KR102461482B1 - The apparatus of harmful contents blocking service platform based on ai agent - Google Patents

Abstract

The present invention relates to an artificial intelligence agent-based harmful content blocking service platform device, which includes: an agent which checks the harmfulness of inspection target content of the management target site and reports the harmfulness confirmation result; and an agent management server which manages the operation of the agent by determining the inspection target of the agent, time, and the response method after inspection, and notifies the site manager of the discovery of harmful content when the agent reports the discovery of harmful content.

Description

AI agent-based harmful content blocking service platform device {THE APPARATUS OF HARMFUL CONTENTS BLOCKING SERVICE PLATFORM BASED ON AI AGENT}

The present invention relates to a harmful content blocking service technology, and more particularly, an artificial intelligence agent-based harmful content that can prevent the distribution of harmful content in advance by inspecting and blocking harmful content on the Internet in real time based on an artificial intelligence agent. It relates to a content blocking service platform device.

The spread of mobile devices and content platforms is spreading, and various types of harmful content are being distributed without being restrained. Currently, many efforts are being made to block harmful content.

For example, content uploaded to servers running SNS and e-commerce such as YouTube, Facebook, Yahoo, etc. are found harmful using various technologies, and measures such as blocking and deletion are taken. However, it is difficult to respond in real time, and if the inspection method is bypassed or if a full inspection is performed by an administrator, the registration process is complicated or the review period is very long, so it is impossible to screen for and block harmful content in real time.

The existing total inspection method is expensive because labor costs are reflected. Some large companies are applying artificial intelligence (AI) to their own harmful content blocking services, but small and medium-sized enterprises are too expensive to introduce it. For this reason, the majority of cases involve post-processing after receiving a report of damage rather than a preemptive response.

In order to keep up with the diversified media along with the development of technology, blocking based on the existing URL (Uniform Resource Identifier) list or network address and the total inspection method by the administrator are time-consuming and costly, and the legal burden of post-processing is also high. is occurring In addition, an increasing number of countries and local governments are strengthening regulations around the world.

Therefore, it is necessary to develop a technology that can respond from the upload stage or the initial stage of distribution, and furthermore, a technology that can respond more actively by automatically inspecting and blocking and warning in advance if the harmfulness standard is not exceeded.

Korean Patent No. 10-2158747 (2020.09.16) Korean Patent No. 10-2259730 (2021.05.27)

An embodiment of the present invention is to provide an artificial intelligence agent-based harmful content blocking service platform device that can prevent the distribution of harmful content in advance by inspecting and blocking harmful content on the Internet in real time based on an artificial intelligence agent. .

An embodiment of the present invention is based on an artificial intelligence agent that can reduce maintenance and management costs and simplify inspection procedures by operating and managing harmful content inspection in the form of an agent to distribute harmful content determination to local terminals. We want to provide a platform device for blocking harmful content.

An embodiment of the present invention is an artificial intelligence agent-based parental content blocking service platform that can quickly respond to harmful contents by configuring agents as an independent object or group and sharing the results of harmfulness inspection and blocking between agents. We want to provide a device.

Among the embodiments, the AI agent-based harmful content blocking service platform device includes an agent that checks the harmfulness of the content to be inspected on the managed site and reports the harmfulness confirmation result, and the inspection target and time of the agent, and a response method after inspection and an agent management server for managing the operation of the agent by determining an activity range including

The agent is installed in the browser in the form of an add-on, checks the harmfulness of the content loaded on the user terminal, and matches through the agent management server as an independent object with general agents that report the site of the confirmed harmfulness to the agent management server It may consist of independent agents that self-execute a content harmfulness test on a managed target and notify the agent management server of the test result to receive a reward according to the test execution.

In the case of an agent subscribing to the compensation system of the agent management server among the general agents, the agent receives a checklist from the agent management server, performs content inspection of websites in the checklist, and manages the agent management results. By notifying the server, you can receive compensation according to the inspection agent.

The agent can preemptively block the harmful content by grouping them according to specific criteria and sharing the harmful content inspection and blocking results of other agents through communication with other agents within the group or between groups.

The agent is installed on the server of the managed site or connected to the internal network according to the installation location, and has an internal agent group having the right to inspect files and folders inside the server of the managed site; It can be divided into an external agent group that is connected to the server of the target site through an external network and is provided as a lease for a certain period of time and has the authority to periodically inspect new and updated content on the server of the managed site. .

The agent is operated in a vertical type that communicates between the agent management server and sub-agents through a leader agent among agents in the group, or a horizontal type in which each of the agents in the group independently communicates with the agent management server It can be operated as a vertical or a mixed type of vertical and horizontal types.

The agent checks the harmfulness of the file to be uploaded when uploading the file, blocks the upload of the file when it is found to be harmful, and when it is confirmed that it is safe, the file can be uploaded including metadata indicating safety authentication based on the management level.

The agent management server may request the server of the management target site to stop service and isolate the content found to be harmful according to the management level.

The agent management server provides a management target list by matching the management target according to the management list request of the independent agent, checks the management list level according to the harmfulness discovery report of the independent agent, and notifies the management target manager of the discovery of harmful contents And if the checked management level is high, the service of contents with harmful contents can be stopped and quarantined.

The agent management server may perform point compensation for an agent that performs a harmfulness check on a management target site in the management target list.

The disclosed technology may have the following effects. However, since a specific embodiment does not mean that it should include all of the following effects or only the following effects, it should not be understood that the scope of the disclosed technology is limited thereby.

The artificial intelligence agent-based harmful content blocking service platform apparatus according to an embodiment of the present invention can prevent the distribution of harmful content in advance by detecting and blocking harmful content on the Internet in real time based on the artificial intelligence agent.

The artificial intelligence agent-based harmful content blocking service platform device according to an embodiment of the present invention operates and manages harmful content inspection in the form of an agent, so that the harmful content determination can be distributed to local terminals, thereby reducing maintenance and management costs and the inspection procedure can be simplified.

The artificial intelligence agent-based harmful content blocking service platform device according to an embodiment of the present invention configures agents as an independent object or group, and shares the results of harmfulness inspection and blocking between agents, so that harmful content can be quickly dealt with. can respond.

1 is a block diagram schematically illustrating an artificial intelligence agent-based harmful content blocking service platform apparatus according to an embodiment of the present invention.
2 is a block diagram showing the basic configuration of an artificial intelligence agent-based harmful content blocking service platform device according to an embodiment of the present invention.
FIG. 3 is a diagram illustrating a connection configuration between the agent in FIG. 1 and an agent management server.
4 is a block diagram illustrating an artificial intelligence agent-based harmful content blocking service platform apparatus according to an embodiment of the present invention.
5 and 6 are flowcharts illustrating an operation process of an artificial intelligence agent-based harmful content blocking service platform device according to the present invention.

Since the description of the present invention is merely an embodiment for structural or functional description, the scope of the present invention should not be construed as being limited by the embodiment described in the text. That is, since the embodiment is capable of various changes and may have various forms, it should be understood that the scope of the present invention includes equivalents capable of realizing the technical idea. In addition, since the object or effect presented in the present invention does not mean that a specific embodiment should include all of them or only such effects, it should not be understood that the scope of the present invention is limited thereby.

On the other hand, the meaning of the terms described in the present application should be understood as follows.

Terms such as “first” and “second” are for distinguishing one component from another, and the scope of rights should not be limited by these terms. For example, a first component may be termed a second component, and similarly, a second component may also be termed a first component.

When a component is referred to as being “connected” to another component, it may be directly connected to the other component, but it should be understood that other components may exist in between. On the other hand, when it is mentioned that a certain element is "directly connected" to another element, it should be understood that the other element does not exist in the middle. On the other hand, other expressions describing the relationship between elements, that is, "between" and "between" or "neighboring to" and "directly adjacent to", etc., should be interpreted similarly.

The singular expression is to be understood to include the plural expression unless the context clearly dictates otherwise, and terms such as "comprises" or "have" refer to the embodied feature, number, step, action, component, part or these It is intended to indicate that a combination exists, and it should be understood that it does not preclude the possibility of the existence or addition of one or more other features or numbers, steps, operations, components, parts, or combinations thereof.

Identifiers (eg, a, b, c, etc.) in each step are used for convenience of description, and the identification code does not describe the order of each step, and each step clearly indicates a specific order in context. Unless otherwise specified, it may occur in a different order from the specified order. That is, each step may occur in the same order as specified, may be performed substantially simultaneously, or may be performed in the reverse order.

The present invention can be embodied as computer-readable codes on a computer-readable recording medium, and the computer-readable recording medium includes all types of recording devices in which data readable by a computer system is stored. . Examples of the computer-readable recording medium include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like. In addition, the computer-readable recording medium may be distributed in a network-connected computer system, and the computer-readable code may be stored and executed in a distributed manner.

All terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs, unless otherwise defined. Terms defined in the dictionary should be interpreted as being consistent with the meaning of the context of the related art, and cannot be interpreted as having an ideal or excessively formal meaning unless explicitly defined in the present application.

1 is a block diagram schematically illustrating an artificial intelligence agent-based harmful content blocking service platform apparatus according to an embodiment of the present invention.

Referring to FIG. 1 , the AI agent-based parental content blocking service platform apparatus 100 includes a user terminal 110 , a web server 130 , an agent 150 , and an agent management server 170 .

First, the user terminal 110 is a terminal device used by a user, and may be, for example, any computing device including a mobile device such as a smart phone or a tablet PC as well as a PC and a laptop computer. A user may subscribe to various contents such as new articles, images, or videos uploaded to an online bulletin board through the user terminal 110 . In this case, the user terminal 110 may correspond to a subscriber terminal. A user may upload content such as a new article, image, or video to an online bulletin board through the user terminal 110 . In this case, the user terminal 110 may correspond to a publisher terminal. Here, the user terminal 110 may be connected to the web server 130 and the agent management server 170 through a network, and may be connected to a plurality of user terminals 110 at the same time.

The web server 130 corresponds to a web service provider that provides content subscribed to by the user terminal 110 . The web server 130 operates a database in which various contents are registered and updated.

The agent 150 corresponds to a program that determines the harmfulness of the content to be used by the user terminal 110 . The agent 150 is distributed and operated by the agent management server 170 . In this case, the agent 150 may be distributed and operated by being divided into corporate and personal use. The agent 150 is installed in the user terminal 110 to provide a service that blocks harmful content for content intended to be used in the user terminal 110 . The agent 150 may be separately implemented as a device for inspecting harmful content.

The agent management server 170 may provide the agent 150 so that the harmfulness check of the content is performed in the user terminal 110 . The agent management server 170 may be connected to the agent 150 installed in the user terminal 110 through a wired network or a wireless network such as Bluetooth or WiFi, and the agent 150 installed in the user terminal 110 through a wired or wireless network. ) can communicate with The agent management server 170 may retain various pieces of information necessary to inspect and block harmfulness of content. For example, the agent management server 170 may maintain a list of harmful URLs and a plurality of harmfulness judgment model files. The agent 150 may interwork with the agent management server 170 to determine in real time whether the subscription content of the user terminal 110 is harmful through a harmful URL list and a plurality of harmfulness determination model files. The agent 150 may deliver the content inspection and corresponding actions to the web server 130 and the agent management server 170 . The agent 150 inspects the harmfulness of the content uploaded to the web server 130 and reports the inspection result to the web server administrator. In addition, when a network and server to be monitored, a specific folder, etc. are designated, the agent 150 periodically identifies a usage trend to monitor and manage it, and reports the result to an automatically set administrator. The agent 150 can share the inspection results with other agents, so that as the number of agents increases, monitoring and inspection can be operated in a 24/7 system like a net. Accordingly, even if harmful content is distributed, damage can be minimized and reproduction and redistribution of harmful content can be minimized.

2 is a block diagram showing the basic configuration of an artificial intelligence agent-based harmful content blocking service platform device according to an embodiment of the present invention.

Referring to FIG. 2 , content 210 such as the Internet, broadcasting, and cable TV is input to the intelligent agent 230 , and the intelligent agent 230 receives an authentication key from the agent operation and management center 220 and is input The harmfulness of the content is inspected and the degree of harmfulness, blocking, reporting, and notification are implemented in real time according to the policy set by the agent operation and management center 220 . As a result of the real-time harmfulness test, only contents determined to be non-harmful are output through the plurality of terminals 240 . The intelligent agent 230 is installed and operated in each of a plurality of terminals 240 composed of a computer or various devices.

FIG. 3 is a diagram illustrating a connection configuration between the agent in FIG. 1 and an agent management server.

Referring to FIG. 3 , the agent 150 may include general agents 151 and independent agents 153 . The general agents 151 are user agents installed in the browser as an add-on. Each of the general agents 151 visits a browser, inspects and blocks the harmfulness of the content loaded on the individual's local PC, and reports the site for which the harmfulness is confirmed to the agent management server 170 . Among the general agents 151, in the case of an agent subscribing to the compensation system, a checklist may be provided from the agent management server 170 to inspect the contents of a website in the checklist, and the test result may be transmitted to the agent management server ( 170) to receive compensation in various forms, such as points, according to the inspection agency.

Independent agents 153 are independent objects capable of executing functions such as content harmfulness inspection, blocking and isolation by themselves. You can progress and be rewarded accordingly. The independent agents 153 may be managed through the independent control center 330 .

The agent 150 may also be divided into an internal type and an external type. The internal type may be in the same local network as the server to be managed, directly connected to the server to be managed, or installed, and may have the authority to inspect all files inside the server. Among the independent agents 153 , in the case of an agent installed inside a server, it is possible to check for harmful substances on files other than the contents being serviced by the corresponding server.

External type is a case in which it is located on a network different from the server of the target to be managed, or is physically installed in a device (PC, server) different from the target to be managed. This is a server or PC that is mainly operated for the purpose of inspection, and is separated from the server where the web service is running.

On the other hand, the agent 150 is connected to the central control center 310 corresponding to the agent management server 170 and RTMS (Real Time Messaging Service) to create a specific communication channel to communicate. The independent agents 153 may be communicatively connected to the central control center 310 via the independent control center 330 or through a reader agent.

4 is a block diagram illustrating an artificial intelligence agent-based harmful content blocking service platform apparatus according to an embodiment of the present invention.

Referring to FIG. 4 , the AI agent-based parental content blocking service platform apparatus 100 may include an agent group unit 410 , an agent operation management unit 430 , and a web service management unit 450 .

The agent group unit 410 may configure at least one or more agents 150 into groups according to specific criteria. Of course, the agent 150 may act as an independent object. In one embodiment, the agent group unit 410 may be composed of an internal agent group 410a and an external agent group 310b according to the installation location of the agent. The internal agent group 410a is a bundle of at least one agent installed or connected to a management target such as the user terminal 110 and the web server 130, or having the authority to perform content inspection in an internal network such as an intranet. is composed of The external agent group 410b is composed of a bundle of inspection agents that perform inspection on content provided to a management target in an external network. The internal agent group 410a enables stronger management than the external agent group 410b. For example, the internal agent group 410a may manage folders and files in a specific local directory other than the actual web-serviced content. The external agent group 410b may be provided as a rental type for a certain period of time for a company.

The agent group unit 410 may further include an independent agent group 410c in terms of agent operation and management. The independent agent group 410c may be used when there are many management targets, when external management control cannot be received, and when it is necessary to introduce a separate independent management center inside.

The agent group unit 410 may be operated in at least one of a vertical type, a horizontal type, and a mixed type among the agents in the group when the agents are grouped into a group. In the vertical type, there is a leader among agents, and through this leader, agents can communicate and exchange information about progress and roles. The leader agent 411 may receive a command from the agent operation management unit 430 , deliver it to the sub-agent 413 , and propagate the command to the sub-agent 413 . The leader agent 411 may also receive the task performance and results from the sub-agents 413 and report it to the agent operation management unit 430 . In the horizontal type, each agent operates independently, and directly communicates with the agent operation management unit 430 to receive and report work. In addition, information is exchanged in real time with each other as needed. The agent group unit 410 may share the results of mutual inspection and blocking between agents within a group or between groups.

Each of the agents included in the agent group unit 410 may inspect the content in real time and block it according to the degree of harmfulness. Here, each of the agents is “a device and method for managing harmful content using metadata” of Registered Patent No. 10-2213373 previously applied by the present applicant and “Artificial intelligence-based harmful content blocking device” of Registered Patent No. 10-2259730 The harmful content inspection and blocking technology disclosed in the like may be applied.

The agent operation management unit 430 may distribute the agent 150 and manage the operation of the distributed agents 150 . In an embodiment, the agent operation management unit 430 may manage agents for each agent group. In the case of the internal agent group 410a, the agent operation management unit 430 may manage agents distributed free of charge to all registered or unregistered individuals. For example, it is possible to manage the agent version, the harmfulness determination AI model being used, customer management, authentication key, and the like. In the case of the external agent group 410b, the agent operation management unit 430 manages the leased agents to be used for a specific period and special purpose in the same way as the management of the internal agent group 410a, but determines the harmfulness according to the purpose AI model type , it is possible to provide options such as a processing method and response method, and an authentication key when a response occurs. The agent operation management unit 430 may be connected to each agent 150 through a dedicated communication channel (RTMS, Real Time Messaging Service). The independent agent group 410c may be independently managed through the independent agent operation management unit 435 . The independent agent operation management unit 435 may operate in the same manner as the agent operation management unit 430 . The independent agent operation management unit 435 may manage and supervise the independent agent group 410c and manage the web server 130 such as websites and other contents to be managed by the company. In this case, the independent agent group 410c is authenticated for use by the agent operation and management unit 430, but the control over the scope of activities such as the inspection target, inspection time, and post-inspection response method is managed by the independent agent operation and management unit 335. receive

The agent operation management unit 430 may prepare a report on the contents of the agent's content inspection and blocking results and provide it to the web server management unit 450 . The agent operation management unit 430 enables the agents grouped in the agent group unit 410 to communicate with each other through a specific channel and to provide content inspection and actions therefor. For example, the agent operation management unit 430 divides all the contents in a specific local directory among the grouped agents for inspection or shares information according to the inspection result to prevent duplicate inspection. The agent operation and management unit 430 checks the harmfulness judgment AI model and then updates it, commands to adjust content inspection items, ungroups grouped agents and concludes a group, dissemination of site and harmful content information to be blocked within the group, conducts full inspection and After re-inspection of only the contents within the specified range, update the safety certification, list the contents to be inspected when performing inspection on the leader agent, distribute it to sub-agents, and collect inspection instructions and results.

In one embodiment, the agent operation and management unit 430 may match the management object to the agent 150 according to the management list request of the agent 150 and provide the management object list to the agent 150 . For example, when the user of the installed local PC is absent, the agent 150 automatically requests the management list from the agent operation and management unit 430 to automatically inspect the site to be managed in the management list for harmfulness and compensate accordingly. can receive The agent operation management unit 430 may receive a report of the agent's harmfulness test result and reward points. The purchaser or installer of the agent may use the points paid by the agent operation and management unit 430 in an online shopping mall or the like.

The web service management unit 450 can receive the agent's content inspection report from the agent operation management unit 430, and can alert and alarm the corresponding content publisher according to the web service policy, and block the bulletin board, blog, etc. where the content is posted. , release, switch to private, etc.

5 and 6 are flowcharts illustrating an operation process of an artificial intelligence agent-based harmful content blocking service platform apparatus according to the present invention.

Referring to FIG. 5 , a harmful content blocking service process according to a content request from the user terminal 110 is illustrated.

First, a user may access a website through the user terminal 110 and request content from the corresponding web server 130 . In an embodiment, the user terminal 110 may request desired content through a web browser, an app, an IoT device, or the like. Here, the requestable content may be variously included, such as a website, an image, a video, and a smartphone text message. In the user terminal 110 , the agent program 150 for inspection may be installed through the agent management server 170 .

The agent 150 may preferentially check the presence or absence of safety authentication before the content is loaded into the user terminal 110 (step S510). In one embodiment, the agent 150 may check the presence or absence of safety authentication and, if there is no authentication content, notify the agent management server 170 of this, and if the authentication content is invalid or damaged, it may be updated. Here, the safety certification may correspond to the metadata of the content, and the metadata may include content-related information and a rating based on the degree of harmfulness, and the publisher's email address, ID, IP address, posting date, harmfulness test date, It may further include information including at least one of a registered website address and whether or not copyright is protected.

The agent 150 may determine whether to register a harmful URL by preferentially performing a URL check on the content requested by the user (step S520). In one embodiment, the agent 150 may interwork with the agent management server 170 to perform a URL check in real time by collating it with a site checklist held in the agent management server 170 . Here, in the site checklist, not only a list of public harmful URLs but also a list of personal harmful URLs registered based on a user's personal preference are stored together, so that the agent 150 can perform a URL inspection that satisfies the user's criteria. As a result of the URL inspection, the agent 150 may display the result on the user terminal 110 if it is registered as a harmful URL, and may load the corresponding content if it is not registered as a harmful URL.

The agent 150 may proceed with whether or not to determine whether the content is harmful in real time before being output to the screen of the user terminal 110 at the same time as loading the content (step S530). In an embodiment, the agent 150 may perform a harmful content inspection in real time based on the harmfulness determination AI models. As a result of the harmfulness check, when it is determined that the harmful content is harmful, the agent 150 may request the agent management server 170 to register the harmful URL and notify the web server 130 of the detection of the harmful content.

The agent 150 may directly request the web server 150 to stop the service and inspect the content containing the harmful substances according to the degree of harmfulness (step S540).

Referring to FIG. 6 , a harmful content blocking service process according to a content upload request of the user terminal 110 is shown.

First, a user may access a website through the user terminal 110 and request content upload to the corresponding web server 130 . Here, the upload type is a general file upload, in which a file is attached to a blog, a bulletin board, an e-mail, or the like. Adding pictures in the blog editor is also considered an upload. In the user terminal 110 , the agent program 150 for inspection may be installed through the agent management server 170 .

The agent 150 may check the member status by inquiring the member management list held in the agent management server 170 before uploading the contents from the user terminal 110 (step S610). In one embodiment, if the agent 150 is a non-member, it may notify the agent management server 170 of this and block the upload according to the instructions of the agent management server 170 . If the agent 150 is a member, the agent 150 may check the management level.

The agent 150 checks the presence or absence of safety authentication and, if there is no authentication content, informs the agent management server 170 of this so that safety authentication-related measures can be taken. It can be done (step S620). When there is no safety authentication app in the user terminal 110, the agent 150 may block the upload according to the agent user's settings without safety authentication-related measures.

The agent 150 may select the content to be uploaded after confirming the safety authentication (step S630), and may proceed to determine whether the selected content is harmful before uploading it (step S640). In an embodiment, the agent 150 may perform a harmful content inspection in real time based on the harmfulness determination AI models. The agent 150 may block upload of harmful content when it is determined as harmful content as a result of the harmfulness test. The agent 150 may upload to the web server 130 including metadata corresponding to the member management level when it is confirmed that the content is safe.

As described above, it is possible to inspect the content through the agent and request a report and action from the corresponding content manager, and there is an effect of stopping the service of harmful content or blocking uploading according to the agent manager setting.

Information on content that has been identified as harmful is stored in the agent and shared between agents, so that pre-existing blocking can be enforced when reconnecting to a site with the corresponding content, and duplicate inspection can be avoided.

It is possible to reduce the burden of costs for harmful content inspection for individuals and businesses.

Although the above has been described with reference to preferred embodiments of the present invention, those skilled in the art can variously modify and change the present invention within the scope without departing from the spirit and scope of the present invention as set forth in the claims below. You will understand that it can be done.

100: AI agent-based parental content blocking service platform device
110: user terminal 130: web server
150: agent
151: generic agents 153: standalone agents
170: agent management server
210: content 220: agent operation, management center
230: intelligent agent 240: terminal
310: central control center 330: stand-alone control center
410: agent group unit
410a: internal agent group 410b: external agent group
410c: Standalone Agent Group
411: leader agent 413: subagent
430: agent operation and management unit 435: stand-alone agent operation and management unit
450: web service management unit

Claims (10)

an agent that checks the harmfulness of the content to be inspected on the managed site and reports the harmfulness check result; and
An agent management server that manages the operation of the agent by determining an activity range including the target and time of the agent's inspection, and a response method after inspection, and notifies a site administrator of the discovery of harmful content when the agent reports the discovery of harmful content;
the agent
By grouping according to specific criteria and sharing the results of inspection and blocking of harmful contents of other agents through communication with other agents within or between groups, preemptively blocking the harmful contents and preventing duplicate scanning;
When the specific criterion is the installation location, an internal agent group installed on the server of the managed site or connected to the internal network and having the authority to inspect files and folders inside the server of the managed site; It is connected to the server of the management site through an external network and is provided as a lease for a certain period of time and is divided into an external agent group having the authority to periodically inspect new and updated content on the server of the management site. ,
It operates in a vertical type that communicates between the agent management server and sub-agents through a leader agent among agents in the group, or operates in a horizontal type that communicates with the agent management server independently of each of the agents in the group. An artificial intelligence agent-based harmful content blocking service platform device, characterized in that it operates in a mixed form of vertical and horizontal.
The method of claim 1, wherein the agent
General agents that are installed in the browser as an add-on and check the harmfulness of the content loaded on the user terminal and report the site for which the harmfulness is confirmed to the agent management server;
As an independent object, it is composed of independent agents that self-execute the content harmfulness test on the matched management target through the agent management server and notify the test result to the agent management server to receive compensation according to the test execution. An artificial intelligence agent-based parental content blocking service platform device.
3. The method of claim 2, wherein the agent
Among the general agents, in the case of an agent subscribing to the compensation system of the agent management server, the agent management server receives the checklist, performs the inspection of the contents of the website in the checklist, and reports the test result to the agent management server An artificial intelligence agent-based harmful content blocking service platform device, characterized in that it receives compensation according to the inspection agency.
delete delete delete The method of claim 1, wherein the agent
Artificial intelligence, characterized in that when a file is uploaded, the file to be uploaded is checked for harmfulness, and if harmfulness is found, the file upload is blocked, and when it is confirmed that it is safe, the file is uploaded including metadata indicating safety certification based on the management level. Agent-based parental content blocking service platform device.
The method of claim 1, wherein the agent management server
An artificial intelligence agent-based harmful content blocking service platform device, characterized in that it requests the server of the managed site to stop service and isolate the content found to be harmful according to the management level.
The method of claim 2, wherein the agent management server
According to the management list request of the independent agent, the management object is matched and the management object list is provided, the management list level is checked according to the harmfulness discovery report of the independent agent, and the discovery of harmful contents is notified to the management object manager and the confirmed management level If this is high, an artificial intelligence agent-based harmful content blocking service platform device, characterized in that it stops and isolates the service of harmful content.
10. The method of claim 9, wherein the agent management server
An artificial intelligence agent-based harmful content blocking service platform device, characterized in that points are rewarded for an agent that performs a harmfulness check on a management target site in the management target list.

Images