JP2008234041A - Context sharing system, context sharing method, client and server - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a context sharing system easy in control and setting of privacy policy in context sharing. <P>SOLUTION: In this context sharing system, context information is configured by a node ID, a property, and a value according to a triple of an RDF (Resource Description Framework), and node information is configured by an ID, a class and a value. The privacy policy is configured by a node condition, a target user, and an access right. When the produced context information or node information matches the node condition of the policy, a client copies the access right and the target user of the policy to access information, and performs transmission together with the produced information. A server stores the transmitted information. When a context of an arbitrary user is requested from a certain user, the request source user returns the context information when the request source user is permitted by the access right related to the requested context of a request destination user. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a context sharing system, a context sharing method, a client, and a server. For example, the present invention can be applied to a method for controlling a privacy policy when sharing a context including a personal profile, a keyword, an action history, a current situation, and the like. It is.

  Conventionally, as a method for controlling a privacy policy when sharing a context, there is a method disclosed in paragraph “0009” of Patent Document 1. This method allows specific keywords, noun phrases, and contacts that should not be sent to the server computer to be identified in a privacy preference (which defines information that can be sent to the server). It is a thing.

There is also a method disclosed in paragraphs “0009” and “0010” of Patent Document 2. This method is a method of selecting information elements that satisfy preset privacy requirements corresponding to the user, and generating presence information based on the selected information elements, and information communication used by the user. When fetching detection elements relating to the state of the terminal, this is a method of selectively fetching only detection elements corresponding to privacy requirements set in advance corresponding to the user and generating presence information based on this.
JP 2006-92540 A JP-A-2005-202789

  However, in the method of the above-mentioned patent document 1, as the location information of the action history, the location described in the schedule is disclosed in the company, the location described in the email is disclosed only to the recipient of the email, and the information outside the working hours is not disclosed. In order to be able to identify a complicated privacy policy such as that by a specific keyword or noun phrase, there is a problem that it takes time to set it.

  In addition, the method of Patent Document 2 does not describe a specific example of how to describe privacy requirements, and how to extract information elements and presence information that satisfy the privacy requirements. No specific way to describe a complex privacy policy is disclosed.

  Therefore, there is a demand for a context sharing system, a context sharing method, a client, and a server that can easily set and control a privacy policy for sharing a context and that have a clear description of the privacy policy.

  According to a first aspect of the present invention, one or more clients that acquire context information related to individual users and intermittently transmit the information to a server, and context information of an arbitrary user requested by an arbitrary client, under a predetermined condition In a context sharing system having a server that sends a reply, (1) each of the clients has (1-1) (1-1-1) a node ID, one or more property / value pair information, and a processing flag. Including context information having 0 or 1 or more records, and (1-1-2) 0 or 1 or more records including ID, class and value corresponding to the node ID or property value of the context information. Node information, (1-1-3) ID, and a target user that describes a user or group that permits or does not permit reading. A first context holding means for holding access information having 0 or 1 or more records including an access right describing permission / non-permission of reading, and (1-2) a target class or keyword Manages privacy policy information with one or more records, including node conditions that describe, users who are allowed to read, or users who are not allowed to be grouped, target users that are described, and read rights that are permitted or denied. (1-3) collecting the context and creating the context information and the node information, matching the created context information and the node information with the privacy policy information, Privacy policy information node conditions and container If the target information or node information matches, the target user and access of the privacy policy information are copied to the access information, and then the context information, the node information, and the access information are transmitted to the server, and the user First context management means for transmitting a context request to the server in response to the input of (2) the server, (2-1) the context information transmitted from the client, and the node information. And a second context holding unit that holds the access information for each user, and (2-2) causing the second context holding unit to hold the transmitted context information, the node information, and the access information. Along with any client's content from any client user. A second context management means for returning the context information when the requesting user is permitted to read with the access right related to the requested context of the requested user when the request is requested It is characterized by providing.

  According to a second aspect of the present invention, there is provided one or more clients that acquire context information relating to individual users and intermittently transmit the information to a server, and any user context information requested by any client under a predetermined condition. In the context sharing system having a server for returning, (1) each of the clients receives first context management means for managing transmission of context, first context holding means for holding context information, and privacy policy information. (1-1) The context holding means holds context information, node information, and access information. (1-1-1) The context information includes a node ID, 0 or more including one or more property and value pair information and processing flags (1-1-2) The node information includes zero or one or more records including an ID, a class, and a value corresponding to a node ID or property value of the context information. (1-1-3) The access information includes an ID, a target user that describes a user or group that permits or does not permit reading, and an access right that describes permission or non-permission of reading. (1-2) The above privacy policy information includes a node condition describing a target class or keyword, a target user describing a user or group that permits or does not permit reading, and permission to read. , Having one or more records including the access right describing the non-permission, and (1-3) the first context The management means collects the context and creates the context information and the node information, matches the created context information and node information with the privacy policy information, and node conditions of the privacy policy information, When the context information or the node information matches, after copying the target user and access of the privacy policy information to the access information, the context information, the node information and the access information are transmitted to the server, and the user And (2) the server has a second context management unit and a second context holding unit, and (2-1) the second context management unit is transmitted to the server. The context holding means is transmitted from the client. The above-mentioned context information, the node information, and the access information are stored for each user. (2-2) The second context management means transmits the transmitted context information, the node information, and the access The information is held in the second context holding unit, and when the user's context is requested from the user of an arbitrary client, the requesting user relates to the requested context of the requested user. When reading is permitted by the access right, the context information is returned.

  According to a third aspect of the present invention, there is provided one or more clients that acquire context information related to individual users and intermittently transmit the information to a server, and information on the context of an arbitrary user requested by an arbitrary client under predetermined conditions. In the above-mentioned client in the context sharing system having a server that sends back, 0 or 1 or more records including (1) (1-1) node ID, one or more sets of property and value information, and a processing flag (1-2) node information having zero or one or more records including ID, class and value corresponding to the node ID or property value of the context information; and (1-3) ID, target user that describes the user or group that permits or does not permit reading, permission to read Context holding means for holding access information having 0 or 1 or more records including an access right describing non-permission, and (2) a node condition describing a target class or keyword, and reading is permitted. Or a privacy policy management means for managing privacy policy information having one or more records including a target user describing a user or a group that is not permitted, an access right describing permission of reading and permission, and (3) The context is collected to create the context information and the node information, and the created context information and node information are matched with the privacy policy information, and the node condition of the privacy policy information and the context information or node Information is In this case, after copying the target user and access of the privacy policy information to the access information, the context information, the node information and the access information are transmitted to the server, and in response to a user input, Context management means for transmitting a context request to the server.

  According to a fourth aspect of the present invention, there is provided one or more clients that acquire context information related to individual users and intermittently transmit the information to a server, and information on the context of an arbitrary user requested by an arbitrary client under a predetermined condition. In the above server in the context sharing system having a server to send back, (1) includes (1-1) node ID, one or more property / value pair information, and a processing flag transmitted from the client. , Context information having 0 or 1 or more records, and (1-2) a node having 0 or 1 or more records including ID, class and value corresponding to the node ID or property value of the context information Information, (1-3) ID, and the user or group that permits or does not permit reading A context holding means for holding access information having zero or one or more records including an access right describing the target user, read permission, and non-permission, and (2) the transmitted context information, The node holding information and the access information are held in the context holding means, and when the user's context is requested by the user of an arbitrary client, the request source user requests the requested user's context. And a context management means for returning the context information when reading is permitted with the access right.

  According to the present invention, it is possible to provide a context sharing system, a context sharing method, a client, and a server that are easy to set and control a privacy policy when sharing a context and in which a description of the privacy policy is clear. Can do.

(A) First Embodiment Hereinafter, a first embodiment of a context sharing system, a context sharing method, a client, and a server according to the present invention will be described with reference to the drawings.

(A-1) Configuration of First Embodiment FIG. 1 is a block diagram showing a configuration of a context sharing system according to the first embodiment. In FIG. 1, a plurality of clients 1 (1A, 1B, 1C,...) And a server 2 are connected via a network.

  Each client 1 transmits and manages a context to the server 2 and requests and retrieves a desired context from the server 2. Each client 1 corresponds to, for example, an information processing apparatus such as a personal computer or a mobile phone having a communication function in which a program for a context sharing system is installed. It has a detailed configuration as shown in the client 1A in FIG.

  Each client 1 holds a user ID 10, a context management unit 11 that manages context transmission, a context holding unit 12 that holds context information, a privacy policy management unit 13 that manages a privacy policy, and a security policy. Security policy management means 14 to manage, context estimation means 15 to estimate context from sensor and application information, 0 or one or more sensors 16 (16A, 16B, 16C,...), And zero or one or more computer applications 17 (17A, 17B, 17C,...).

  Connected to the context management unit 11 are a context holding unit 12, a privacy policy management unit 13, a security policy management unit 14, a context estimation unit 15, sensors 16, and computer applications 17.

  In FIG. 1, the configuration for extracting the context from the server 2 side in each client 1 is omitted.

  The context management unit 11 holds class definition information 111 as shown in FIG. 2, and class definition is performed for data and information input from the sensor 16 and data and information given from the activated application 17. The information 111 is applied, and the context held by the context holding unit 12 is formed and managed.

  The class definition information 111 is composed of one or more class definitions 111A, 111B, 111C,. Each class definition includes a class name 1111, 0 or 1 or more property definitions 1112, 0 or 1 or more method definitions 1113. Each class definition consisting of a set of class name 1111, property definition 1112, and method definition 1113 is defined such that context information and node information described later can form a triple of RDF (Resource Description Framework), and a plurality of It is defined so that an RDF graph can be easily constructed from RDF triples.

  In the case of this embodiment, the description of “RDF triple” is not limited to the description according to the description method according to the W3C recommendation, and may be a description method capable of realizing the same notation as the RDF triple or RDF graph. (For example, it may not be expressed in the URL reference method).

  For example, as a class definition whose class name 1111 is “location information context”, the property definition 1112 is defined by the property “date / time” which is a date / time class. As another class definition whose class name 1111 is “location information context”, the property definition 1112 is defined by the property “position” which is an address class. As yet another class definition whose class name 1111 is “location information context”, a property “movement situation” which is a movement situation class is defined as its property definition 1112.

  Further, for example, a method “date ()” for obtaining a date class is defined as a method definition 1113 as one class definition whose class name 1111 is “date and time”. In addition, as another class definition whose class name 1111 is “date and time”, a method “time ()” for acquiring a time class is defined as a method definition 1113. Further, as another class definition whose class name 1111 is “date and time”, a method “day of week ()” for obtaining the day class is defined as the method definition 1113.

  Next, a general configuration of information held by the context holding unit 12 will be described with reference to FIGS. A specific configuration example will be referred to in the operation section described later.

  As shown in FIG. 3, the context holding unit 12 mainly holds context information 121, node information 122, and access information 123.

  The context information 121 is composed of zero or one or more records 121A, 121B, 121C,... And describes the context. As shown in FIG. 4, each record includes a node ID 1211, a set of one or more properties 1212 and a value 1213, and a processing flag 1214. The node ID 1211, the property 1212, and the value 1213 are RDF triples, and the RDF triples having the same node ID 1211 are described in the same record. As the value 1213 of the property 1212, another node ID may be described. The processing flag 1214 is described in units of RDF triples, and describes whether or not information related to the RDF triples has been updated from the client 1 to the server 2.

  The node information 122 is composed of zero or one or more records 122A, 122B, 122C,... And describes node information that defines the RDF triple of the context information 121. As shown in FIG. 5, each record includes an ID 1221 corresponding to the node ID 1211 or the value 1213 of the context information 121, a class 1222, and a value 1223. Again, the ID 1221, class 1222, and value 1223 are RDF triples.

  The access information 123 is composed of zero or one or more records 123A, 123B, 123C,..., And describes conditions for refusing or permitting access rights, presence / absence of disclosure (publication) of context information, etc. It is. As shown in FIG. 6, each record includes an ID 1231, a target user 1232, an access right 1233, and an anonymous flag 1234. Node ID is described in ID1231. The target user 1232 is a field for specifying target users such as “all”, “external”, and “internal”. The access right 1233 describes rejection or permission. Anonymity flag 1234 describes the presence / absence or the like of the user ID.

  The privacy policy management means 13 holds the privacy policy information 131 that forms the basis of the access information 123, and the security policy management means 14 holds the security policy information 141 that forms the basis of the access information 123. Is. The privacy policy information 131 is policy information for protecting personal information or small group information about users, and the security policy information 141 is for all users or a large number of users (for example, company-wide or departmental). This is policy information for information protection that is commonly applied.

  The privacy policy information 131 is composed of one or more records 131A, 131B, 131C,. As shown in FIG. 7, each record has a node condition 1311 describing a target class or keyword, a target user 1312 describing a user or group that permits reading, or permits reading, or It consists of an access right 1313 that describes “not permitted” and an anonymous flag 1314 that describes whether to publish anonymously or with a registered name.

  The security policy information 141 is also composed of one or more records 141A, 141B, 141C,. As shown in FIG. 8, each record has a node condition 1411 describing a target class or keyword, a target user 1412 describing a user or a group that permits reading, or permits reading, or It consists of an access right 1413 describing not permitted, and an anonymous flag 1414 describing whether to publish anonymously or with a registered name.

  The sensor 16 is a position sensor, a temperature sensor, an acoustic sensor, a light intensity sensor, an acceleration sensor, or the like using a GPS sensor or wireless communication. If the sensor 16 is related to the acquisition of the context about a user, the kind of the sensor will not be ask | required. When the client 1 is configured by an information processing apparatus such as a personal computer, the sensor 16 may be built in the information processing apparatus, and can be connected to a USB terminal or the like included in the information processing apparatus. Also good.

  The application 17 is an application program such as mail software, telephone software, chat software, document creation software, an Internet browser, and a scheduler. The type of application 17 is not limited as long as it is related to the acquisition of context about the user.

  The context estimation unit 15 estimates a context from information from the sensor 16 and the application 17. The estimation method will be clarified in the operation section.

  The server 2 receives the context and transmits the context in response to the request. The server 2 corresponds to, for example, an information processing apparatus such as a personal computer having a communication function in which a program for a context sharing system is installed. From the viewpoint of the context sharing function, the server 2 is functionally shown in FIG. It has such a detailed configuration.

  That is, the server 2 includes a context management unit 21, a context holding unit 22 that holds context information of each user, and a security policy management unit 24 that manages a security policy. A context holding unit 22 and a security policy management unit 24 are connected to the context management unit 21.

  The context holding unit 22 shares and holds information related to the context for each user. As shown in FIG. 9, the context holding unit 22 holds one or more user IDs 220 and context information 221, node information 222, and access information 223 corresponding to each user ID 220.

  Although not shown in the figure, the context information 221 has the same configuration as the context information 121 in the client 1 and is composed of zero or one or more records 221A, 221B, 221C,..., And each record has a node ID 2211 and It consists of a set of one or more properties 2212 and values 2213 (see FIG. 4).

  Although not shown, the node information 222 has the same configuration as that of the node information 122 in the client 1 and includes 0 or 1 or more records 222A, 222B, 222C,..., And each record has an ID 2221, a class 2222, It consists of a value 2223 (see FIG. 5).

Although the access information 223 is not illustrated, the access information 12 in the client 1 is not shown.
, And is composed of zero or one or more records 223A, 223B, 223C,..., And each record is composed of an ID 2231, a target user 2232, an access right 2233, and an anonymous flag 2234 (see FIG. 6). ).

  The security policy management means 24 holds security policy information 241. Although not shown, the security policy information 241 has the same configuration as the security policy information 141 in the client 1 and is composed of one or more records 241A, 241B, 241C,..., Each record having a node condition 2411, a target It consists of a user 2412, an access right 2413, and an anonymous flag 2414 (see FIG. 8).

(A-2) Operation of First Embodiment Next, the operation of the context sharing system according to the first embodiment will be described with reference to the drawings.

  First, the security policy unification operation between the server 2 and the client 1 will be described.

  As shown in the sequence diagram of FIG. 10, the client 1 accesses the server 2 to fetch the security policy information 241 (S 1, S 2), and the client 1 security policy information 141 of the client 1 and the imported server 2 It is determined whether or not the security policy information 241 is the same (S3). If the security policy information 241 is not the same, the security policy information 241 of the server 2 is copied and used as the security policy information 141 of the client 1 (S4). Note that downloading of security policy information 241 for loading and comparison processing of both security policy information 141 and 241 may be executed in parallel.

  Although the server 2 does not show the sequence diagram or the flowchart, when the security policy information 241 of the server 2 is changed, the server 2 accesses each client 1A, 1B, 1C,. The information 241 is copied and given to be stored as security policy information 141 in each client 1A, 1B, 1C,.

  Through the operation as described above, it is possible to unify security policies in the context sharing system.

  Next, a method for generating the context information 121 in the client 1 will be described with reference to FIG. In the following description, it is assumed that the sensor 16 is a GPS sensor or a position information sensor using wireless communication, and the application 17 is mail software.

  The sensor 16 acquires the current position data periodically (for example, every one minute) and sends it to the context management unit 11 together with the acquired date / time data (S51). For example, “data = position information; date = 2006/10/22 12:34; latitude = 35/38 / 26.185; longitude = 139/44 / 49.47” is sent.

  The context management unit 11 sends the data sent from the sensor 16 to the context estimation unit 15 (S52). The context estimation unit 15 estimates the context according to the sent data (S53), and context management is performed as context information. Return to means 11 (S54).

  On the other hand, the application 17 sends the type of the transmitted mail or the read mail and the entire mail to the context management unit 11 at the time of mail transmission and mail browsing or periodically (for example, every 15 minutes) (S61). .

  Also at this time, the context management unit 11 sends the data sent from the application 17 to the context estimation unit 15 (S52), and the context estimation unit 15 estimates the context according to the sent data (S53), It returns to the context management means 11 as context information (S54).

  Hereinafter, an estimation method of the context estimation unit 15 will be described.

  When estimating the position information context from the received position information data, the context estimation means 15 first, if the current position data is coordinate data, using the received position information data as a key, an address or landmark Information (eg, Tokyo Station, Keio University Hiyoshi Campus, Tomei Expressway Ebina SA, etc.) is searched for, and an address and, if present, a landmark are obtained. In contrast, when the sent position information data is address or landmark information, the context estimation means 15 obtains coordinate data using them as keys.

  The context estimation unit 15 may perform the above-described data conversion at the time of estimation based on internal information prepared in advance or may use a general-purpose Web service. As such a general-purpose Web service, for example, Geocoding (http://www.geocoding.jp/) of a personal site, Geocoding API (http://api.knecht.jp/geocoding/) of kneecht, etc. are applied. it can.

  In the case of the above-described position data “data = position information; date = 2006/10/22 12:34; latitude = 35/38 / 26.185; longitude = 139/44 / 49.47”, the address is “Tokyo The landmark is “Oki Electric Industry Co., Ltd.”.

  Next, the context estimation means 15 determines the movement status (stopped, walking, getting on a train or a car) from the time series of the current position data (the past current position data is buffered for a predetermined time, for example). Is determined). As such a determination method, for example, a method described in Japanese Patent Laid-Open No. 2003-0776818 can be applied. For example, if the current position data and the previous position data are the same, the speed is 0 and the movement status is “still”.

  Thereafter, the context estimation unit 15 returns the date / time data, the address and landmark of the position at the date / time, and the movement status to the context management unit 11 as the position information context. In the above example, the data includes “context: location information; date: 2006/10/22 12:34; address = 4-10-16 Shibaura, Minato-ku, Tokyo; landmark = Oki Electric Industry Co., Ltd .; = "Still".

  On the other hand, the context estimation means 15 performs the following estimation operation when the mail type and the entire mail are sent. In the following, for example, “data = email; type = reception; transmission date = 2006/10/22 12:34; sender = Ichiro Tanaka; transmission address=ichiro.tanaka@abc.com; recipient = Taro Yamada; Address=taro.yamada@xyz.com; title = 11/1 meeting; mail body = Tanaka @ ABC 杜. Sato also participated in the 11/1 meeting, and CTI ... ”was sent.

  The context estimation means 15 estimates the mail history context from the mail data sent via the context management means 11 (S53).

  First, attributes such as inside and outside are obtained from the email address. In the above example, it is estimated that the attribute of the sender is “external” from [abc] below the transmission address @. Next, from the email body, using the unique expression extraction technique and keyword extraction technique, the unique expression as a keyword (person name, organization name, place name, product name, event name, address, telephone number, email address, etc.), and Extract other keywords. Also, the unique expression as a keyword and other keywords are extracted from the mail title.

  For example, a method described in Japanese Patent Application Laid-Open No. 2004-046775 can be applied as a method for extracting a proper expression. In the above example, “11/1” as the date, “Tanaka” and “Sato” as the personal names, and “ABC 杜” as the organization name are extracted from the mail text, and “11/1” as the date and time is extracted from the mail title. Extracted. As a keyword extraction method, for example, a method described in JP-A-200235061 can be applied. In the above example, “meeting” and “CTI” are extracted from the mail text, and “meeting” is extracted from the mail title.

  Thereafter, the context estimation means 15 includes, as a mail history context, data such as mail type, transmission date / time, reception date / time, mail sender, recipient, title, keyword (specific expression and other keywords), etc. “context = mail history” Date = 2006/10/22 12:34; Sender = Ichiro Tanaka; Sender attribute = External: Recipient = Taro Yamada; Title = 11/1 meeting; Keyword = 11/1 (date) , Tanaka (person name), Sato (person name), ABC company (organization name), meeting (not set), and CTI (not set) ”are returned to the context management means 11.

  The context management unit 11 converts the returned context into a format corresponding to the context information 121 and stores it in the context holding unit 12 (S54). Hereinafter, the operation of step S54 will be described in detail.

  The context management unit 11 first acquires the context information 121 and the node information 122 by the context holding unit 12.

  In the example of the position information context described above, the following five records 122A to 122E are created in the node information 122 as shown in FIG. The record 122A has a unique ID (eg, “CP0001”) that does not overlap the ID 1221, the class 1222 is “location information context”, and the value 1223 is “none (blank)”. The record 122B is a unique ID (eg, “D0001”) in which the ID 1221 does not overlap, the class 1222 is “date / time”, and the date / time information (“2006/10/22 12:34”) at which the value 1223 is transmitted. The record 122C is a unique ID (for example, “P0001”) in which the ID 1221 does not overlap, the class 1222 is “address”, and the address to which the value 1223 is transmitted (“4-10-16 Shibaura, Minato-ku, Tokyo”). The record 122D is a unique ID (for example, “P0002”) in which the ID 1221 is not duplicated, the class 1222 is “landmark”, and the landmark (“Oki Electric Industry Co., Ltd.”) in which the value 1223 is transmitted. The record 122E is a unique ID (for example, “C0001”) in which the ID 1221 does not overlap, the class 1222 is “movement status”, and the movement status (for example, “still”) in which the value 1223 is transmitted.

  However, when creating a record of node information 122, if a record with the same class 1222 and the same value 1223 exists, a new record is not created.

  Further, a record 121A of the context information 121 as shown in FIG. As the node ID 1211 in the new record 121A of the context information 121, the same value as the ID 1221 of the first record 122A in the node information 122, that is, “CP0001” is inserted. Further, as a set of the property 1212 and the value 1213 of the context information 121, “date and time” and “D0001” corresponding to the second record 122B in the node information 122 and “position” corresponding to the third record 122C in the node information 122 And “P0001”, “position” and “P0002” corresponding to the fourth record 122D in the node information 122, “movement status” and “CO001” corresponding to the fifth record 122E in the node information 122, and processing flag Reference numeral 1214 denotes “unprocessed”.

  In the above, setting the properties of the class “address” and “landmark” to “position” is defined by the class definition information shown in FIG.

  When the configuration of the context information 121 and the node information 122 shown in FIG. 12 is drawn with a graph using nodes and arcs, it is as shown in FIG. That is, the graph is an RDF graph in which “node, arc, node” is an RDF triple.

  FIG. 14 shows the acquired context information 121 and node information 122 for the mail history context example described above. Note that FIG. 14 shows some records for both the context information 121 and the node information 122.

  In the example of the mail history context, the node information 122 includes the ID 1221 of “CM0001”, the class 1222 of “mail history context”, the record 122F having no value 1223, the ID 1221 of “MT0001”, and the class 1222 of “mail transmission / reception”. A record 122G having a “type” and a value 1223 of “reception” is created, and the node information 122 has an ID 1221 of “D0002”, a class 1222 of “date and time”, and a value 1223 of “10/22 12:34”. A record 122H is created. Similarly, a record of node information 122 is created for the sender, transmission address, recipient, reception address, title, and keyword. At that time, for each keyword, the node information 122 includes a record 122I whose ID 1221 is “K0001”, class 1222 is “person”, value 1223 is “Tanaka”, ID 1221 is “K0002”, and class 1222 is Records 122J,... With “person” and value 1223 “Sato” are created. Thereafter, in the context information 121, the node ID 1211 is “CM0001”, and the property 1212 and value 1213 pairs are “type”, “MT0001”, “transmission / reception date / time”, “D0002”, “keyword”, “K0001”, “keyword”. , “K0002” (hereinafter abbreviated), a record 121B having a process flag 1214 of “unprocessed” is created.

  When the process of creating (acquiring) the context information 121 and the node information 122 is completed, the context holding unit 12 stores the context information 121 and the node information 122 as a file, for example, in a database.

  Next, prior to description of context transmission processing in consideration of privacy and security, security policy and privacy policy setting will be described in order.

  Security policies include: “Personal information is not distributed at all even within the company”, “Employee location information is shared internally, and only the location information of the company (including head office and branch offices) is disclosed outside the company”, “ The case of setting the mail history as “transmission / reception distinction, reception / transmission date / time, title, and keyword only” will be described with reference to a setting example shown in FIG. The security policy information 241 in the server 2 is described by a user (administrator) according to a predetermined description method.

  As described above, the security policy information 241 in the server 2 has the same configuration as the security policy information 141 in the client 1 shown in FIG. 8, and each record includes a node condition 2411, a target user 2412, an access right 2413, It consists of an anonymous flag 2414.

  First, the user (administrator) sets the node policy 2411 to “X; X. class == person”, the target user 2412 is “all”, the access right 2413 is “deny”, and anonymous in the security policy information 241 of the server 2. A record 241A in which the flag 2414 is “not set” is created. This means "Personal information is never distributed even within the company".

  Next, a record 241B in which the node condition 2411 is “X; X. class == location information context”, the target user 2412 is “in-house”, the access right 2413 is “permitted”, and the anonymous flag 2414 is “name”, Condition 2411 is “X; X. class == location information context AND (X, location, @internal landmark)”, target user 2412 is “external”, access right 2413 is “permitted”, and anonymous flag 2414 is “name” Record 241C, node condition 2411 is “X; X. class == location information context”, target user 2412 is “external”, access right 2413 is “denied”, and anonymous flag 2414 is “unset” 241D is created. These records 241 </ b> B to 241 </ b> D mean “employee location information is shared within the company, and only location information within the company (including the head office and branch office) is disclosed outside the company”.

  Further, the node condition 2411 is “X, Y, Z, U, V; X. class == mail history context AND (X, type, Y) AND (X, date, Z) AND (X, title, U) AND (X, keyword, V) ”, the target user 2412 is“ in-house ”, the access right 2413 is“ permitted ”, the anonymous flag 2414 is“ record name ”, and the node condition 2411 is“ Y; X. class == A mail history context AND (X, *, Y) ", a record 241F in which the target user 2412 is" in-house ", the access right 2413 is" denied ", and the anonymous flag 2414 is" not set "is created. These records 241 </ b> E and 241 </ b> F mean “the mail history is made publicly disclosed only in transmission / reception, reception or transmission date / time, title, and keywords, and other than that”.

  In the notation of the security policy information 241, “X;” means that X is a node for which a policy is set, and “X. class ==” is equal to a value in which a class of a certain node is described. Although it does not exist in FIG. 15, “X ==” means that the value of a certain node is equal to the described value. Also, for example, “(X, position, value)” means that the node connected by the property whose class is “position” from the target node is a certain value, and “*” matches everything. It means to do. “@ In-house landmark” is a set of in-house landmarks, that is, “self-seat”, “first meeting room”, “second meeting room”, “branch 1”, “branch 2”, and “in-house”. . Each record has a higher priority on the top side.

  The security policy information 141 of each client 1 has the same content as the security policy 241 of the server 2 by the above-described security policy unification method.

  Next, the setting shown in FIG. 16 is set when the client 1 is set as a privacy policy “location information is disclosed to the company from 9:30 to 17:30 on weekdays but is not disclosed to others including the company”. This will be described with reference to an example. The privacy policy of the client 1 is described by the user of the client 1 according to a predetermined description method.

  As described above, each record of the privacy policy information 131 in the client 1 includes the node condition 1311, the target user 1312, the access right 1313, and the anonymous flag 1314.

  The user of the client 1 sets the node condition 1311 in the privacy policy information 131 of the client 1 as “X, Y, Z, U; X. class == location information context AND (X, date and time, Y) AND (Y, day of week, Z) AND Z == @ Weekday AND (Y, Time, U) AND U> = Time (9:00) AND U <= Time (17:30) ”, the target user 1312 is“ internal ”, and the access right 1313 is “Allow”, record 131A with anonymous flag 1314 “name”, node condition 1311 “X; X. class == location information context”, target user 1312 “all”, and access right 1313 “rejected” ", The record 131B in which the anonymous flag 1314 is" not set "is created.

  Here, “(Y, day of the week, Z)” means a node (Z) having a value connected with the “day of the week” property from the node (Y), but the “date and time” class is a method for returning the “day of the week” class. If “day of week ()” is defined in the class definition information 111 and the “day of week” property does not exist, a node of the “tomorrow” class is acquired by this method. “Time (9 o'clock)” is a constructor that returns a node (time class) corresponding to 9 o'clock and is defined in the class definition information 111.

  The context transmission process is executed in consideration of (reference to) the security policy and privacy policy set as described above.

  Next, the context disclosure process will be described with reference to FIG. In the context disclosure process, the context management unit 11 periodically (for example, every 15 minutes) matches the context information 121 in the order of the security policy and the privacy policy to determine the disclosure range of the context information. .

  The context management unit 11 first refers to the records of the context information 121 in order. When the processing flag 1214 of the record is “unprocessed”, the context management unit 11 refers to the information of the node information 122 and sets the processing flag 1214 to “processed”. (S101). In the example of FIG. 12A, the record 121A is referred to, and the processing flag 1214 is “unprocessed”. Therefore, the node ID 1211, that is, “CP0001”, and the set of the property 1212 and the value 1213, that is, “date and time”. And “D0001”, “Address” and “P0001”, “Landmark” and “P0002”, “Movement Status”, and “C0001”, the corresponding ID1221 record of the node information 122, that is, the ID1221 is “CP0001”. Record 122A of node information 122 whose class 1222 is "location information context" and value is "CP0001", record 122B of node information 122 whose ID 1221 is "D0001", class 122 is "date and time", and value is date information. Refer to

  Next, the records of the context information 121 and the node information 122 that have been referred to are matched with the node conditions 1411 of the records 141A, 141B, 141C,... Of the security policy information 141 in order. A record is created in 123 (S102).

  FIG. 18A is an explanatory diagram illustrating an example of a record created in the access information 123 in the case of the above example.

  In the above example, since the record 121A of the context information 121 matches the node condition 1411 “X.class == location information context” of the record 141B of the security policy information 141, the ID 1221 of the record of the context information 121 that matches the ID 1231. “CP0001” is set, and the target user 1412 “internal”, access right 1413 “permitted”, and anonymous flag 1414 “name” of the record of the security policy information 141 in which the target user 1232, the access right 1233, and the anonymous flag 1234 are matched. The set record 123A is created in the access information 123. However, when creating a record in the access information 123, if a record having the same ID 1231 and target user 1232 already exists, a new record is not created.

  Similarly, since the record 122A of the node information 122 matches the node condition 1411 “X. class == location information context” of the record 141D of the security policy information 141, the ID 1231 is “CP0001” and the target user 1232 is “external”. The record 123B in which the access right 1233 is “denied” and the anonymous flag 1234 is “not set” is created in the access information 123.

  Similarly, matching with the privacy policy is performed (S103). In the above example, when October 22nd is Sunday, the record 122A of the node information 122 matches the node condition 1311 “X. class == location information context” of the record 131B of the privacy policy information 131. A record 123C in which “CP0001”, the target user 1232 is “all”, the access right 1233 is “denied”, and the anonymous flag 1234 is “not set” is created in the access information 123.

  Next, for each record of the access information 123, the context management unit 11 obtains a logical product of the records having the same ID 1231 and replaces the original record with the obtained one (S104). FIG. 18B is a replacement of FIG.

  Therefore, first, when the access right 1233 of each record is “denied”, the target user 1232 is changed to a complementary set, and the access right 1233 is set to “permitted”. In the above example, the target user 1232 of the record 123B is changed to “NOT outside”, the access right 1233 is changed to “permitted”, the target user 1232 of the record 123C is changed to “none”, and the access right 1233 is changed to “permitted”. When the logical product is obtained, the target user 1232 becomes “internal AND NOT external AND no”, that is, “none”, and the record of the access information 123 indicates that the ID 1231 is “CP0001”, the target user 1232 is “none”, and the access right 1233 is “Permitted” and the anonymous flag 1234 are “not set”. This is the location information context that indicates that you were in Shibaura 4-10-16, Minato-ku, Tokyo, at 12:34 on October 22nd, because that day is Sunday, so it will not be disclosed by the privacy policy. It means that.

  As for the example of the mail history context, when matching with the security policy, the node corresponding to the transmission / reception distinction, transmission or reception date / time, title, keyword is recorded in the records 141E, 141F, 141G, 141H of the security policy information 141, respectively. For a match, the sender, the transmission address, the receiver, the reception address, and the corresponding node match the record 141I of the security policy information 141.

  After the logical product processing, the access information 123 includes a record in which the ID 1231 is “CM0001”, the target user 1232 is “internal”, the access right 1233 is “permitted”, and the anonymous flag 1234 is “signed”, and the ID 1231 is transmitted / received. ID of node corresponding to distinction, transmission or reception date / time, title, keyword, target user 1232 is “internal”, access right 1233 is “permitted”, anonymous flag 1234 is “record name”, and ID 1231 is transmitted , Sender address, recipient address, received address, ID corresponding to body node, target user 1232 is “none”, access right 1233 is “permitted”, and anonymous flag 1234 is “not set”. Records are created.

  Thereafter, the context management unit 11 sends context information other than private information to the server 2 together with the user ID 10 based on the access information 123 (S105).

  In the above example, the record whose ID 1231 of the access information 123 is “CP0001” does not send context information because the target user 1232 is “None”, and the record whose ID 1231 is “CM0001” is “ Because “in-house”, the record, the record 121B related to the context information 121, and the record related to the node information 122 are sent to the server 2 together with the user ID 10 (for example, “U0001”). However, when the IDs 1221 and 1231 of the records related to the node information 122 and the access information 123 are records of the access information 123 and the target user 1232 is “none”, the records of the node information 122 and the access information 123 are recorded. Will not send.

Next, the processing in the server 2 will be described using the flowchart of FIG.
When the server 2 receives the user ID, the record of context information, the record of node information, and the record of access information from the client 1, the context management unit 21 corresponds to the received user ID by the context holding unit 22. The information is stored in the context information 221, the node information 222, and the access information 223, respectively (S201).

  In the above example, as context information 221 corresponding to the user ID “U0001”, the node ID 2211 is “CM0001”, and the set of the property 2212 and the value 2213 is “transmission / reception date”, “D0002”, “keyword”, and “K0001”. , “Keyword” and “K0002”, “keyword” and “K0003”, “keyword” and “K0004”, and related node information 222 and access information 223, ID 2231 is “CM0001”, user A record in which the information 2232 is “in-house”, the access right 2233 is “permitted”, and the anonymous flag 2234 is “name” is stored.

  When the user accesses the server 2 and requests a context, the context is returned according to the access right (S202).

  When a user in the company (ID = U0002) requests the current location information context of the user ID “U0001” (October 22, 12:45) from the client 1B, the client 1B receives “requested user ID: U0002; Request context = location information context; user ID = U0001 ”is sent to the server 2.

  Since the server 2 cannot find context information within a certain time (for example, 15 minutes, which is the time interval for the client 1 to send context information to the server 2) for “U0001”, “user ID = U0001; class = position Information context; result = none is returned.

  In addition, when an in-house user (ID = U0002) requests a mail history context with a user ID “U0001”, there is a record whose ID 2231 is “CM0001” in the context information 221 as a context within a certain period of time. Since the target user 2232 of the access information 223 to be “in-house”, the access right 2233 is “permitted”, and the anonymous flag 2234 is “named”, the context information is “user ID = U0001; class = mail history context; type = received” ; Date: 2006/10/22 12:34; Title = 11/1 meeting; Keyword = 11/1 (date), ABC company (organization name), meeting (not set), CTI (not set) .

  As another example, the location information context where the date and time is Monday, October 23 and the other contents are the same (I was in 4-10-16 Shibaura, Minato-ku, Tokyo at 12:34 on October 23 In the case of matching the context information with the privacy policy in the client 1, the record of the node information 122 is the node condition 1311 “X” of the record 131A of the security policy information 131. Class == Location information context AND (X, Date, Y) AND (Y, Day of the week, Z) AND Z == @ Weekday AND (Y, Time, U) AND U> = Time (9 o'clock) AND U <= Time (17:30) ”, the access information 123 has the ID 1231“ CP0002 ”, the target user 12 2 is “internal”, the access right 1233 is “permitted”, and the anonymous flag 1234 is “record name”, and a record 123D is created, and the logical product is that the target user 1232 is “internal AND NOT external external AND internal”, that is, “internal”. In the record of the access information 123, the ID 1231 is “CP0002”, the target user 1232 is “in-house”, the access right 1233 is “permitted”, and the anonymous flag 1234 is “signed”.

  When this context is stored in the server 2, when a user in the company (ID = U0002) requests the current location information context (October 23, 12:45) of the user ID “U0001” from the client 1B. As a context within a certain time, there is a record whose ID 2211 is “CP0002” in the context information 221, the target user 2232 of the corresponding access information 223 is “internal”, the access right 2233 is “permitted”, and the anonymous flag 2234 is “ “User ID = U0001; Class = Location information context; Date = 2006/10/23 12:34; Address = 4-10-16 Shibaura, Minato-ku, Tokyo; Landmark = Oki Electric “Industry Co., Ltd .; Movement status = stationary” is returned.

(A-3) Effects of the First Embodiment As described above, according to the first embodiment, the context information 121 is composed of the node ID 1211, the property 1212, and the value 1213, and the node information 122 is changed to the ID 1221, The security policy information 141 is composed of a node condition 1411, a target user 1412, and an access right 1413, and the privacy policy information 131 is also composed of a node condition 1311, a target user 1312, and an access right 1313. Since it did in this way, the effect that it becomes possible to specify the class of the context to be disclosed or not, to specify the value, and to specify the condition of a certain node can be obtained.

  For example, “Personal information will not be distributed at all even within the company”, “Employee location information will be shared internally, and only the location information of the company (including head office and branch offices) will be disclosed outside the company”, “The mail history is only disclosed within the company for the distinction between transmission and reception, the date and time of reception or transmission, the title, and the keyword.” “The location information is disclosed internally from 9 am to 17:30 on weekdays. It is possible to set a complicated security policy such as “private” and a privacy policy.

  In addition, by preparing “person”, “name”, “address”, “phone number”, etc. as classes, and using the specific expression extraction technology for extracting keywords corresponding to the classes in the context estimation means 15, When the keyword extracted from the text part is used as the context, when publishing the context, even if the general keyword is made public, it is possible to keep only personal information private, maximizing while protecting security and privacy. The effect that the context of can be disclosed is obtained.

(B) Second Embodiment Next, a second embodiment of the context sharing system, the context sharing method, the client, and the server according to the present invention will be described with reference to the drawings.

  The configuration of the context sharing system according to the second embodiment is almost the same as that of the first embodiment. However, as shown in FIG. The point which can call the estimation means 15 differs from 1st Embodiment. In the second embodiment, it is assumed that the application 17 can request and receive a context from the server 2 via the context management unit 11.

  In the case of the second embodiment, it is assumed that the server 2 has a function of accepting a request for a context to which a condition is added and returning a context that matches the condition.

  An example of characteristic operations will be described assuming that the application 17 of the client 1 is mail software, as in the case of the first embodiment. FIG. 21 is a sequence diagram showing an example of this characteristic operation.

  When mail is transmitted, the mail software (application 17) first calls the context estimation means 15 to obtain the mail context (mail history context).

  For example, if the destination is the section manager (Jiro Suzuki), the title is “ABC company meeting”, and the text is “ABC company has contacted ...”, the application 17 sends the mail data to the context estimation unit 15 and the context estimation unit 15 Extracts “ABC company” as the organization name, and “context = mail history; type = unsent; date = 2006/10/22 15:00; sender = Taro Yamada; recipient = Jiro Suzuki”; “Receiver attribute = internal; title = ABC company meeting; keyword = ABC company (organization);...” Is returned to the application 17 (S301).

  Thereafter, the application 17 calls the server 2 and obtains a context corresponding to the mail history context of the mail to be transmitted (S302).

  First, the application 17 adds, for example, the keyword “ABC 杜” of the mail history context of the mail to be transmitted and the recipient attribute “in-house” as a context disclosure range to the request data, “request user ID = U0001; request context = *; User ID = U0001; keyword = ABC company; disclosure range = in-house ”is sent to the server 2. Here, both the requesting user ID and the user ID being “U0001” means requesting its own context, and “requesting context = *” is requesting all contexts. Means.

  The server 2 searches the context relating to the user “U0001” including the keyword “ABC company”, and “user ID = U0001; class = mail history context; type = reception; date = 2006/10/22 12:34. Title = 11/1 about meeting; keyword = 11/1 (date), ABC company (organization name), meeting (not set), CTI (not set),. If another keyword “XYZ company” is requested as a condition at this time, a context matching the keyword is not found, and the server 2 determines that “user ID = U0001; class = mail history context; result = Returns “none”.

  The application 17 adds the obtained context information to the mail to be sent and sends it (S303).

  According to the second embodiment, since the server 2 has a function of accepting a request for a context to which a condition is added and returning a context that matches the condition, a service using the function can be provided.

  For example, by obtaining the user's own context regarding the mail to be sent and assigning it to the outgoing mail, the user's context can be communicated to the other party, and communication can be made more efficient.

(C) Third Embodiment Next, a third embodiment of the context sharing system, context sharing method, client, and server according to the present invention will be described with reference to the drawings.

  The configuration of the context sharing system according to the third embodiment is almost the same as that of the second embodiment. In the case of the third embodiment, the server 2 is different from the above-described embodiment in that it has a function of receiving a request for a context not specifying a user ID and returning contexts of a plurality of users that match the conditions.

  A characteristic operation example in the third embodiment will be described assuming that the application 17 of the client 1 is presentation software. FIG. 22 is a sequence diagram showing an example of this characteristic operation.

  When the user creates the material, the presentation software (application 17) calls the server 2 and obtains an anonymous context related to the keyword specified by the user.

  First, the application 17 adds, for example, the keyword “CTI” as a condition to the request data, “request user ID = U0001; request context = *; anonymous flag = ON; (X; X. class = person AND X, schedule, Y) AND (Y, date and time, 10/25 10:00) AND (Y, title, ABC company preliminary meeting)); keyword = CTI "is sent to the server 2 (S401). Here, “anonymous flag = ON” means an anonymous context, and “(X; X.class = person ...)” means that a context of a prospective participant is requested.

  The server 2 searches the context including the keyword “CTI”, and if there is a matching context, assigns a temporary user ID, for example, “class = mail history context; anonymous user ID = UTMP0001; type = received” Date and time = 2006/10/22 12:34; Title; 11/1 meeting; Keyword = 11/1 (date), ABC (organization name), meeting (not set), CTI (not set), ...; Class = file history context; anonymous user ID = UTMP0002; type = document;...; Keyword = CTI (unset) ”is returned (S402).

  The application 17 also obtains an anonymous context for “VoIP” (S403, S404).

  The application 17 displays that there are two contexts for “CTI” and one context for “VoIP” by counting the contexts by anonymous user ID from the obtained context information (S405).

  According to the third embodiment, the server 2 has a function of accepting a request for a context not specifying a user ID and returning contexts of a plurality of users that match the conditions, so that a service using the function is possible. It becomes.

  For example, it is possible to estimate the degree of recognition of conference participants' keywords in advance by obtaining the context of the prospective participants during the creation of the materials, and counting by keyword, thereby improving communication efficiency. be able to.

(D) Other Embodiments In each of the above embodiments, the context management unit 11 of the client 1 transmits the context information to the server 2, but instead of or in addition to this, the context management unit 11 An operation mode may be provided in which a context that can be disclosed is returned in response to a request from the application 17 in the client 1 without transmitting the context information to the server 2. For example, when the client 1 is a mobile phone, since the user always carries the client 1, this function is effective.

  Further, in each of the above embodiments, what is obtained by automatically estimating the context to be transmitted to the server is shown, but in addition to this, context information may be obtained according to user input information.

  Further, in each of the above embodiments, the context information and the like are periodically acquired. However, the context information may be intermittently acquired as a matter of course.

  Furthermore, in the description of each of the above embodiments, information such as context information has been illustrated and described in a tabular format. However, the storage format is not limited to a tabular format, and is the same unit as described above. Other storage formats may be used as long as they can be distinguished. In the claims, the term “record” is used for convenience.

It is a block diagram which shows the structure of the context sharing system which concerns on 1st Embodiment. It is explanatory drawing which shows the structure of the class definition information which the context management means in the client of 1st Embodiment manages. It is explanatory drawing which shows schematic structure of the information which the context holding | maintenance means in the client of 1st Embodiment hold | maintains. It is explanatory drawing which shows the structure of the context information in the client of 1st Embodiment. It is explanatory drawing which shows the structure of the node information in the client of 1st Embodiment. It is explanatory drawing which shows the structure of the access information in the client of 1st Embodiment. It is explanatory drawing which shows the structure of the privacy policy information in the client of 1st Embodiment. It is explanatory drawing which shows the structure of the security policy information in the client of 1st Embodiment. It is explanatory drawing which shows schematic structure of the information which the context holding | maintenance means in the server of 1st Embodiment hold | maintains. It is a sequence diagram which shows the operation | movement which the client of 1st Embodiment confirms security policy information at the time of starting. It is a sequence diagram which shows the production | generation method of the context information in the client of 1st Embodiment. It is explanatory drawing which shows the production example (1) of the context information and node information in the client of 1st Embodiment. It is explanatory drawing which expressed the context information of FIG. 12 with the RDF graph. It is explanatory drawing which shows the production example (2) of the context information and node information in the client of 1st Embodiment. It is explanatory drawing which shows the example of a setting of the security policy in 1st Embodiment. It is explanatory drawing which shows the example of a setting of the privacy policy in 1st Embodiment. It is a flowchart which shows the disclosure process of the context in 1st Embodiment. It is explanatory drawing which shows the production | generation example and integration example (logical product example) of the access information in the client of 1st Embodiment. It is a flowchart which shows the process of the server which concerns on 1st Embodiment. It is a block diagram which shows the structure of the context sharing system which concerns on 2nd Embodiment. It is a sequence diagram which shows the characteristic operation | movement in the context sharing system of 2nd Embodiment. It is a sequence diagram which shows the characteristic operation | movement in the context sharing system of 3rd Embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 (1A, 1B, 1C, ...) ... Client, 10 ... User ID, 11 ... Context management means, 12 ... Context holding means, 13 ... Privacy policy management means, 14 ... Security policy management means, 15 ... Context estimation means, 16 ... sensor, 17 ... application,
2 ... server, 21 ... context management means, 22 ... context holding means, 24 ... security policy management means.

Claims (8)

A context having one or more clients that acquire context information related to individual users and intermittently transmit the information to a server, and a server that returns information on the context of an arbitrary user requested by an arbitrary client under a predetermined condition In a shared system,
Each client above
Context information having zero or one or more records including node ID, one or more property and value pair information, and a processing flag;
Node information having zero or one or more records including the ID, class and value corresponding to the node ID or property value of the context information;
Holding access information having zero or more records, including an ID, a target user that describes a user or group that permits or does not permit reading, and an access right that describes permission or denial of reading 1 context holding means;
It has one or more records including node conditions that describe the target class or keyword, target users that describe users or groups that allow or do not permit reading, and access rights that describe permission or disapproval of reading A privacy policy management means for managing privacy policy information;
The context is collected to create the context information and the node information, and the created context information and node information are matched with the privacy policy information, and the node condition of the privacy policy information and the context information or node If the information matches, after copying the target user and access of the privacy policy information to the access information, the context information, the node information, and the access information are transmitted to the server, and in response to the user input. And a first context management means for transmitting a request for the context to the server,
The server
Second context holding means for holding the context information, the node information, and the access information transmitted from the client for each user;
The context information, the node information, and the access information that have been transmitted are held in the second context holding unit, and when the context of an arbitrary user is requested from a user of an arbitrary client, the request source A context sharing system, comprising: a second context management unit that returns the context information when the user is permitted to read with the access right related to the requested context of the requested user.   When the context of a certain condition of an arbitrary user is requested from the user of an arbitrary client, the second context management means matches the condition specified for the context information of the requested user, and the request source 2. The context sharing system according to claim 1, wherein the requested context information is returned when an access right is granted to the user. Each record of the access information further includes an anonymous flag, and each record of the privacy policy information further includes an anonymous flag,
When the context of a certain condition is requested from a user of an arbitrary client without designating the user, the second context management means matches the designated condition with respect to all context information, 3. The context sharing system according to claim 1, wherein the context information is returned when the access right of the requesting user is permitted. A context having one or more clients that acquire context information related to individual users and intermittently transmit the information to a server, and a server that returns information on the context of an arbitrary user requested by an arbitrary client under a predetermined condition In the sharing method,
Each client has first context management means for managing context transmission, first context holding means for holding context information, and privacy policy management means for managing privacy policy information,
The context holding means holds context information, node information, and access information,
The context information has 0 or 1 or more records including a node ID, 1 or more sets of property and value information, and a processing flag.
The node information has zero or one or more records including an ID, a class, and a value corresponding to the node ID or property value of the context information,
The access information has 0 or 1 or more records including an ID, a target user that describes a user or group that permits or does not permit reading, and an access right that describes permission or denial of reading,
The privacy policy information includes a node condition describing a target class or keyword, a target user describing a user or group that permits or does not permit reading, and an access right that describes permission or non-permission of reading. Have one or more records,
The first context management means includes:
Collect the context to create the context information and the node information,
The created context information and node information are matched with the above privacy policy information. When the node condition of the privacy policy information matches the context information or the node information, the target user and access of the privacy policy information are After copying to the access information, the context information, the node information and the access information are transmitted to the server, and in response to a user input, a context request is transmitted to the server.
The server has second context management means and second context holding means,
The second context holding unit holds the context information, the node information, and the access information transmitted from the client for each user,
The second context management means includes:
While the transmitted context information, the node information and the access information are held in the second context holding means,
When an arbitrary user context is requested by a user of an arbitrary client, when the requesting user is allowed to read with the access right related to the requested context of the requested user, the context A context sharing method characterized by returning information.   When the context of a certain condition of an arbitrary user is requested from the user of an arbitrary client, the second context management means matches the condition specified for the context information of the requested user, and the request source 5. The context sharing method according to claim 4, wherein the requested context information is returned when an access right is granted to the user. Each record of the access information further includes an anonymous flag, and each record of the privacy policy information further includes an anonymous flag,
When the context of a certain condition is requested from a user of an arbitrary client without designating the user, the second context management means matches the designated condition with respect to all context information, 6. The context sharing method according to claim 4, wherein the context information is returned when the access right of the requesting user is permitted. A context having one or more clients that acquire context information related to individual users and intermittently transmit the information to a server, and a server that returns information on the context of an arbitrary user requested by an arbitrary client under a predetermined condition In the above client in the shared system,
Context information having zero or one or more records including node ID, one or more property and value pair information, and a processing flag;
Node information having zero or one or more records including the ID, class and value corresponding to the node ID or property value of the context information;
Context that holds ID, access target that describes the user or group that permits or does not permit reading, and the access right that describes permission or denial of reading, and holds access information Holding means;
It has one or more records including node conditions that describe the target class or keyword, target users that describe users or groups that allow or do not permit reading, and access rights that describe permission or disapproval of reading A privacy policy management means for managing privacy policy information;
The context is collected to create the context information and the node information, and the created context information and node information are matched with the privacy policy information, and the node condition of the privacy policy information and the context information or node If the information matches, after copying the target user and access of the privacy policy information to the access information, the context information, the node information, and the access information are transmitted to the server, and in response to the user input. And a context management means for transmitting a context request to the server. A context having one or more clients that acquire context information related to individual users and intermittently transmit the information to a server, and a server that returns information on the context of an arbitrary user requested by an arbitrary client under a predetermined condition In the above server in the shared system,
Sent from the above client,
Context information having zero or one or more records including node ID, one or more property and value pair information, and a processing flag;
Node information having zero or one or more records including the ID, class and value corresponding to the node ID or property value of the context information;
Context that holds ID, access target that describes the user or group that permits or does not permit reading, and the access right that describes permission or denial of reading, and holds access information Holding means;
The context information, the node information, and the access information that have been transmitted are held in the context holding unit, and when an arbitrary user context is requested from an arbitrary client user, the requesting user A server comprising: context management means for returning the context information when reading is permitted with the access right related to the requested context of the requested user.

Images