KR20190056352A - A collaborative personal assistant system for delegating providing of services supported by third party task providers and method therefor - Google Patents

Abstract

Provided is a collaborative personal assistant (PA) server for delegating provision of services by third party task providers. The collaborative PA server comprises: a task delegation database for receiving a plurality of pattern descriptors including natural language input patterns and service queries from a plurality of external devices, and storing the pattern descriptors; a context inference unit for acquiring an input of a user requiring execution of at least one task which can be performed by one of the external devices, and identifying at least one pattern descriptor among the pattern descriptors based on the acquired input of the user; and a task delegation unit for transmitting a query request to at least one external device corresponding to the identified at least one pattern descriptor, and receiving a final response including an action list associated with the at least one task in response to the query request, from the at least one external device, wherein each of the natural language input patterns is mapped to at least one of the service queries, the input of the user is a form of voice or text, the at least one pattern descriptor includes a natural language input pattern matched with the acquired input of the user, and the at least one task is executed based on the query request.

Description

TECHNICAL FIELD [0001] The present invention relates to a collaborative personal assistant system for providing service of a third party task provider and a method thereof,

The present invention relates to a collaborative personal assistant system acting as a service provider of third party task providers and a method therefor.

In general, a virtual personal assistant (hereinafter, referred to as 'PA') system understands a user's intention and learns user profile and context information, And performs the tasks required by the users by applying them. The virtual PA system uses natural language processing to respond to the user's query, makes recommendations and performs actions corresponding to the task by proxying requests in an extension set of web services do.

1 is a structural diagram of a general virtual PA system.

Referring to FIG. 1, a virtual PA system includes a PA server 102, a network 104, a user device 106 and a third party task server 108. Here, the PA server 102 includes three technical components not shown in the figure. The technical components include a conversational interface, a personal context awareness unit, and a service agent. The dialog interface is a dialog-style natural language user interface that obtains natural language input from both a human speech or a user-typed query type query. The PA server 102 is aware of the personal contexts in advance. The personal contexts may include, for example, information about a place where the user is currently located, or where he or she lives or works. The PA server 102 acts as a proxy server for providing services to the corresponding user, which the third party task server 108 supports. Here, as an example, the case where there is one third-party task server is shown, but the PA server 102 can provide services that are supported by a plurality of third party task servers. The provision of the services supported by the third party task server 108 may include, for example, searching for small businesses / restaurants, local events and movie theaters, reserving restaurants etc., And answering the answers to the questions.

A typical virtual PA system 100 comprises a PA application running on the PA server 102, the user device 106 and the third party server 108. The user device 106 may be a personal digital assistant (PDA), an intelligent terminal that adds a computer support function such as internet communication and information search to a mobile phone, for example, a smart phone, a tablet ) And the like.

2 is a signal flow diagram for operation of a general virtual PA system.

2, the virtual PA system includes a user device 200, a PA server 202, and an external third party task server 204. [

In step 206, the user device 200 receives a user input from a user. At this time, the user input may be inputted in the form of a human speech (dictation) or a characterized query. The operating system (OS) or PA application of the user device 200 will pre-process the dormant speech dictated from the text by applying a speech to text technique.

In operation 208, the PA application operating on the user device 200 obtains a user request from the user input and transmits the user request to the PA server 202. The user request may include user queries and, in one example, additional context information (such as the location of the device), such as the location of the user device 200. At this time, the user query may be either a textual query form or an oral query form.

For example, Table 1 below shows the user request in the form of a message.

And analyzing the received user request with personal context information for identifying a domain to which the user is connected. At this time, an artificial intelligence technology such as 'ontology', 'inference' and 'pattern matching supporting unified modeling, dictionaries and dialog models' Lt; / RTI >

In step 214, the PA server 202 transmits an intermediate response to the user device 200 as a response to the PA application. The intermediate response is displayed to the user in the form of a text message or transferred to a verb. One example of this midterm response is "I'm looking for a Korean BBQ restaurant for your dinner today." Then, in step 216, the user device 200 displays the received intermediate response in the message format shown in Table 2 below.

Then, the PA server 202 constructs a query for delivering the task to the external third party task server 204 in order to perform the task requested by the user request. In step 218, Party task server 204 of the third party. Then, in step 220, the external third-party task server transmits a query response including the task corresponding to the query to the PA server 202. At this time, if there are a plurality of tasks to be provided by the external third party task server 204 in the inquiry response, in step 224, the PA server 202 determines the items to be finally displayed to the user Determines a final response based on the determined priority based on the selected items, and transmits the final response to the user device 200 in step 226. Then, in step 228, the user device 228 displays the final response on the screen of the user device 200.

The conventional virtual PA system as described above has the inflexibility due to the limited number of third party providers to provide the service to the user. And, even though there are a number of actions or services that the third party provider can provide, the virtual PA system is unable to provide other sets of actions or services requested by the user.

As an example, suppose that a user requests a service provided by a third party task server that is a music related site. At this time, the third party server provides various service sets such as playing music using streaming, music download, or purchase service. However, a user using the virtual PA system can use only the services of the third party server that can be provided through the virtual PA system. For example, if the PA application provided by the virtual PA system supports the operation of executing a music play application on a user device or executing a library for playing music contents, the PA application The user can not directly access (download or play) the contents.

If there is a case where temporary context information of the user is required among the actions provided by the third party task server. One example of the actions may be a tarot point, a reservation service, or the like, which is generated based on a channel service or corresponding user information, and the temporary context information may include location information or user profile information (telephone number). The virtual PA system has inconvenience to confirm acceptance as to whether the temporary context information can be provided to the third party task server when the temporary context information is requested to the user.

The present invention proposes a collaborative PA system and a method therefor acting as a service provider of third party task providers.

The present invention proposes an interface for proxy services to be provided by a plurality of third party task servers.

The present invention provides additional services and actions that can be provided through a third party task server that provides a task required by user input.

In the present invention, the collaborative PA system delivers the personal profiles required for the task corresponding to the user input of the user to the corresponding third party task server according to the user's permission.

The apparatus according to an embodiment of the present disclosure includes a Collaborative Personal Assistant server for providing services of third party suppliers, and includes natural language input patterns and service queries from a plurality of external devices A task proxy database for receiving the plurality of pattern descriptors for storing the plurality of pattern descriptors and an input of a user requiring execution of at least one task executable by one of the plurality of external apparatuses A context inference unit for identifying at least one pattern descriptor among the plurality of pattern descriptors based on the obtained input of the user, and a query request unit for transmitting a query request to at least one external device corresponding to the identified at least one pattern descriptor And from the at least one external device And a task agent for receiving a final response including an action list associated with the at least one task in response to the query request, wherein each of the natural language input patterns is mapped to at least one of the service queries, Wherein the user input is in the form of voice or text and the at least one pattern descriptor comprises a natural language input pattern that matches the obtained user input and the at least one task is executed based on the query request.

A method according to an embodiment of the present disclosure is a method for providing a service of a third party provider using a Collaborative Personal Assistant server, the method comprising: receiving natural language input patterns and service queries from a plurality of external devices The method comprising: receiving a plurality of pattern descriptors including a plurality of pattern descriptors and storing the plurality of pattern descriptors; obtaining an input of a user requiring execution of at least one task executable by one of the plurality of external apparatuses; Identifying at least one pattern descriptor among the plurality of pattern descriptors based on the obtained input of the user, transmitting a query request to at least one external device corresponding to the identified at least one pattern descriptor From said at least one external device, Receiving a final response comprising a list of actions associated with the at least one task in response to a request of the user, wherein each of the natural language input patterns is mapped to at least one of the service queries, Wherein the input is in the form of voice or text and the at least one pattern descriptor comprises a natural language input pattern that matches the obtained user input and the at least one task is executed based on the query request.

The present invention relates to a collaborative PA system acting as a service provider of third party task providers, wherein the collaborative PA system comprises a third party task server providing the task required by the user input from the user input of the user and a query matching the task A third party task server that recognizes user input more systematically and provides a task corresponding to the user input by providing an interface for registering natural language input patterns for helping to find commands in a third party task agency, Can be found more easily. The collaborative PA system provides additional services and actions that can be provided through a third party task server that provides a task required by the user input, regardless of the user's request, thereby providing a limited third party task server It is possible to overcome the limit of non-inferiority caused by Finally, the user's permission to use the personal profile information is acquired in advance, and the information is transmitted to the third-party task server. When providing the task to the user, a procedure for requesting and acquiring the personal profile information of the user .

1 is a structural view of a typical virtual personal assistant system.
2 is a signal flow diagram for operation of a general virtual PA system;
3 is a configuration diagram of a collaborative PA system in accordance with an embodiment of the present invention;
4A is a signal flow diagram of a cooperative PA system according to a first embodiment of the present invention.
4B is a diagram illustrating an example of a final response provided to a user according to the first embodiment of the present invention;
5A is an example of a signal flow diagram of a cooperative PA system according to a second embodiment of the present invention.
FIG. 5B is a view showing an example of a screen of a user apparatus on which a service proxy operation of a cooperative PA system according to a second embodiment of the present invention is displayed; FIG.
6A is an example of a signal flow diagram of a cooperative PA system according to a third embodiment of the present invention.
FIG. 6B is a view showing an example of a screen of a user device displaying a request for sharing a user profile according to the third embodiment of the present invention; FIG.
7 is an example of a signal flow diagram of a preprocessing procedure for user identification between a PA server and a third party task server according to an embodiment of the present invention;

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, only parts necessary for understanding the operation according to the present invention will be described, and the description of other parts will be omitted so as not to obscure the gist of the present invention.

The present invention proposes a collaborative PA system acting as a service provider of third party providers and a method therefor.

The collaborative PA system according to the first embodiment of the present invention provides an interface for providing services supported by a plurality of third party task servers. The interface is configured to allow the collaborative PA system to retrieve from the user's input a natural language input pattern to assist in finding a third party task server providing the task required by the user input and query instructions matching the task, To register for the Department. Accordingly, the narrow-minded PA system according to the present invention is configured to automatically generate a query for functions for performing proxy operation of the third party task server through the interface.

The collaborative PA system according to the second embodiment of the present invention provides additional services and actions that can be provided through a third party task server that provides the task required by the user input, regardless of the user's request. Specifically, the collaborative PA system transmits an action service end point URL that can receive additional services other than the corresponding service through a user's touch or click action on the user device. Accordingly, the collaborative PA system further includes a context reasoning unit that can identify the indicated action from a programming language such as play, download, and the like. The context reasoning unit stores contents object meta data including natural language verbal words that identify an action to be taken on the corresponding content object.

The collaborative PA system according to the third embodiment of the present invention provides a third party task agent for registering personal profiles required for performing a task corresponding to a user input of the user. The collaborative PA system obtains a user permission to share personal profile information, such as the location or phone number of the user, during an operation to retrieve a result corresponding to the user input of that user. The collaborative PA system transmits the personal profile information of the user to the third party task server that provides the task to process the task requested by the user.

3 is a configuration diagram of a cooperative PA system according to an embodiment of the present invention.

3, the collaborative PA system 300 includes a PA server 310, a user device 320, a network 330, and a third party task server 340.

First, the PA server 310 includes a personal context database (DB) 312, a chat robot 316, a task proxy DB 317, a natural language processor (NLP) 313 Contextual reasoned and task delegate agent 315. The user device 320 includes a personal assistant application and a mobile operating system (OS) 324 The third party task server 340 includes a query processor 342 and an action processor 344. The user device 320 and the third party task server 340 are connected to a general user device and a third party And operates in the same manner as the task server, and thus a detailed description thereof will be omitted here.

The context reasoning unit 314 analyzes the received user input together with the personal context information of the corresponding user to identify the domain of the tasks satisfying the user request. The contextual reasoning unit 314 performs pre-processing of user input dictated from the text by applying a speech to text technique. The context reasoning unit 314 identifies a domain providing a task corresponding to the user input. Then, the task proxy agent 315 performs the information, which is extracted from the user input obtained using inference or dialog, and the task specified together with the user's personal context. The task proxy DB 317, for example, stores registration information from a plurality of third party task servers including the third party task server 340 of FIG. The third party task server 340 represents web service providers that supply application programming interfaces to provide domains supporting specific services such as knowledge search engines, restaurant bookings / searches, and the like.

Hereinafter, the operation according to the embodiments of the present invention will be described as the operation of the PA server itself, not for each of the detailed configurations constituting the PA server for convenience of explanation.

4A is a signal flow diagram of a cooperative PA system according to a first embodiment of the present invention. Here, user queries and corresponding tasks are provided through pre-registered natural language input patterns. Here, it is assumed that the third party task server 404 provides a task corresponding to the user input when the user inputs a user input through the PA application provided by the PA server 402. [

Referring to FIG. 4A, in step 406, an operator of the third party task server 404 manages a service for providing services supported by the third party task server 404, And transmits a registration request of a pattern descriptor to the PA server 402. The pattern descriptor for the query includes a web service query for proxy operation of a task proxy agent (corresponding to 315 in Fig. 3) included in the PA server 402, and a natural language input pattern mapped to the web service query . Wherein the PA server includes a third party that supports a task corresponding to a web service query mapped to the one input pattern and the natural language input pattern when one of the natural language input patterns stored in the PA server matches a user input, Performs proxy operation of the task server. For example, the pattern descriptor for the query is shown in Table 3 and Table 4 below. The pattern descriptor for the query includes a category of service for the task and an end point uniform resource locator (URL) of the third party task server to be substituted.

In step 408, the PA server 402 registers a descriptor for the query and transmits a registration response to the third party task server 404.

In step 410, the user drives the PA application 400 on the user device and delivers the user input to the PA server 402. Here, the user input is delivered in a natural language form, either in words or text. In step 412, the PA server 402 checks whether the user input requests an agent service of the third party task server. As a result of the inspection, when the user input requests an agent service of the third party task server, context inference of the user input is performed using the pattern descriptor based on the previously registered query. Specifically, when the user input is matched with one of the natural language input patterns stored in the PA server 402, the PA server 402 generates a task corresponding to the one input pattern and the web service query mapped to the input pattern Identify the domain of the supported third-party task server. In step 414, the PA server 402 delivers a message-type interim response to the user through the PA application 400 to indicate 'searching for the identified domain'.

Then, in step 416, the task proxy agent of the PA server 402 configures the personal profile information for the user (including, for example, a query command and parameters for requesting the user identifier) as a query request, To the party task server (404). Then, in step 418, the third party task server 404 transmits a query response to the PA server 402. The query response includes an item object supporting the task requested by the user input and an action set allowed for each item object. For example, the query response is expressed in the form of Table 5 to Table 7 below.

In step 420, the PA server 402 transmits a final response to the user input through the PA application 400. [ The final response is shown, for example, in the form of FIG. 4B.

4B is an example of a final response provided to a user according to the first embodiment of the present invention.

Referring to FIG. 4B, 'what can I do for you?' 430 is an example of a query displayed on the user device by the PA application 400. The 'Find cozy place for Korean BBQ' 435 corresponds to an example of textual representation of the user input 410 transmitted to the PA server 402 in step 410 of FIG. 4A.

Let me check the Korean ~ in reference numeral 440 represents an example of the intermediate response that the PA server 402 provides to the user through the PA application 400 in step 414 of Figure 4A. Finally, reference numeral 445 denotes an example of a final response corresponding to step 420 of FIG. 4A. Specifically, two final responses to user input requesting a venue request for a Korean BBQ are displayed. Then, additional actions for each of the two places, that is, a set of actions capable of performing a call and a reserve action, are displayed together.

5A is an example of a signal flow diagram of a cooperative PA system according to a second embodiment of the present invention. Here, the PA server 502 provides the user with a set of additional services and actions that can be provided through a third-party task server that provides the task required by the user input. At this time, the PA server enables the action to be performed by transmitting a service end point URL of an action that can receive additional services other than the corresponding service through a user's touch or click action on the user device. In addition, the content object metadata previously stored by the inquiry inference unit includes natural language verb words. The words are used to identify actions that can be performed on the content object, and the PA server can analyze and identify the corresponding actions in a programming language such as play, download, etc. through natural language context reasoning.

Referring to FIG. 5A, in step 506, a user drives a PA application 500 on a user device and delivers a user input to the PA server 502. Here, the user input is delivered in a natural language form, either in words or text. Then, in step 508, the PA server 502 checks whether the query requested by the user input requests an agent of the third party task server. If it is determined that the query is a request for substitution of the third party task server, context inference for the user input is performed using the pattern descriptors for the query stored in the procedure described in FIG. 4A. Specifically, when the user input matches one of the natural language input patterns stored in the PA server 502, the PA server 502 generates a task corresponding to the one input pattern and the web service query mapped to the input pattern Identify the domain of the supported third-party task server. In step 510, the PA server 502 transmits a message type interim response to the user through the PA application 500 indicating 'searching for the identified domain'.

Thereafter, in step 512, the PA server 502 configures the personal profile information for the user (including, for example, a query command and parameters for requesting the user identifier) as a query request, To the third party task server (504). At this time, the parameters are configured from the user input with a personal context, a conversation context, and an inferred context. At this time, the inquiry request includes a user identifier, for example, personal profile information of the user in order to identify the user and the permission of the action the user subscribed to, by the operator of the corresponding third party task server. In step 514, the third party task server 504, which has received the inquiry request, transmits a query response corresponding to the inquiry request to the PA server 502. Specifically, the third party task server 504 determines whether or not the user has joined the server, and if the user is subscribed, the performance of the user device supports the object corresponding to the user input , Then the third party task server 504 includes a list of the actions allowed by the subscription of the user in the query response. And may further include further actions that may be performed by the user after the allowed actions are performed. One example of the additional actions corresponds to a natural language text that can be presented to the user through the PA server 502 and an action to play music. As a specific example, assume that a user is not subscribed to a third party task provider that is a music play site. In this case, the third party task provider sends to the PA server a list of actions such as 'one minute free listening' or 'join' action that can be provided to the user. If the user is subscribed to the 'streaming play' service but not the 'download service', the third party task server transmits only 'play' action to the PA server. However, if the user is subscribed to both the 'streaming play' and the 'download service', the third party task server transmits both the 'streaming play' and the 'download service' actions to the PA server.

In step 514, the PA server 502 receives the query response from the third party task server 504, and then stores the language keyword of the action provided by the query response in the personal context DB. The storing operation is for helping understanding of a user input to be input later including the stored keyword.

In step 516, the PA server 502 determines the priorities of the items included in the query response and supports the task, and transmits a final response, which is composed of the determined items, to the user through the PA application 500 . Then, in step 518, the PA application 500 can display the items obtained through the final response and the actions permitted to the user for each item. Then, in step 520, the user enters a second user input to select one of the displayed actions via the PA application 500. [ At this time, the input method corresponds to touch, click, or speaking. In step 522, the PA server 502 receiving the second user input interprets the second user input as 'music play', using the stored personal context information. In step 524, the PA application 500 transmits a response to the user. The response includes, for example, a content object identifier such as an action requested by the user and a URL corresponding to the interpreted second user input.

Thereafter, in step 526, the PA application 500 determines whether the music play library or the music play application on the user device that can access the third party task server 504, which is identified by the end point URL obtained through the response, . Then, in step 528a, the PA application 500 transmits an action request including the user identifier and the object identifier to the third party task server 504 for further action. Then, in step 528b, the third party task server 504 provides the streaming service to the user through the PA application 500 as an example of a corresponding service corresponding to the object identifier.

FIG. 5B is a view showing an example of a screen of a user apparatus on which a service proxy operation of the cooperative PA system according to the second embodiment of the present invention is displayed.

Referring to FIG. 5B, reference numeral 501 denotes a start screen of the PA application 500. That is, the PA application 500 displays " what can I help you with? &Quot; as an example of a query sentence for requesting a user input.

The 'what is lonely?', Which is an example of the response of the query, corresponds to the first user input in step 506 of FIG. 5A. Herein, the case of inputting the response of the query as a sentence is shown as an example, The response may be input to the user device in the form of speech as described above.

Next, the description ("Lonely is a music ~ You can play or download the song.") Provided as a response to the query response in FIG. 5B is obtained in step 510 .

Reference numeral 515 corresponds to the final response in step 516 of FIG. 5A. Here, as an example of the above-mentioned question and answer, a sound source service such as a sound source title and a play time is provided, and a play and a download are given as an example of actions allowed to the user.

Here, it is assumed that the user selects 'play' as an example of a specific action through the second user input in step 520 of FIG. 5A. Then, the PA server 500 displays 'Could you play the song?' As a response to the selection of the action on the user device in a sentence. The response corresponds to step 524 of FIG. 5A. The PA application 500 provides the corresponding music together with a message ("OK I will play the song 'lonely") that provides music play to the user, .

6A is an example of a signal flow diagram of a cooperative PA system according to a third embodiment of the present invention. The collaborative PA system according to the third embodiment of the present invention includes a task corresponding to a request of a user and a proxy service for registering personal profiles required for providing the object by a third party task server providing an object supporting the task to provide. That is, while the PA server retrieves results according to user input, the PA server obtains the user permission to share the user profile information (e.g., location, telephone number, etc.), and transmits the user profile information to the corresponding third party task server It handles the action requested by the user by forwarding.

 Referring to FIG. 6A, in step 606, the user drives a PA application 600 on the user device and delivers the user input to the PA server 602. Here, the user input is delivered in a natural language form, either in words or text. Then, in step 608, the PA server 602 checks whether the query requested by the user input requests an agent of the third party task server.

If it is determined that the query is a request for substitution of the third party task server, context inference for the user input is performed using the pattern descriptors for the query stored in the procedure described in FIG. 4A. Specifically, when the user input is matched with one of the natural language input patterns stored in the PA server 602, the PA server 602 searches for a task corresponding to the one input pattern and the web service query mapped to the input pattern Identify the domain of the supported third-party task server. In step 610, the PA server 602 transmits an interim response in the form of a message indicating 'searching for the identified domain' to the user through the PA application 600.

Thereafter, in step 612, the PA server 602 configures a query command and parameters for requesting the user identifier as a query request, for example, as personal information on the user, and transmits the query request and parameters to the third party task server 604). The parameters are configured from the user input with a personal context, a conversation context, and an inferred context. At this time, the inquiry request includes a user identifier as an example of the personal information of the user in order that the corresponding third party provider can identify the user and the action that the user subscribes. In step 614, the third party task server 604 receiving the inquiry request transmits a query response corresponding to the inquiry request to the PA server 602. Specifically, the third party task server 604 includes an action descriptor including personal profile information required by the inquiry request.

The action descriptor is shown in Table 8 below.

In step 616, the PA server 602 transmits the data corresponding to the results of the query together with the permission request of the user to the PA application 600 as a final response. Then, in step 618, the PA application 600 displays the content objects included in the final response to the user through the screen of the user device.

FIG. 6B is a diagram illustrating an example of a screen of a user device displaying a request to share user profile information of a user according to a third embodiment of the present invention.

Referring to FIG. 6B, for example, a Korean BBQ restaurant near the user location is checked ("Let me check the Korean BBQ restaurant which is cozy you're your location"). The intermediate response corresponds to step 610 of FIG. 6A.

The PA server 602 then displays the final response including the restaurant information retrieved along with the intermediate response through the user device, e.g., name, distance and telephone number, and actions that the user can select. The final response of the above form corresponds to step 618 of FIG. 6A. For example, assume that the user selects 'reservation' when the actions that can be selected by the user include a call and a reserve.

 At this time, the PA server 602 transmits a request for use of the user's personal information together with a call and a reservation of the actions that the user can select.

Assume that in step 620, the user transmits 'allow' for the request to the PA server 602 through the PA application 600 as a second user input. Then, the PA server 602 transmits a response to the reception of the second user input through the PA application 600. In step 624, the PA server 602 transmits to the third party task server 604 an action request including the action requested by the user and personal profile information such as a phone number or location information requested as an example of the action. . As described above, since the user has selected " reservation " as the action, the requested personal profile information is A user telephone number or location information, and the like. In step 626a and step b, the third party task server 604 transmits a user response corresponding to the action to the PA application 600 via the PA server 602. [

7 is an example of a signal flow diagram of a preprocessing procedure for user identification between a PA server and a third party task server according to an embodiment of the present invention. Here, the PA server must be accompanied by a preprocessing procedure for associating user information managed by each of the PA server and the third party task server in order to recognize the user's permission to the task and the content object requested by the user. A typical third-party task server manages identifiers of subscribed users. In the second to third embodiments of the present invention, the pre-processing operation is required.

The preprocessing operation is performed while any user subscribes to any third party task server. For example, if the third party task server displays a list of PA servers to the user during the subscription procedure, the user selects one PA server from the list. At this time, the third party task server requests the user to identify the PA server. Through this, during the subscription procedure, the third party server is able to obtain the identifier of the PA server to which the user is associated.

Referring to FIG. 7, in step 706, the user performs a subscription procedure to the third party task server 704 through the PA application 700 running on the user apparatus. Here, the case where the third party task server 704 is one is described as an example. However, each of the third party task servers providing the service to be PA server 702 should be accompanied by a preprocessing procedure for user identification something to do.

When the subscription procedure is completed, the third party task server 704 delivers the user information of users subscribed to the third party task to the PA server 702 in step 708. For example, the user information managed by the third party task server 704 is shown in Table 9 below.

Meanwhile, the user information managed by the PA server 702 may be represented as shown in Table 10 below.

The collaborative PA system as described above performs a service operation of a third party task server that provides a task and an object corresponding to a user input so that the user directly accesses the third party task server to acquire the task and the object It is possible to reduce unnecessary procedures and to solve the existing problems as compared with the operation.

While the present invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not limited to the disclosed embodiments, but is capable of various modifications within the scope of the invention. Therefore, the scope of the present invention should not be limited by the illustrated embodiments, but should be determined by the scope of the appended claims and equivalents thereof.

Claims (12)

CLAIMS 1. A Collaborative Personal Assistant server for providing services of third party suppliers,
A task proxy database for receiving a plurality of pattern descriptors including natural language input patterns and service queries from a plurality of external devices and storing the plurality of pattern descriptors,
Acquiring an input of a user requiring execution of at least one task executable by one of the plurality of external apparatuses and identifying at least one pattern descriptor of the plurality of pattern descriptors based on the input of the obtained user A context inference unit,
The method comprising the steps of: sending a query request to at least one external device corresponding to the identified at least one pattern descriptor; and sending, from the at least one external device, And a task agent for receiving a response,
Wherein each of the natural language input patterns is mapped to at least one of the service queries,
The input of the user is in the form of voice or text,
Wherein the at least one pattern descriptor includes a natural language input pattern matched with an input of the obtained user,
Wherein the at least one task is executed based on the query request. The method according to claim 1,
Wherein the task agency comprises:
And displays an intermediate response in the form of a text message in response to the obtained input of the user. The method according to claim 1,
Wherein the action list includes additional services and actions associated with the at least one task,
Wherein the task agency comprises:
And displaying additional services and actions associated with the at least one task. The method according to claim 1,
The action list includes:
The items of support for the at least one task and the set of actions allowed for each of the items. The method of claim 3,
Wherein the task agency comprises:
When the additional service and actions request the user's personal information, transmits an acceptance request for use of the user's personal information to the user device, and, when receiving the acceptance of use of the personal information from the user device, And delivering the user's personal information to the at least one external device to provide the additional services and actions to the user device. In a method for providing service of a third party provider using a Collaborative Personal Assistant server,
Receiving a plurality of pattern descriptors including natural language input patterns and service queries from a plurality of external apparatuses and storing the plurality of pattern descriptors;
Obtaining an input of a user requiring execution of at least one task executable by one of the plurality of external apparatuses;
Identifying at least one pattern descriptor among the plurality of pattern descriptors based on the obtained user input;
Transmitting a query request to at least one external device corresponding to the identified at least one pattern descriptor;
Receiving a final response from the at least one external device, the response including a list of actions associated with the at least one task in response to the query request;
Wherein each of the natural language input patterns is mapped to at least one of the service queries,
The input of the user is in the form of voice or text,
Wherein the at least one pattern descriptor includes a natural language input pattern matched with an input of the obtained user,
Wherein the at least one task is executed based on the query request. The method according to claim 6,
And displaying an intermediate response in the form of a text message in response to the obtained input of the user. The method according to claim 6,
Wherein the action list includes additional services and actions associated with the at least one task,
Further comprising displaying additional services and actions associated with the at least one task. ≪ RTI ID = 0.0 > 31. < / RTI > The method according to claim 6,
The action list includes:
Further comprising items supporting the at least one task and an action set allowed for each of the items. 9. The method of claim 8,
Transmitting an acceptance request for use of the personal information of the user to the user device when the additional services and actions request the user's personal information;
And providing the additional service and actions to the user device by transmitting the user's personal information to the at least one external device when the user device receives an acceptance of use of the personal information from the user device To provide services to third party suppliers. The method according to claim 6,
And sending information about an action list associated with the at least one task to a user device. The method according to claim 6,
Wherein the information about the action list associated with the at least one task is transmitted to the user device in response to an input from the user device and displayed on the user device.

Images