5 already described in the first embodiment, and the frequent operation pattern shown in FIG. 33 can be further extracted therefrom. is there.

  In addition, the user agent 202 and the service concierge 203 are configured as shown in FIG. 1 in the first embodiment, and the operation procedure shown in FIGS. 5 and 9 is used to extract frequent operation patterns of operation history data. By using the operation history data and the decisive operation information as shown in FIG. 34 without using any special user identification means, it is possible to estimate which user the operation is. Further, it is possible to provide services such as device control and information provision based on operation prediction using a frequent operation pattern to which information related to these operators is given.

  Further, the user agent 202 and the service concierge 203 are configured as shown in FIG. 14 in the second embodiment, and the operation procedure shown in FIG. By using the operation history data and the definitive operation information without using the user identification means, it is possible to estimate which user is performing the operation. Further, it is possible to provide services such as device control and information provision based on operation prediction using a frequent operation pattern to which information related to these operators is given.

  Further, the user agent 202 and the service concierge 203 are configured as shown in FIG. 17 in the third embodiment, and the operation procedure shown in FIG. Provided after determining services such as device control and information provision based on operation prediction based on the defined priority for each user and operation even when provision of multiple services that cannot be performed is predicted It becomes possible to do.

  Further, the user agent 202 and the service concierge 203 are configured as shown in FIG. 20 in the fourth embodiment, and the operation procedure shown in FIG. Provided after determining services such as device control and information provision based on operation prediction based on the probability distribution for each defined user and operation even when provision of multiple services that cannot be performed is predicted It becomes possible to do.

  Further, the user agent 202 and the service concierge 203 are configured as shown in FIG. 23 in the fifth embodiment, and the operation procedure shown in FIG. For operations that deviate from the pattern, it is possible to prevent an operator's erroneous operation or unintended operation by issuing a warning to the operator, interrupting the operation, or executing an alternative operation expected from the frequent operation pattern. It becomes.

  In this embodiment, the user agent 202 and the service concierge 203 are configured independently as shown in FIG. 27, but the present invention is not limited to this. For example, as shown in FIG. 35, the user agent 202 and the service concierge 203 are separated at the function level, but the same apparatus may be provided with these functions.

  In this embodiment, the terminal home appliance 101 is exemplified as a device used in the home. However, the present invention is not limited to this, and it is possible to collect information by connecting to a network. For example, it may be a device mounted on a car or the like, a personal digital assistant (PDA), a mobile phone, or the like.

  In addition, the above embodiments are not only implemented individually, but also in combination with other embodiments, for example, the first embodiment and the third embodiment are combined. Is also possible.

  Further, when only one terminal home appliance 101 exists in each of the above-described embodiments, the terminal home appliance 101 is provided with both the device agent 201 and the user agent 202 to form an integrated device, or a service concierge. It is also possible to implement it as an integrated apparatus having three members with 203 added.

  As is clear from the above description, according to the operation history utilization system according to the present invention, operation history data of individual devices are collected, the operation history data is centrally managed, and the situation at that time is Combined services and services that better reflect the characteristics of each individual's life.

  Therefore, today's practical value is very high in which a situation-dependent service that provides a service according to the situation at that time using information obtained by analyzing user preferences and states based on various history information is extremely high.

  The present invention is an operation history utilization system comprising a communication system in which various devices, a user server device, an application server device, and the like are connected via a transmission line, in particular, a digital TV, a video, an air conditioner, a lighting fixture, an MD component, and a PC. It can be used as an operation history utilization system comprising a communication system composed of application home appliances such as terminal home appliances such as home servers and home gateways and computers and application server devices such as computers.

FIG. 1 is a block diagram showing a configuration of an operation history utilization system according to the first embodiment of the present invention. FIG. 2 is a flowchart showing an operation flow when registering operation history data in the first embodiment. FIG. 3 is a diagram illustrating an example of an operation to be stored as a history. FIG. 4 is a diagram illustrating an example of data recorded in the operation history database. FIG. 5 is a flowchart showing the flow of operations when extracting frequent operation patterns in the first embodiment. FIG. 6 is a diagram illustrating an example of episode data from the operation history data according to the first embodiment. FIG. 7 is a diagram illustrating an example of a frequent operation pattern extracted from operation history data according to the first embodiment. FIG. 8 is a diagram illustrating an example of the definition of the decisive action used for user identification according to the first embodiment. FIG. 9 is a flowchart illustrating a flow of an operation when specifying which user has operated the frequent operation pattern according to the first embodiment. FIG. 10 is a diagram illustrating an example of a frequent operation pattern in which a user is specified in the first embodiment. FIG. 11 is a diagram illustrating another example of the definition of the decisive action used for user identification according to the first embodiment. FIG. 12 is a block diagram showing the configuration of FIG. 1 including devices in the operation history utilization system according to the first embodiment. FIG. 13 is a block diagram illustrating another configuration of the operation history utilization system according to the first embodiment. FIG. 14 is a block diagram showing a configuration of an operation history utilization system according to the second embodiment of the present invention. FIG. 15 is a diagram illustrating an example of received operation history data according to the second embodiment. FIG. 16 is a flowchart illustrating a flow of an operation when specifying which user has operated the frequent operation pattern according to the second embodiment. FIG. 17 is a block diagram showing a configuration of an operation history utilization system according to the third embodiment of the present invention. FIG. 18 is a diagram illustrating an example of the priority order of service provision in the third embodiment, and illustrates (a) a priority order for each user and (b) a priority order for each operation. FIG. 19 is a flowchart illustrating an operation flow when providing the estimated service according to the third embodiment. FIG. 20 is a block diagram showing a configuration of an operation history utilization system according to the fourth embodiment of the present invention. FIG. 21 is a diagram illustrating an example of a distribution probability of service distribution in the fourth embodiment, and illustrates (a) a distribution probability for each user and (b) a distribution probability for each operation. FIG. 22 is a flowchart showing an operation flow when providing the estimated service according to the fourth embodiment. FIG. 23 is a block diagram showing a configuration of an operation history utilization system according to the fifth embodiment of the present invention. FIG. 24 is a flowchart showing an operation flow when the operation history data is received in the fifth embodiment. FIG. 25 is a diagram illustrating an operation example when an operation that is expected to be an erroneous operation in the fifth embodiment occurs. FIG. 26 is a conceptual diagram showing an overall configuration of an operation history utilization system according to the sixth embodiment of the present invention. FIG. 27 is a block diagram showing an arrangement of main functional modules constituting the system of the sixth embodiment. FIG. 28 is a block diagram illustrating a configuration of a device agent according to the sixth embodiment. FIG. 29 is a diagram illustrating an example of an operation to be stored as a history in the device agent according to the sixth embodiment. FIG. 30 is a diagram illustrating an example of operation history data transmitted from the device agent according to the sixth embodiment. FIG. 31 is a diagram illustrating an example of operation history data recorded in the operation history database. FIG. 32 is a diagram illustrating an example of episode data from the operation history data according to the sixth embodiment. FIG. 33 is a diagram illustrating an example of a frequent operation pattern extracted from operation history data according to the sixth embodiment. FIG. 34 is a diagram illustrating an example of a definition of a decisive action used for user identification in the sixth embodiment. FIG. 35 is a block diagram illustrating another configuration of the operation history utilization system according to the sixth embodiment.

Claims (3)

An operation history management device that manages operation history data transmitted from at least one device, and a service providing device that performs operation control of the device according to a user based on operation history data managed by the operation history management device Connected operation history utilization system,
The operation history management device includes:
The operation history data transmitted from the device and registered in the operation history database is sorted by operation date and time, and the operation history data in which the interval between adjacent data is equal to or less than a predetermined time is classified as the same group, Pattern extraction means for extracting a partially matching operation sequence and registering it as a frequent operation pattern in a pattern database;
A user who defines an operator with a certainty in advance and defines a deterministic operation in advance, extracts a frequent operation pattern including the deterministic operation from the frequent operation patterns registered in the pattern database, and corresponds to the decisive operation The operator's certainty is registered in the pattern database, frequent operation patterns that occur following the frequent operation pattern having the operator's certainty are extracted, and the operator's certainty is determined according to the frequency of occurrence. The frequency is registered in the pattern database, and frequent operation patterns that occur following the frequent operation patterns having the certainty of the operator are extracted until the corresponding frequent operation patterns disappear. By registering the certainty factor in the pattern database recursively, the operator's confirmation can be confirmed. And pattern classification means for registering degrees in the pattern database,
Pattern monitoring means for searching for an operation sequence of operation history data newly transmitted from the device for finding a partial match with the frequent operation pattern of the pattern database,
The service providing apparatus includes:
When an operation sequence of the operation history data newly transmitted from the device is found by the pattern monitoring unit, when the part that matches the frequent operation pattern is found, the frequent operation pattern and the operator of the frequent operation pattern A service decision for determining whether to perform operation control of the device according to an operation subsequent to the matched operation sequence included in the frequent operation pattern or to output information according to a user who is an operator based on An operation history utilization system comprising: means. An operation history management device that manages operation history data transmitted from at least one device, and a service providing device that performs operation control of the device according to a user based on operation history data managed by the operation history management device Connected operation history utilization system,
The operation history management device includes:
The operation history data transmitted from the device and registered in the operation history database is sorted by operation date and time, and the operation history data in which the interval between adjacent data is equal to or less than a predetermined time is classified as the same group, Pattern extraction means for extracting a partially matching operation sequence and registering it as a frequent operation pattern in a pattern database;
Pattern monitoring means for searching for an operation sequence of operation history data newly transmitted from the device, which partially matches the frequent operation pattern of the pattern database;
An operator is defined in advance in association with a certainty level with certainty, and a frequent operation pattern including the deterministic operation is extracted from the operation history data registered in the operation history database, and the decisive operation is supported. Registering the operator's certainty factor in the pattern database as a user, extracting a frequent operation pattern having the operator's certainty factor from the operation history data newly transmitted from the device, and continuing to the frequent operation pattern Frequent operations patterns that appear without intervening other operations or that some operations overlap, and the operator's certainty is registered in the pattern database according to the degree of adjacency that is the degree of overlap. Until the corresponding frequent operation pattern no longer appears, recursively sandwiches other operations following the frequent operation pattern. Frequent operations patterns that appear in a part of or overlapped with some operations and register the operator's certainty in the pattern database according to the degree of adjacency that is the degree of overlap. Pattern classification means for registering the certainty factor in the pattern database;
The service providing apparatus includes:
When an operation sequence of the operation history data newly transmitted from the device is found by the pattern monitoring unit, when the part that matches the frequent operation pattern is found, the frequent operation pattern and the operator of the frequent operation pattern A service decision for determining whether to perform operation control of the device according to an operation subsequent to the matched operation sequence included in the frequent operation pattern or to output information according to a user who is an operator based on An operation history utilization system comprising: means. At least one device, an operation history management device that manages operation history data transmitted from the device, and a service that performs operation control of the device according to a user based on operation history data managed by the operation history management device An operation history utilization system connected to a providing device,
The device transmits operation history data in which operation details for operating the device are associated with information on the operation date and time,
The operation history management device includes:
The operation history data transmitted from the device and registered in the operation history database is sorted by operation date and time, and the operation history data in which the interval between adjacent data is equal to or less than a predetermined time is classified as the same group, Pattern extraction means for extracting a partially matching operation sequence and registering it as a frequent operation pattern in a pattern database;
An operator is defined in advance in association with a certainty level with certainty, and a frequent operation pattern including the deterministic operation is extracted from the operation history data registered in the operation history database. Pattern classification means for registering the certainty factor of each operator in the pattern database as being a user corresponding to the decisive operation,
Pattern monitoring means for searching for an operation sequence of operation history data newly transmitted from the device for finding a partial match with the frequent operation pattern of the pattern database,
The service providing apparatus includes:
When an operation sequence of the operation history data newly transmitted from the device is found by the pattern monitoring unit, when the part that matches the frequent operation pattern is found, the frequent operation pattern and the operator of the frequent operation pattern Based on the service determination, it is determined to control the operation of the device according to the subsequent operation of the matched operation sequence included in the frequent operation pattern or to output information according to the user who is the operator An operation history utilization system comprising: means.

Images