KR20210116741A - High developed intelligence collaboration assistant system - Google Patents

Abstract

The present invention relates to a highly developed intelligent collaboration support system which intelligently integrates and manages various messages, that are scattered and sporadic and received from a variety of devices, and analyzes them according to an intention, is Integrated with a variety of internal and external collaboration services in a group to support and can be associated with a plurality of internal or external collaboration systems, thereby increasing user convenience.

Description

Advanced intelligent collaboration support system {High developed intelligence collaboration assistant system}

The present invention relates to groupware, and more particularly, to an advanced intelligent collaboration support system.

Recently, due to various communication-oriented work types, there is a trend of changing to a complex work form that crosses multiple collaboration spaces using multiple collaboration tools that can be individually selected. Existing distributed and fragmented collaboration tools have a problem in that it is difficult to adequately respond to changes in these complex work types. In addition, when a large number of messages are generated in a state where formal and informal collaboration tools are mixed, there is a risk of lowering work productivity.

Therefore, intelligent collaboration support that can improve user convenience by intelligently integrating and managing various scattered and sporadic messages input from various devices, analyzing them according to the intention, and integrating and supporting various internal and external collaboration services The need for a system is emerging.

Korean Patent Laid-Open Patent No. 10-2018-0081021 (July 13, 2018) discloses a business strategy map management server that supports collaboration when a plurality of business performing organizations are involved in the messenger interworking with ERP. When the business strategy map management server manages the business project, the project message thread includes a technical means called 'to-do-evaluation item'.

A project message thread refers to a series of messages related to a specific project. Multiple message threads can be related to one project, and this information is managed as a list. The task list is related to this 'to-do-evaluation item' (this association is determined according to the user's task), and the users use the messenger on the task strategy map terminal, and the 'to-do-evaluation' specific to a specific project message thread items can be inserted. At this time, the task list is updated according to the inserted item.

This prior art presents the concept of a messenger thread, and presents the concept of managing the task/to-do by using the messenger thread. It discloses a management technology with the concept of In this prior art, responsibilities are assigned by manual assignment in which messenger users insert to-do-evaluation items.

Korean Patent Registration No. 10-1881114 (July 17, 2018) proposes a technology for automatically filling a related user interface by processing a message in natural language. A related task is determined based on scoring for a candidate user task and a rule path, and the user interface of the task is selected and presented and this user based on message analysis The data entry fields of the interface are automatically populated.

US Patent No. 9,928,229 (March 27, 2018) discloses a technology for pre-filling a form with information related to an action command found in a message. In this technology, information extracted by analyzing a natural language message is compared with input fields required in a candidate target form, and a target form is determined by scoring.

However, in these prior arts, the execution of the task is necessarily handled by the user. In addition, since the determination of a task is determined by a single message, there is a limit to be applied to the modern complex and complex business tasks.

Republic of Korea Patent Publication No. 10-2018-0081021 (2018.07.13) Republic of Korea Patent Registration No. 10-1881114 (2018.07.17) US registered patent US 9,928,229 (2018.3.27)

The present invention intelligently integrates and manages various scattered and sporadic messages input from various devices, analyzes them according to the intention, integrates them into various internal and external collaboration services and supports them, but works with multiple internal or external collaboration systems. Its purpose is to provide an advanced intelligent collaboration support system that can improve user convenience by interworking with

Another object of the present invention is to provide an advanced intelligent collaboration support system capable of intelligently determining a collaboration task through message intent analysis.

Another object of the present invention is not only to fill the individual work slots constituting the work story for each collaborative work unit, but to execute a collaborative work action when it is filled, so that a more reliable advanced intelligent collaboration support system is to provide

Another object of the present invention is to provide an advanced intelligent collaboration support system with excellent cosmetic properties that is linked to a natural language processing platform capable of supporting multiple languages.

According to an aspect of the present invention for achieving the above object, an advanced intelligent collaboration support system performs collaborative tasks through context-based message intent analysis of messages delivered between a plurality of collaboration participant terminals. An intelligent collaboration support platform for performing asynchronous session management and interaction with a collaborative task corresponding to message intent; A collaboration scenario is defined by defining a work story for each collaborative work unit, defining individual work slots constituting a work story for each defined collaborative work unit, and defining work processing steps for individual work slots. ), but according to the collaboration task call from the intelligent collaboration support platform, the task processing steps included in the individual task slots constituting the task story for each collaborative task unit are executed. A channel forming unit including an intelligent collaboration system, wherein the intelligent collaboration support platform connects a communication channel with external collaboration systems; Executes commands of internal systems or external collaboration systems including intelligent collaboration systems, interprets requests from internal systems or external collaboration systems, and calls conversions to APIs Converted and displayed command cell (Command Shell) and; an SNS linkage unit for exchanging real-time messages with a social network service (SNS) server that provides a social network service; a web linkage unit for exchanging real-time messages with web servers providing web services; a mail link unit for exchanging real-time messages with mail servers providing mail services; an internal system interworking unit for exchanging real-time messages with an internal system including an intelligent collaboration system; Collaboration interlocking control unit that controls overall collaboration interworking including communication channel setting to be connected through the channel forming unit, command processing through command cells, message exchange control through SNS interworking unit, web interworking unit, mail interworking unit, and internal system interworking unit includes

According to an additional aspect of the present invention, an intelligent collaboration support platform includes a communication unit that performs frame/slot-based S2S (System to System) communication with an internal system or external collaboration systems including an intelligent collaboration system. an interaction unit comprising; It further includes software robotics (SW Robotics) for executing an application corresponding to a frame/slot-based S2S (System to System) interaction performed by the interactor.

According to an additional aspect of the present invention, the NLP platform that performs natural language processing (NLP) including morpheme analysis, entity name analysis, dialogue act analysis, and intention analysis required for the intelligent collaboration support platform to analyze a message is interlocked It further includes an NLP platform adapter.

According to an additional aspect of the present invention, the interaction unit comprises: a text analysis unit for deriving a collaboration participant associated with the message through text analysis of messages transmitted between a plurality of cooperative participant terminals; a session analysis unit for deriving a collaborative work unit related to the message through session analysis of messages transmitted between a plurality of collaboration participant terminals, and managing the continuity of the derived collaborative work unit; It further includes a context analysis unit including an intention analysis unit for extracting the intent of the collaboration participant derived by the text analysis unit through the analysis of intent (intent) for messages transmitted between a plurality of collaboration participant terminals.

According to an additional aspect of the present invention, software robotics performs a task processing step corresponding to the intention of the collaboration participant among the task processing steps included in individual task slots constituting the task story of the collaborative task unit determined by the session analysis unit. The running business process is called from the intelligent collaboration system.

According to an additional aspect of the present invention, software robotics defines CRUD (Create, Read, Update, Delete) according to URI (Uniform Resource Identifier) and basic methods (POST, GET, PUT, DELETE) of http, and the corresponding method When called, it calls a REST (REpresentational State Transfer)-based business process that accesses an objectified business processing step.

According to an additional aspect of the present invention, the intention analysis unit may be implemented to extract multiple intentions based on a rank.

According to an additional aspect of the present invention, the intention analysis unit may be implemented to determine the intention according to priority when the intention of the collaboration participant has multiple intentions.

According to an additional aspect of the present invention, software robotics performs NER (Named Entity Recognition) and fills individual task slots constituting a task story for each collaborative task unit from messages transmitted between collaboration participant terminals. It will be implemented to extract the intentions required for the task, fill the extracted intentions in individual task slots, and call the business process that executes the collaborative task action from the intelligent collaboration system when the individual task slots are all filled. can

According to an additional aspect of the present invention, the interactor may further include a learning unit that analyzes the atypical messages transmitted from the external collaboration system through learning, converts it into a standardized intention usable in the internal intelligent collaboration system, and learns.

The present invention intelligently integrates and manages various scattered and sporadic messages input from various devices, analyzes them according to the intention, integrates them into various internal and external collaboration services and supports them, but works with multiple internal or external collaboration systems. It has the effect of improving user convenience by being able to interlock.

In addition, since the present invention can execute a business process by intelligently determining a collaboration task through message intent analysis, it is possible to minimize user selection for collaboration.

In addition, the present invention has an effect that more reliable collaboration support is possible by not only filling individual task slots constituting the task story for each collaborative task unit, but executing a collaborative task action when they are filled.

In addition, the present invention can be linked with a natural language processing platform capable of supporting multiple languages, thereby enabling advanced intelligent collaboration support with excellent scalability.

1 is a block diagram showing the configuration of an embodiment of the advanced intelligent collaboration support system according to the present invention.
2 is a diagram illustrating a frame/slot-based structure for performing S2S (System to System) communication.
3 is a diagram illustrating a context-based asynchronous session management concept.
4 is a diagram illustrating extracting multiple intentions based on a rank.
5 is a diagram illustrating a work story of a schedule-related collaboration task unit in which schedules are coordinated and registered among collaboration participants.
6 is a block diagram showing the configuration of an embodiment of the collaboration interworking adapter of the intelligent collaboration support platform of the advanced intelligent collaboration support system according to the present invention.

Hereinafter, the present invention will be described in detail so that those skilled in the art can easily understand and reproduce it through preferred embodiments described with reference to the accompanying drawings. While specific embodiments are illustrated in the drawings and the related detailed description is set forth, they are not intended to limit the various embodiments of the present invention to a specific form.

In describing the present invention, if it is determined that a detailed description of a related well-known function or configuration may unnecessarily obscure the gist of the embodiments of the present invention, the detailed description thereof will be omitted.

When a component is referred to as being “connected” or “connected” to another component, it is understood that the other component may be directly connected or connected to the other component, but other components may exist in between. it should be

On the other hand, when it is mentioned that a certain element is "directly connected" or "directly connected" to another element, it should be understood that the other element does not exist in the middle.

1 is a block diagram showing the configuration of an embodiment of an advanced intelligent collaboration support system according to the present invention. As shown in FIG. 1 , the advanced intelligent collaboration support system according to this embodiment includes an intelligent collaboration support platform 100 and an intelligent collaboration system 200 . At this time, the intelligent collaboration support platform 100 and the intelligent collaboration system 200 may be implemented in the form of software, and may be mounted on the same or separate device.

Intelligent collaboration support platform 100 manages an asynchronous session for collaborative tasks through context-based message intent analysis for messages delivered between a plurality of collaboration participant terminals (not shown) and performing an interaction with a collaborative task corresponding to the message intent.

In this case, the message transmitted between the plurality of collaboration participant terminals may include a message transmitted and received through internal or external mail, a message transmitted and received through a social network service (SNS), a message shared on a group sharing bulletin board, etc., but limited to this it doesn't happen

For example, the intelligent collaboration support platform 100 may be implemented to intercept messages transferred between a plurality of collaboration participant terminals, and to perform context-based message intent analysis on the intercepted messages.

Meanwhile, collaboration tasks include payment collaboration tasks for sharing approval, mail collaboration tasks for mail sharing, schedule collaboration tasks for sharing schedules, meeting collaboration tasks for meetings, and IPA collaboration tasks for personal assistant (IPA) functions. can, but is not limited thereto.

On the other hand, asynchronous session management for collaborative tasks means that each collaborative task is divided to ensure continuity of each collaborative task because a number of different collaborative tasks are performed at random, and each collaborative task is processed. It means that the details are stored in the session memory and managed individually.

On the other hand, the interaction with the collaborative task refers to calling and executing the business process for performing the collaborative task from the intelligent collaboration system 200 .

The intelligent collaboration support platform 100 may be built in the form of a Micro Service Architecture (MSA)-based framework. By applying MSA, it is possible to apply an architecture that provides a development API by dividing a single application program, and integrates them to implement a combination in a service unit.

In addition, it is possible to minimize changes by applying the REST (REpresentational State Transfer) protocol to encapsulate scalability, versatility, independent distribution of components, and legacy systems.

In addition, the response type can be determined based on the application, device, location, etc. used by the user through the user's application and data-based context adaptive context management, and user profile information, application information, and user state information are managed. possible, and through asynchronous session management for collaborative tasks, the continuity of collaborative task processing can be guaranteed.

The intelligent collaboration system 200 performs task processing steps included in individual task slots constituting a work story for each collaborative task unit according to a collaborative task call from the intelligent collaboration support platform 100 . execute (execution).

To this end, the intelligent collaboration system 200 defines a work story for each collaborative work unit, defines individual work slots constituting a work story for each defined collaborative work unit, and performs tasks for individual work slots. Define a processing step to create a business scenario.

The intelligent collaboration system 200 may include a plurality of business processes for processing business processing steps defined in individual business slots constituting a business story for each cooperative business unit.

For example, these business processes include an approval collaboration business process 210 for sharing approval, a mail collaboration business process 220 for sharing mail, a schedule collaboration business process 230 for sharing a schedule, and a meeting collaboration business process for a meeting ( 240), an IPA collaboration work process 250 for a personal assistant (IPA) function, etc., but is not limited thereto.

On the other hand, the intelligent collaboration system 200 can overcome the password use problem by separating the authentication protocol and the authentication means and authenticate without a password. It may further include a Logger process 270 for logging to 300 .

Meanwhile, according to an additional aspect of the invention, the intelligent collaboration support platform 100 may include the interactor 110 and software robotics (SW Robotics) 120 .

The interaction unit 110 includes a communication unit 111 that performs frame/slot-based S2S (System to System) communication with internal or external collaboration systems including the intelligent collaboration system 200 .

2 is a diagram illustrating a frame/slot-based structure for performing S2S (System to System) communication.

The intelligent collaboration system 200 defines a work story for each collaborative work unit, defines individual work slots constituting a work story for each defined collaborative work unit, and performs work processing steps for individual work slots Define (step) to create a business scenario.

The intelligent collaboration support platform 100 is a frame / A request is made to the intelligent collaboration system 200 using a slot-based structure.

A frame/slot-based structure is a knowledge expression technique that expresses typical knowledge about a specific object or concept as a set of slots. In the present invention, a frame is described in units of stories, and is composed of one or more slots and actions. A slot is a business process step (step) and intent (intent) is described. The action (Action) describes a business process called to perform a collaborative business action that is executed after all slots are filled. In this case, the action type may be described as an execution application, API, question, suggestion, message template, etc., but is not limited thereto.

Specifically, the details to be processed for each business processing step are described in the slot. The details to be processed for each business processing step are defined by REST API and registered as URLs in the slot. Manages REST API parameters and execution result sets as meta information and binds them at the time of execution.

The software robotics 120 executes an application corresponding to a Frame/Slot-based System to System (S2S) interaction performed by the interactor 110 .

For example, the application executed by the software robotics 120 performs a classification application 121 that performs data classification including task classification, a response application 122 that responds to a query including a task response, and a task recommendation. It may include a recommendation application 123 that performs data recommendation including, a processing application 124 that performs command processing including a business process call, and an analysis application 125 that performs data analysis including context analysis. , but is not limited thereto.

The interactor 110 requests the software robotics 120 to drive an application required by the frame/slot based S2S (System to System) interaction, and the software robotics 120 is the interactor 110 . The application is driven and executed, and the application execution result is reported to the interaction unit 110 .

Meanwhile, according to an additional aspect of the present invention, the interactor 110 may further include a context analyzer 112 . 3 , the context analyzer 112 may include a text analyzer 112a, a session analyzer 112b, and an intention analyzer 112c.

The text analysis unit 112a derives a collaboration participant associated with the message through text analysis of messages transmitted between a plurality of collaboration participant terminals. In this case, time information may be derived through the text analysis unit 112a.

The session analysis unit 112b derives a collaboration task unit associated with the message through session analysis of messages transmitted between a plurality of collaboration participant terminals, and manages the business continuity of the derived collaboration task unit.

3 is a diagram illustrating a context-based asynchronous session management concept. As shown in FIG. 3 , the session analysis unit 112b manages the collaborative work processing details in progress to determine whether to start/continue/end the session, and stores and manages the collaborative work processing details in the session memory for collaboration It can be implemented to manage business continuity.

The intent analysis unit 112c extracts the intent of the collaboration participant derived by the text analysis unit 112a through intent analysis on messages transmitted between a plurality of collaboration participant terminals.

A technique is known in which the analysis of intent from a message is scored based on morphological analysis and a decision tree, and an intended pattern is determined through this.

Meanwhile, the intention analysis unit 112c may be implemented to extract multiple intentions based on a rank. In this case, when the intention analysis unit 112c has multiple intentions of the collaboration participant, the intention analysis unit 112c may be implemented to determine the intention according to the priority.

4 is a diagram illustrating extraction of multiple intentions based on a rank. Referring to FIG. 4 , by analyzing the message content, all intents included in the content are derived, and priorities are determined for the derived intentions according to a predefined rule, The intention is determined according to a predefined rank.

On the other hand, according to an additional aspect of the invention, the software robotics 120 is the intention of the collaboration participant among the task processing steps included in the individual task slots constituting the task story of the collaborative task unit determined by the session analysis unit 112b. It may be implemented to call a business process for executing a business processing step corresponding to the intelligent collaboration system 200 .

At this time, the software robotics 120 defines CRUD (Create, Read, Update, Delete) according to the URI (Uniform Resource Identifier) and the basic method of http (POST, GET, PUT, DELETE), and objectifies when the method is called It can be implemented to call a REST (REpresentational State Transfer)-based business process that accesses the designated business processing step.

5 is a diagram illustrating a work story of a schedule-related collaboration task unit in which schedules are coordinated and registered among collaboration participants. The software robotics 120 collaborates among the task processing steps S1 (confirm the recipient's schedule), S2 (send possible schedule mail), and S3 (send confirmation mail) included in the schedule coordination individual task slot of the work story of the schedule-related collaborative task unit. A business process for executing a business processing step corresponding to the intention of a participant (a mail message recipient) is called from the intelligent collaboration system 200 .

On the other hand, according to an additional aspect of the invention, the software robotics 120 performs NER (Named Entity Recognition) of individual task slots constituting a work story for each collaboration task unit from messages transmitted between collaboration participant terminals. The intelligent collaboration system (intelligent collaboration system ( 200) can be implemented.

For example, if it is a meeting-related collaboration work unit in which a plurality of collaboration participants participate, the intentions required in Named Entity Recognition (NER) may be a place, a time, and a list of participants.

Meeting-related messages are transmitted between the collaboration participant terminals, and as the intentions of the collaboration participants are analyzed by the intention analysis unit 112c, the software robotics 120 performs NER (Named Entity Recognition) to create a work story of a meeting-related collaboration task unit. The extracted intentions are filled in individual task slots by extracting the place, time, and participant list, which are intentions necessary for the filling of individual task slots constituting the .

Through this process, when the intentions of place, time, and participant list required for individual work slots are all filled, the software robotics 120 executes conference schedule registration, which is a collaborative work action of a conference-related collaborative work unit. A business process is called from the intelligent collaboration system 200 to register a meeting schedule.

Meanwhile, according to an additional aspect of the invention, the interactor 110 may further include a learning unit 113 . The learning unit 113 analyzes the unstructured messages delivered from the external collaboration system 400 through learning, and converts them into standardized intentions usable in the internal intelligent collaboration system and learns them.

Since the messages delivered from the external collaboration system 400 are unstructured messages that are different from the standardized messages delivered from the internal intelligent collaboration system 200 in message format, etc., in order to be available in the intelligent collaboration support platform 100, Atypical intentions included in the atypical messages delivered from the collaboration system 400 should be analyzed and learned through the learning unit 113 to be converted into a standardized intention and accumulated.

Meanwhile, according to an additional aspect of the invention, the intelligent collaboration support platform 100 may further include a collaboration interworking adapter 130 . The collaboration interworking adapter 130 is a configuration for interworking the external collaboration system 400 .

When the external collaboration system 400 is linked through the collaboration interworking adapter 130 , external messages delivered from the external collaboration system 400 are analyzed, and intentions are obtained from external messages using the information learned by the learning unit 113 . is extracted

6 is a block diagram showing the configuration of an embodiment of the collaboration interworking adapter of the intelligent collaboration support platform of the advanced intelligent collaboration support system according to the present invention. As shown in FIG. 6 , the cooperative interworking adapter 130 includes a channel forming unit 131 , a command shell 132 , an SNS interworking unit 133 , and a web interworking unit 134 , It includes a mail interworking unit 135 , an internal system interworking unit 136 , and a cooperative interworking control unit 137 .

The channel forming unit 131 connects external collaboration systems and communication channels. In this case, the external collaboration system may be a social network service (SNS) server providing a social network service, a web server providing a web service, a mail server providing a mail service, and the like.

The command shell 132 executes commands of internal systems or external collaboration systems including the intelligent collaboration system, interprets requests from internal systems or external collaboration systems, converts them to APIs, and processes commands The result is converted and displayed in the form of a structured document that can be understood by the user. For example, the structured document may be a document such as Jason or XML.

The SNS linkage unit 133 exchanges real-time messages with a social network service (SNS) server that provides a social network service. For example, the social network service (SNS) server may be KakaoTalk, Slack, Facebook, WeChat, or the like.

The web interworking unit 134 exchanges real-time messages with web servers that provide web services. For example, the web interworking unit 134 may be implemented to exchange real-time messages with web servers that provide web services based on webhooks.

The mail link unit 135 exchanges real-time messages with mail servers that provide mail services. For example, the mail interworking unit 135 may be implemented to exchange real-time messages with mail servers that provide mail services based on push.

The internal system interworking unit 136 exchanges real-time messages with an internal system including an intelligent collaboration system.

The cooperative interworking control unit 137 sets up a communication channel to be connected through the channel forming unit 131, command processing through the command cell 132, the SNS interworking unit 133, the web interworking unit 134, and the mail interworking unit ( 135) and controls the overall cooperative interworking, including message exchange control through the internal system interworking unit 136.

In addition to this, the collaboration interworking control unit 137 may perform message history management, user management, statistics management, and the like. By implementing in this way, the present invention is multi-linked with a plurality of internal or external collaboration systems to enable advanced intelligent collaboration support and management.

Meanwhile, according to an additional aspect of the invention, the intelligent collaboration support platform 100 may further include an NLP platform adapter 140 . The NLP platform adapter 140 is a configuration for interworking the NLP platform 500 .

The NLP platform 500 that performs natural language processing (NLP) including morpheme analysis, entity name analysis, dialogue act analysis, and intention analysis required for message analysis is linked through the NLP platform adapter 140 .

In this case, the NLP platform adapter 140 may be implemented to perform natural language processing (NLP) in which multiple languages such as Chinese and English are supported as well as Korean.

Meanwhile, according to an additional aspect of the invention, the intelligent collaboration support platform 100 may further include a control unit 150 . The control unit 150 controls the overall intelligent collaboration support platform 100 including intention analysis control.

When analyzing a message through the intelligent collaboration support platform 100, the present invention first determines session continuity based on priority, and performs intent analysis and NER (Named Entity Recognition) extraction. Through this, it is decided whether or not to execute the collaborative work action.

At this time, since a new session may be generated or an existing session may be terminated during message exchange, the session is managed asynchronously for continuity correction. On the other hand, one message may relate to multiple sessions.

As described above, the present invention intelligently integrates and manages various scattered and sporadic messages input from various devices, analyzes them according to the intention, and integrates and supports various internal and external collaboration services, but multiple internal or external User convenience can be improved by interworking with external collaboration systems.

In addition, since the present invention can execute a business process by intelligently determining a collaboration task through message intent analysis, user selection for collaboration can be minimized.

In addition, the present invention does not stop at filling the individual work slots constituting the work story for each collaborative work unit, but when they are filled, more reliable collaboration support is possible by executing a collaborative work action.

In addition, since the present invention can be linked with a natural language processing platform capable of supporting multiple languages, advanced intelligent collaboration support with excellent scalability is possible.

The various embodiments disclosed in the present specification and drawings are merely presented as specific examples to aid understanding, and are not intended to limit the scope of various embodiments of the present invention.

Accordingly, the scope of various embodiments of the present invention, in addition to the embodiments described herein, all changes or modifications derived based on the technical idea of various embodiments of the present invention are included in the scope of various embodiments of the present invention. should be interpreted as being

The present invention can be industrially used in the field of collaboration-related technology and its applied technology.

100: Intelligent Collaboration Support Platform
110: interaction unit
111: communication department
112: context analysis unit
112a: text analysis unit
112b: session analysis unit
112c: intention analysis unit
113: study department
120: Software Robotics
121 Classification application
122: response application
123: recommended application
124: processing application
125: analysis application
130: collaboration interworking adapter
131: channel forming part
132: command cell
133: SNS linkage
134: web linkage
135: mail linkage
136: internal system linkage
137: Collaboration interlocking control unit
140 : NLP Platform Adapter
150: control unit
200: intelligent collaboration system
210: payment collaboration work process
220: mail collaboration work process
230: schedule collaboration work process
240: meeting collaboration work process
250: IPA collaborative work process
260: FIDO Security Process
270 : logger process
300: Big Data Analysis Platform
400: external collaboration system
500: NLP platform

Claims (10)

Asynchronous session management for collaborative tasks and interaction with collaborative tasks corresponding to message intentions through context-based message intent analysis of messages delivered between multiple collaboration participant terminals ) and an intelligent collaboration support platform that performs;
A collaboration scenario is defined by defining a work story for each collaborative work unit, defining individual work slots constituting a work story for each defined collaborative work unit, and defining work processing steps for individual work slots. ), but according to the collaboration task call from the intelligent collaboration support platform, the task processing steps included in the individual task slots constituting the task story for each collaborative task unit are executed. intelligent collaboration system;
including,
The intelligent collaboration support platform has:
a channel forming unit for connecting communication channels with external collaboration systems;
Executes commands of internal or external collaboration systems including intelligent collaboration systems, interprets requests from internal systems or external collaboration systems, and converts them to API Converting and displaying a command cell (Command Shell) and;
an SNS linkage unit for exchanging real-time messages with a social network service (SNS) server that provides a social network service;
a web interworking unit for exchanging real-time messages with web servers that provide web services;
a mail link unit for exchanging real-time messages with mail servers providing mail services;
an internal system interworking unit for exchanging real-time messages with an internal system including an intelligent collaboration system;
Collaboration interworking control unit that controls overall collaboration interworking including communication channel setting to be connected through channel forming unit, command processing through command cell, SNS interworking unit, web interworking unit, and message exchange control through mail interworking unit and internal system interworking unit cast;
Advanced and intelligent collaboration support system including the included collaboration interworking adapter. The method of claim 1,
The intelligent collaboration support platform has:
an interaction unit including a communication unit for performing frame/slot-based S2S (System to System) communication with an internal system including an intelligent collaboration system or external collaboration systems;
Software Robotics that executes an application corresponding to a Frame/Slot-based S2S (System to System) interaction performed by the interactor;
Advanced intelligent collaboration support system that includes more. The method of claim 1,
The intelligent collaboration support platform has:
an NLP platform adapter interworking with an NLP platform that performs natural language processing (NLP) including morpheme analysis, entity name analysis, dialogue act analysis, and intention analysis required for message analysis;
Advanced intelligent collaboration support system that includes more. 3. The method of claim 2,
Interactions:
a text analysis unit for deriving a collaboration participant associated with the message through text analysis of messages transmitted between a plurality of collaboration participant terminals;
a session analysis unit for deriving a collaborative work unit associated with the message through session analysis of messages transmitted between a plurality of collaboration participant terminals and managing the continuity of the derived collaborative work unit;
an intention analysis unit for extracting the intent of the collaboration participant derived by the text analysis unit through intent analysis on messages transmitted between a plurality of collaboration participant terminals;
Advanced intelligent collaboration support system that further includes a context analysis unit that includes. 5. The method of claim 4,
Software Robotics:
Advancement of calling from the intelligent collaboration system the business process that executes the task processing step corresponding to the intention of the collaboration participant among the task processing steps included in the individual task slots constituting the task story of the collaborative task unit determined by the session analysis unit intelligent collaboration support system. 5. The method of claim 4,
Software Robotics:
Defines CRUD (Create, Read, Update, Delete) according to URI (Uniform Resource Identifier) and http basic methods (POST, GET, PUT, DELETE) Advanced intelligent collaboration support system that calls REpresentational State Transfer)-based business processes. 5. The method of claim 4,
The intent analysis unit:
Advanced intelligent collaboration support system that extracts multiple intentions based on rank. 8. The method of claim 7,
The intent analysis unit:
Advanced intelligent collaboration support system that determines the intention according to priority when the intention of the collaboration participant has multiple intentions. 5. The method of claim 4,
Software Robotics:
NER (Named Entity Recognition) is performed to extract the intentions necessary for filling of individual task slots constituting the task story for each collaboration task unit from messages transmitted between the terminals of the collaboration participants to create individual task slots An advanced intelligent collaboration support system that fills the intentions extracted from the fields and calls a business process that executes a collaborative task action from the intelligent collaboration system when individual task slots are filled. 3. The method of claim 2,
Interactions:
a learning unit that analyzes atypical messages delivered from an external collaboration system through learning, converts them into standardized intentions usable in an internal intelligent collaboration system, and learns;
Advanced intelligent collaboration support system that includes more.

Images