Evidence-Validated RAG Agent-Based System for Free-Drawing Emotional Observation and Story Development

Abstract

The present invention relates to a free drawing emotion observation and story development system that supports emotion observation and story development regarding a user's free drawing, comprising: a multimodal data input unit that collects image data of the free drawing and production process data generated during the production process of the free drawing; a free drawing feature extraction unit that calculates feature values ​​regarding one or more of line, color, shape, spatial arrangement, margin, repetition, relationships between shapes, and production sequence from the image data and production process data; a context information structuring unit that generates context information by linking one or more of the user's age, developmental level, production materials, activity purpose, production context, previous works, and user self-description with the feature values; a RAG knowledge search unit that searches for supporting data corresponding to the feature values ​​and context information from a knowledge base; and an emotion observation agent that generates objective observation facts based on the feature values ​​and generates a plurality of emotion observation hypotheses based on the objective observation facts, context information, and searched supporting data.

Description

Technology Field

[0001] The present invention relates to a technology for observing emotions and developing stories using free drawing, and more specifically, to a system that collects not only the completed image of a free drawing created by a user but also production process data such as stroke coordinates, pen pressure, speed, direction, stopping, repeating, erasing, and modification generated during the production process of the free drawing, and extracts visual characteristics and process characteristics of the free drawing from the image data and production process data to observe the user's emotional expression. Background Technology

[0003] Free drawing is a creative activity that allows users to express their thoughts or experiences using free lines, colors, shapes, spatial arrangements, and symbols, without relying on standardized answers or verbal explanations. In particular, for users whose verbal expression skills are not sufficiently developed or who find it difficult to directly explain their thoughts or feelings in words, free drawing can be utilized as a useful resource for observing their thoughts, interests, and modes of expression.

[0004] Traditionally, in analyzing freehand drawings or paintings, the primary method used has been to analyze the objects, colors, shapes, sizes, or locations appearing in the finished image and to correlate these characteristics with specific meanings or emotional states.

[0005] However, even when identical or similar finished images are formed, the process by which users created those images may differ. For example, even for the same black area, the characteristics that appear during the production process may differ depending on whether it was colored once in a short time or if the same area was repeatedly drawn over over a long period of time.

[0006] In addition, actions such as deleting or modifying specific shapes while the user is creating a drawing, pausing the work for a certain period of time, or repeatedly expressing specific areas are difficult to verify solely from the completed image.

[0007] Therefore, it is necessary to analyze not only the finished image but also production process data such as stroke order, pen pressure, speed, pauses, repetitions, erasing, and modifications.

[0008] On the other hand, when a single visual feature is directly correlated with a single emotional state, there is a possibility that an interpretation different from the user's actual production intent may occur.

[0009] For example, a drawing that uses black extensively may be due to the expression of a specific emotion, but it may also be due to the user's color preference, the subject of the task, the materials used for production, or the depiction of a scene in the story.

[0010] Furthermore, even when a small figure is positioned at the edge of the screen, rather than assuming it represents a specific psychological state, it is necessary to consider various possibilities such as an observer within the story, perspective representation, differences in scene importance, or the user's intentional screen composition.

[0011] In particular, when analyzing freehand drawings using artificial intelligence, a problem may arise where it is difficult to verify the observations or supporting data upon which the results generated by the AI ​​are based. There is a possibility that the generated interpretation may be presented as a valid analysis result even if it has low relevance to the actual characteristics of the freehand drawing, is based on data targeting a different age group or environment than the user, or conflicts with the user's own description.

[0012] Furthermore, conventional automatic analysis systems have a problem in that they fail to sufficiently preserve the difference between the user's own explanation and the AI's interpretation by selecting or averaging one of conflicting pieces of information to output a single final result.

[0013] On the other hand, while free drawing can serve not only as a subject for emotional observation but also as a starting point for story creation, conventional story generation systems have the potential to excessively add elements not present in the original drawing during the process of generating new content based on objects detected in the drawing.

[0014] Therefore, technology is required to manage elements directly detected from freehand drawing, elements directly described by the user, and extension elements proposed by artificial intelligence separately, and to enable the user to select or modify a desired candidate among multiple story candidates. Prior art literature

[0015] Republic of Korea Published Patent 10-2022-0013468 The problem to be solved

[0016] The present invention was devised to solve the above-mentioned problems and aims to provide a free drawing sentiment observation and story development system capable of analyzing not only the completed image of the free drawing but also the production process data generated during the production process of the free drawing.

[0017] Another objective of the present invention is to provide a system that generates objective observation facts and multiple emotional observation hypotheses by distinguishing between them, without directly correlating individual features extracted from free drawing with specific emotional states.

[0018] Another objective of the present invention is to provide a system capable of analyzing identical features while considering the user-specific context by linking the user's age, developmental level, production materials, production context, previous works, and user self-description with feature values ​​of free drawing.

[0019] Another objective of the present invention is to provide a system capable of searching for evidence related to an emotion observation hypothesis using RAG knowledge search, and verifying whether the said evidence matches or conflicts with the characteristics of the free drawing, the production process, and the user's self-description.

[0020] Another objective of the present invention is to provide a system that, when an emotion observation hypothesis generated by AI conflicts with a user's self-explanation, preserves both along with their respective grounds without deleting either one, and generates additional verification questions if necessary.

[0021] Another objective of the present invention is to provide a system that generates multiple story candidates using freehand drawing and user self-description, and can manage each story component by distinguishing whether it was directly detected from the freehand drawing, described by the user, or suggested by AI.

[0022] Another objective of the present invention is to provide a system capable of distinguishing between AI analysis results and the final judgment of experts by having experts review, modify, or approve sentiment observation hypotheses and story candidates generated by AI.

[0023] The objectives of the present invention are not limited to those mentioned above, and other unmentioned objectives will be clearly understood by a person skilled in the art from the description below. means of solving the problem

[0025] To achieve the above objectives, the present invention relates to a free drawing emotion observation and story development system that supports emotion observation and story development regarding a user's free drawing, comprising: a multimodal data input unit that collects image data of the free drawing and production process data generated during the production process of the free drawing; a free drawing feature extraction unit that calculates feature values ​​regarding one or more of line, color, shape, spatial arrangement, margin, repetition, relationships between shapes, and production sequence from the image data and production process data; a context information structuring unit that generates context information by linking one or more of the user's age, developmental level, production materials, activity purpose, production context, previous works, and user self-description with the feature values; a RAG knowledge search unit that searches for supporting materials corresponding to the feature values ​​and context information from a knowledge base; an emotion observation agent that generates objective observation facts based on the feature values ​​and generates multiple emotion observation hypotheses based on the objective observation facts, context information, and searched supporting materials; a supporting material verification agent that verifies whether one or more of the multiple emotion observation hypotheses, the user self-description, and searched supporting materials match or conflict; and from the free drawing It includes a story structuring agent that generates multiple story candidates including one or more of a protagonist, background, goal, conflict, helper, emotional change, and ending based on one or more of the extracted characters, objects, spaces, colors, symbols, and the user self-description, and an expert review unit that enables an expert to review, modify, or approve one or more of the multiple emotional observation hypotheses and story candidates.

[0026] The multimodal data input unit described above can collect production process data including one or more of the coordinates of the stroke constituting the free drawing, start time, end time, pen pressure, tilt, color, thickness, speed, direction of progress, stop, repeat, erase, and modification, and one or more of the production video, audio, artwork description, self-report data, expert observation data, gaze information, facial expression information, and non-verbal response information of the free drawing.

[0027] In one embodiment of the present invention, two or more of the image data, production process data, voice and non-verbal response information are aligned along a common time axis, and an event division unit may be further included that divides one or more of the start or end of a stroke, color change, tool change, pen pressure change, pause for a certain period of time or longer, repeated overdrawing of the same area, erasing, modification, speech and non-verbal response into production events.

[0028] In one embodiment of the present invention, the individual difference normalization unit may further include one or more of the user's age, developmental level, motor ability, visual ability, communication method, production material, task conditions, and individual baseline calculated from previous works to normalize the feature value.

[0029] The knowledge base of the above RAG knowledge search unit is divided into two or more areas among the official education and development materials area, the academic and verification materials area, the institution-approved practical materials area, the individual portfolio area, and the creative extension materials area, and the above RAG knowledge search unit may select or restrict the areas subject to search based on one or more of the user's age, purpose of activity, purpose of analysis, consent status, access rights, and production context.

[0030] The above-described sentiment observation agent stores objective observation facts generated from the above-described feature values ​​and multiple sentiment observation hypotheses generated for the above-described objective observation facts separately from each other, and can generate two or more different sentiment observation hypotheses for the same objective observation fact.

[0031] In one embodiment of the present invention, a self-description guidance agent may further include generating questions regarding one or more of the names of shapes included in the free drawing, relationships between shapes, emotions of characters, causes of events, previous scenes, next scenes, important parts, and parts to be modified, and collecting the user's self-description in one or more ways among text, voice, picture cards, facial expression cards, color selection, gaze selection, shape indication by finger, gestures, and augmentative and alternative communication symbols.

[0032] The above-mentioned evidence verification agent calculates interpretation reliability for each of the above-mentioned multiple sentiment observation hypotheses based on two or more of the visual feature consistency, production process consistency, user self-description consistency, search basis suitability, consistency with previous works, and conflicting data ratio, and may assign one or more status information among additional verification required, existence of multiple possibilities, and judgment pending for sentiment observation hypotheses where the interpretation reliability does not satisfy preset conditions or where conflicting data exists.

[0033] If the above-mentioned evidence verification agent finds that the above-mentioned sentiment observation hypothesis and the above-mentioned user self-description conflict, it does not delete either of the above-mentioned sentiment observation hypothesis or the above-mentioned user self-description, but stores them as conflict information along with the corresponding evidence for each, and can generate additional verification questions based on the conflict information.

[0034] The story structuring agent above can manage each story component constituting the story candidate by classifying it into one of the elements directly detected from the free drawing, elements obtained from the user's self-description, and extension elements proposed by artificial intelligence.

[0035] The story structuring agent above can generate two or more different story candidates or ending candidates for the same free drawing and receive input from the user for selection, modification, combination, or rejection of the story candidates or ending candidates.

[0036] The above expert review unit provides the above expert with one or more of the emotion observation hypothesis generated by the above emotion observation agent, the verification result generated by the above evidence verification agent, the above user self-description, the searched evidence data, and the story candidate, and can store the initial analysis content generated by the artificial intelligence and the analysis content modified or approved by the above expert separately.

[0037] In one embodiment of the present invention, one or more of the following are stored in chronological order among artwork images, production process data, feature values, user self-descriptions, emotional observation hypotheses, retrieved evidence, expert review content, and story candidates regarding the user's free drawing generated at a plurality of points in time, and may further include an individual growth portfolio section that analyzes temporal changes among color, line, spatial composition, margins, repetitive symbols, relationships between shapes, story themes, and user self-descriptions.

[0038] In one embodiment of the present invention, when an expression related to self-harm, harm to others, abuse, violence, or the possibility of a crisis is detected from the free drawing, the danger state is not determined solely by said expression, but may further include a safety verification unit that additionally requires one or more of the user's direct statement, whether it is related to actual experience, whether it is related to media experience, whether it is repeated in multiple works, and expert verification.

[0039] In one embodiment of the present invention, the scope of access for at least two of the user, guardian, expert, researcher, and administrator is set differently for one or more of the image data, production process data, user self-description, searched evidence, emotion observation hypothesis, expert review content, and story candidate; the data included in the search target of the RAG knowledge search unit or the analysis target of the free drawing emotion observation and story development system is restricted according to the user's consent status; and the consent management unit may further include one or more actions among viewing, modifying, approving, externally transmitting, and deleting the data as an audit record. Effects of the invention

[0041] According to the present invention, by analyzing production process data together with the completed image of the free drawing, information such as the production sequence, changes in pen pressure, pauses, repetitions, and modifications that are difficult to confirm in the completed image can be utilized in the analysis of the free drawing.

[0042] In addition, by providing a distinction between objective observational facts and multiple emotional observation hypotheses, rather than directly determining individual visual features as specific emotional states, excessive interpretation of a single feature can be reduced.

[0043] In addition, the characteristics of free drawing can be analyzed contextually by utilizing the user's age, developmental level, production materials, production context, previous works, and user self-description.

[0044] In addition, by linking the evidence retrieved via the RAG method with sentiment observation hypotheses, the basis for the interpretation generated by the AI ​​can be traced.

[0045] In addition, when the emotion observation hypothesis and the user's self-description conflict, neither is removed but preserved as conflicting information, allowing experts to compare and review the different information.

[0046] In addition, by generating story candidates by distinguishing elements directly detected in free drawing, elements obtained from user self-descriptions, and extension elements proposed by artificial intelligence, it is possible to suppress story expansion unrelated to free drawing and user intent.

[0047] In addition, user-centered story creation can be supported by allowing users to select, modify, combine, or reject multiple story candidates.

[0048] In addition, by storing AI analysis content and expert review content separately, the process of generating and modifying analysis results can be tracked. Brief explanation of the drawing

[0050] FIG. 1 is a block diagram showing the overall configuration of a free drawing emotion observation and story development system according to one embodiment of the present invention. FIG. 2 is a drawing showing an example of a user screen displayed on a user terminal according to an embodiment of the present invention. FIG. 3 is a drawing showing an example of an expert screen displayed on an expert terminal according to an embodiment of the present invention. Specific details for implementing the invention

[0051] The present invention is capable of various modifications and may have various embodiments, and specific embodiments are illustrated in the drawings and described in detail. However, this is not intended to limit the present invention to specific embodiments, and it should be understood that it includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the invention.

[0052] The terms used in this application are used merely to describe specific embodiments and are not intended to limit the invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "having" are intended to indicate the presence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0053] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the present invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application.

[0054] Furthermore, in the description referring to the attached drawings, identical components are assigned the same reference numeral regardless of drawing symbols, and redundant descriptions thereof are omitted. In describing the present invention, if it is determined that a detailed description of related prior art could unnecessarily obscure the essence of the present invention, such detailed description is omitted.

[0055] The present invention relates to a free drawing emotion observation and story development system that supports emotion observation and story development regarding a user's free drawing.

[0057] FIG. 1 is a block diagram showing the overall configuration of a free drawing emotion observation and story development system according to one embodiment of the present invention.

[0058] Referring to FIG. 1, a free drawing emotion observation and story development system (100) according to one embodiment of the present invention includes a multimodal data input unit (110), a free drawing feature extraction unit (120), a context information structuring unit (130), a RAG knowledge search unit (140), an emotion observation agent (150), an evidence verification agent (160), a story structuring agent (170), and an expert review unit (180). The free drawing emotion observation and story development system (100) may further include one or more of an event division unit (111), an individual difference normalization unit (121), a self-explanatory induction agent (161), an individual growth portfolio unit (190), a safety confirmation unit (195), and a consent management unit (199).

[0059] Each of the above components may be implemented by integrating it into a single server, or by distributing it across multiple servers, cloud servers, user terminals, and expert terminals. Additionally, each component may be implemented by one or more processors and a program or artificial intelligence model executed by said processor.

[0060] In this specification, "free drawing" is not limited to reproducing a predetermined shape or correct answer exactly as it is, but may include drawings, doodles, line drawings, colored drawings, abstract expressions, or combinations thereof that a user freely creates on paper, a tablet, a smart device, or a digital canvas, etc.

[0061] Furthermore, in this specification, "emotional observation" does not mean a medical diagnosis or the confirmation of a mental illness, but may mean providing multiple possibilities for observation related to emotional expression using features objectively detected in free drawing, user self-description, and related supporting materials.

[0062] In this specification, "expert" means a person authorized to review the analysis results or story development results regarding free drawing, and is not limited to a specific occupation or qualification.

[0063] The multimodal data input unit (110) collects image data of the free drawing and production process data generated during the production process of the free drawing.

[0064] The free drawing feature extraction unit (120) calculates feature values ​​regarding one or more of lines, colors, shapes, spatial arrangements, margins, repetitions, relationships between shapes, and production sequences from image data and production process data.

[0065] The context information structuring unit (130) generates context information by associating one or more of the user's age, developmental level, production materials, activity purpose, production context, previous works and user self-description with the above feature values.

[0066] The RAG knowledge search unit (140) searches for source data corresponding to feature values ​​and context information from the knowledge base.

[0067] The emotion observation agent (150) generates objective observation facts based on feature values ​​and generates multiple emotion observation hypotheses based on objective observation facts, context information, and retrieved evidence.

[0068] The evidence verification agent (160) verifies whether there is a match or conflict between one or more of the multiple emotional observation hypotheses, user self-descriptions, and searched evidence.

[0069] A story structuring agent (170) generates multiple story candidates including one or more of a protagonist, background, goal, conflict, helper, emotional change, and ending based on one or more of a character, object, space, color, symbol, and user self-description extracted from a free drawing.

[0070] The expert review department (180) allows experts to review, modify, or approve one or more of the multiple emotional observation hypotheses and story candidates.

[0071] The event division unit (111) aligns two or more of image data, production process data, voice and non-verbal response information along a common time axis and divides one or more of the start or end of a stroke, color change, tool change, pen pressure change, pause for a certain period of time or longer, repeated overdrawing of the same area, erasing, modification, speech and non-verbal response into production events.

[0072] The individual difference normalization unit (121) normalizes feature values ​​based on one or more of the user's age, developmental level, motor skills, visual skills, communication style, production materials, task conditions and individual baselines derived from previous works.

[0073] The self-description inducing agent (161) generates questions regarding one or more of the names of shapes included in the free drawing, relationships between shapes, emotions of characters, causes of events, previous scenes, next scenes, important parts and parts to be modified, and collects user self-description in one or more of the following ways: text, voice, picture cards, facial expression cards, color selection, gaze selection, shape indication by finger, gestures and augmentative and alternative communication symbols.

[0074] The individual growth portfolio section (190) stores, in chronological order, one or more of the following: artwork images, production process data, feature values, user self-description, emotional observation hypothesis, searched evidence, expert review content, and story candidates regarding the user's free drawing created at multiple points in time, and analyzes one or more of the following temporal changes: color, line, spatial composition, margin, repetitive symbol, relationship between shapes, story theme, and user self-description.

[0075] When an expression related to self-harm, harm to others, abuse, violence, or the possibility of danger is detected in a free drawing, the safety verification unit (195) does not determine the danger state based solely on the expression, but additionally requires one or more of the following: a direct statement from the user, whether it is related to actual experience, whether it is related to media experience, whether it is repeated in multiple works, and expert verification.

[0076] The consent management unit (199) sets different access scopes for at least two of the users, guardians, experts, researchers, and administrators regarding one or more of image data, production process data, user self-descriptions, searched evidence materials, emotion observation hypotheses, expert review contents, and story candidates, and restricts the data included in the search targets of the RAG knowledge search unit (140) or the analysis targets of the free drawing emotion observation and story development system according to the user's consent status, and stores one or more of the actions of viewing, modifying, approving, externally transmitting, and deleting the data as audit records.

[0077] The multimodal data input unit (110) can collect production process data including one or more of the coordinates of the strokes constituting the free drawing, start time, end time, pen pressure, tilt, color, thickness, speed, direction of progress, stop, repeat, erase, and modification, and one or more of the production video, audio, description of the work, self-report data, expert observation data, gaze information, facial expression information and non-verbal response information of the free drawing.

[0078] The knowledge base of the RAG knowledge search unit (140) is divided into two or more areas among the official education and development materials area, the academic and verification materials area, the institution-approved practical materials area, the individual portfolio area, and the creative expansion materials area, and the RAG knowledge search unit (140) can select or restrict the areas to be searched based on one or more of the user's age, purpose of activity, purpose of analysis, consent status, access rights, and production context.

[0079] The emotion observation agent (150) stores objective observation facts generated from feature values ​​and multiple emotion observation hypotheses generated for objective observation facts separately from each other, and can generate two or more different emotion observation hypotheses for the same objective observation fact.

[0080] The basis verification agent (160) calculates the reliability of interpretation for each of the multiple emotional observation hypotheses based on two or more of the visual feature consistency, production process consistency, user self-description consistency, search basis suitability, consistency with previous works, and conflicting data ratio, and for emotional observation hypotheses where the reliability of interpretation does not meet pre-set conditions or where conflicting data exists, it may assign one or more status information such as need for additional verification, existence of multiple possibilities, and judgment pending.

[0081] When the evidence verification agent (160) finds that the emotion observation hypothesis and the user self-description conflict, it does not delete either the emotion observation hypothesis or the user self-description, but saves them as conflicting information along with the corresponding evidence for each, and can generate additional verification questions based on the conflicting information.

[0082] The story structuring agent (170) can manage each story component constituting a story candidate by classifying it into one of the elements directly detected from free drawing, elements obtained from user self-description, and extension elements proposed by artificial intelligence.

[0083] The story structuring agent (170) can generate two or more different story candidates or ending candidates for the same free drawing and receive input from the user to select, modify, combine, or reject the story candidates or ending candidates.

[0084] The expert review unit (180) provides one or more of the following to the expert: an emotion observation hypothesis generated by the emotion observation agent (150), a verification result generated by the evidence verification agent (160), a user self-description, searched evidence data, and a story candidate, and can store the original analysis content generated by the artificial intelligence and the analysis content modified or approved by the expert separately.

[0086] The multimodal data input unit (110) can collect different types of data related to the user's free drawing.

[0087] The multimodal data input unit (110) can collect image data such as images of free drawing produced on paper that are photographed or scanned, images generated on a digital canvas, layer information, or vector information. Additionally, the multimodal data input unit (110) can collect one or more of the coordinates, start time, end time, pen pressure, tilt, color, thickness, speed, direction of progress, stop, repeat, erase, and modification of each stroke constituting the free drawing as production process data.

[0088] The multimodal data input unit (110) can further collect production video and audio. For example, the progress of free drawing, selection of tools, or user actions can be obtained from the production video, and the user's utterances related to free drawing can be obtained from the audio. Additionally, the title of the work, description of the work, self-report data, expert observation data, gaze information, facial expression information, and other non-verbal response information can be further collected.

[0089] The event division unit (111) can sort multiple types of data collected by the multimodal data input unit (110) based on time information and classify changes occurring during the production process of free drawing into production event units. For example, production events may be the start or end of a new stroke, a change in color, a change in tool, a change in pen pressure, a pause for longer than a predetermined time, repeated overdrawing of the same area, erasing, modification, speech, or non-verbal response.

[0090] The event division unit (111) can correspond the time when a specific area of ​​free drawing is produced with the utterance or non-verbal response that occurred at that time. For example, if the user utters an explanation about the cloud at the time when the user is repeatedly tracing the cloud, the event division unit (111) can store the repeated tracing and the utterance as a single production event or as linked production events.

[0091] The free drawing feature extraction unit (120) can calculate feature values ​​for analyzing free drawing from image data and production process data. Line features may include one or more of line length, thickness, direction, curvature, intersection, repetition, connection, or disconnection. Motion features may include one or more of average pen pressure, maximum or minimum pen pressure, amount of change in pen pressure, stroke speed, acceleration, deceleration, spacing between strokes, and pause time. Color features may include one or more of the type of color used, area ratio by color, brightness, saturation, color contrast, color repetition, and color change order. Spatial features may include one or more of the position of a shape within the free drawing, screen centeriness, margin ratio, shape density, shape size, overlap, directionality, or balance. Relationship features may include one or more of the distance between multiple shapes, contact status, connection relationship, size ratio, inclusion relationship, and separation relationship.

[0092] The free drawing feature extraction unit (120) may also detect shapes or symbols such as people, houses, trees, clouds, the sun, animals, roads, doors, stairs, letters, numbers, or repeating symbols.

[0093] The individual difference normalization unit (121) can normalize the feature values ​​calculated by the free drawing feature extraction unit (120) based on the user's personal conditions or production conditions.

[0094] The individual difference normalization unit (121) may consider one or more of the user's age, developmental level, motor skills, visual skills, communication method, production materials, and task conditions. For example, even if the line thickness is the same, a free drawing made using a thick crayon and a free drawing made using a precise digital pen may not be evaluated on the same standard.

[0095] In addition, the individual difference normalization unit (121) can set an individual baseline using the user's previous works. For example, if a user has repeatedly used black widely in previous works, the difference from the individual baseline can be calculated without judging the fact that black appears widely in the current work itself as a new change.

[0096] The context information structuring unit (130) can link free drawing feature values ​​with context information necessary for interpreting the corresponding features. The context information may include one or more of the user's age, developmental level, production materials, activity purpose, task conditions, production environment, previous works, user self-description, and expert observation data.

[0097] The context information structuring unit (130) can correspond a specific shape or a specific production event with the user's self-description. For example, a structured context information can be generated by linking the shape information regarding a black cloud with repeated overdrawing, high pen pressure, the user's self-description regarding the cloud, and related previous works.

[0098] The RAG knowledge search unit (140) can search for source materials corresponding to free drawing feature values ​​and context information from the knowledge base. The knowledge base may be divided into multiple material areas. As an example, the knowledge base may include an official education and development material area, an academic and verification material area, an institution-approved practical material area, an individual portfolio area, and a creative expansion material area.

[0099] The RAG knowledge search unit (140) can select or restrict the data area to be searched by using one or more of the user's age, purpose of activity, purpose of analysis, consent status, scope of access, and production context. For example, when performing analysis for the purpose of story creation, it can be set to prioritize the search for storytelling or creative expansion materials, and when performing analysis for emotional observation, it can be set to prioritize the search for educational and verification materials applicable to the user. The source of the data, the target of application, the scope of application, and the degree of relevance to the corresponding free drawing characteristics can be stored together with the searched source data.

[0100] The emotion observation agent (150) can first generate facts that are objectively detected from free drawing. Objective observation facts may not be expressions that directly indicate a specific emotional state, but rather facts confirmed in the data. For example, information such as “black cloud areas are expressed large,” “pen pressure in cloud areas is higher than the overall average of the work,” and “repeated overdrawing occurs in the same cloud area” can be generated as objective observation facts.

[0101] The emotion observation agent (150) can generate multiple emotion observation hypotheses for a single objective observation fact, rather than generating only one interpretation. For example, regarding the same characteristic of repeated overdrawing, different hypotheses can be generated, such as emphasis on the subject, high kinetic energy, expression of tension in the scene, and expression intention in color or material.

[0102] Therefore, the emotion observation agent (150) can store objective observation facts and emotion observation hypotheses as separate data.

[0103] The self-explanation inducing agent (161) may provide the user with questions to confirm the meaning of the free drawing before the emotion observation agent (150) unilaterally determines the meaning of the free drawing. The questions may be questions regarding one or more of the title of the free drawing, names of shapes, relationships between shapes, emotions of characters, events occurring, previous scenes, next scenes, important parts, or parts to be modified.

[0104] Users can respond to questions via text or voice, and may also respond using picture cards, facial expression cards, color selection, selection of specific locations in free drawing, gaze selection, finger pointing, gestures, or augmentative and alternative communication symbols.

[0105] The response obtained by the self-explanation induction agent (161) is stored as user self-explanation and can be used for verification by the basis verification agent (160).

[0106] The evidence verification agent (160) can verify the evidence for each of the multiple emotion observation hypotheses generated by the emotion observation agent (150).

[0107] Specifically, the evidence verification agent (160) can determine whether visual features corresponding to the hypothesis exist in the actual free drawing, whether production process data supports the hypothesis, whether user self-descriptions match the hypothesis, and whether the evidence data searched by the RAG knowledge search unit (140) is applicable to the context of the user. In addition, it can determine whether free drawing features, user self-descriptions, or evidence data that contradict the sentiment observation hypothesis exist.

[0108] The evidence verification agent (160) can calculate the interpretation reliability for each sentiment observation hypothesis using two or more of the visual feature consistency, production process consistency, user self-description consistency, search basis suitability, consistency with previous works, and conflicting data ratio. The interpretation reliability may be provided in the form of a numerical value, a grade, or status information.

[0109] If the reliability of the interpretation does not satisfy the pre-set conditions or there are many conflicting data, the basis verification agent (160) may not treat the hypothesis as a definitive result and may assign status information such as "additional verification needed," "multiple possibilities exist," or "judgment withheld."

[0110] In one embodiment of the present invention, the basis verification agent (160) can determine whether there is a match or conflict between the sentiment observation hypothesis and the user's self-description.

[0111] Even if the user's self-description conflicts with the sentiment observation hypothesis, the evidence verification agent (160) may not delete either one or modify it to match the other. In this case, the sentiment observation hypothesis, the evidence supporting the hypothesis, the user's self-description, and the evidence corresponding to the user's self-description may be distinguished from each other and stored as conflicting information.

[0112] In addition, if conflicting information exists, additional verification questions to identify the cause of the conflict can be generated and provided to the user through a self-explanatory inducing agent (161). If the conflict is resolved by the user's additional response, the verification result can be updated, and if the conflict is not resolved, multiple possibilities can be maintained in parallel.

[0113] A story structuring agent (170) can extract elements that constitute a story from freehand drawings and user self-descriptions. The story components may include one or more of a protagonist, supporting characters, a temporal setting, a spatial setting, a goal, a conflict, a helper, the beginning of an event, the development of an event, emotional changes, and a conclusion.

[0114] The story structuring agent (170) can distinguish and manage the sources of elements constituting a story candidate. Specifically, each story component can be classified into one of the elements directly detected from free drawing, elements obtained from user self-description, and extension elements proposed by artificial intelligence. For example, if "cloud" and "person" are directly detected in free drawing and the user explains "the cloud is angry because it cannot make it rain," the cloud and person can be managed as directly detected elements, and the state and reason of the cloud can be managed as user-description elements. In addition, if the AI ​​suggests an event where the cloud and the person have a conversation, that event can be separately marked as an AI extension element.

[0115] The story structuring agent (170) can generate multiple different story candidates or ending candidates for the same free drawing. The user can select one of the multiple candidates, modify some of the content, combine multiple candidates, or reject all candidates.

[0116] The expert review unit (180) can provide information generated or verified by AI to the expert terminal. The information provided to the expert may include one or more of the following: original freehand drawing, production process data, objective observation facts, sentiment observation hypotheses, user self-description, RAG search basis, basis verification results, conflicting information, interpretation reliability, and story candidates.

[0117] Experts can review the provided information and approve, modify, withhold, or delete sentiment observation hypotheses or story candidates.

[0118] The expert review unit (180) can store the analysis content originally generated by the artificial intelligence and the analysis content modified or approved by the expert separately. Therefore, it is possible to track whether the AI ​​analysis result is identical to the expert's final review result or has been modified.

[0119] The individual growth portfolio section (190) can store the user's free drawings and analysis information regarding them generated at multiple points in time in chronological order. The information stored may include one or more of the following: artwork images, production process data, feature values, user self-descriptions, emotional observation hypotheses, search grounds, expert review content, and generated stories.

[0120] The individual growth portfolio section (190) can analyze the temporal changes in colors, lines, spatial composition, blank spaces, repetitive symbols, relationships between shapes, story themes, or user self-descriptions appearing in multiple works. However, the user's emotional state is not determined to have improved or worsened solely based on the fact that specific characteristics have changed over time, but rather the production materials, task conditions, expression techniques, user self-descriptions, and expert review results can be considered together.

[0121] The safety verification unit (195) may perform an additional verification procedure if an expression requiring safety verification is detected in the free drawing or user description. For example, even if an expression related to self-harm, harm to others, abuse, violence, or the possibility of a crisis is detected, the dangerous state may not be determined based solely on that expression.

[0122] The safety verification unit (195) may additionally require one or more of the following: a direct statement from the user, whether it is related to actual experience, whether it is related to media experience, whether it is repeated in multiple works, or confirmation by an expert. Accordingly, it is possible to prevent a picture expressing imagination, story creation, or experience from external media from being automatically confirmed as an actual dangerous state.

[0123] The consent management department (199) can manage the scope of access to materials used in the free drawing emotion observation and story development system (100) and the user's consent status. For example, the subjects of access can be set differently for image data, production process data, user self-description, search basis, emotion observation hypothesis, expert review content, and story candidate.

[0124] The consent management unit (199) may restrict the information that the RAG knowledge search unit (140) can use for search or as an analysis target according to the user's consent status. In addition, the entity that performed actions such as viewing, modifying, approving, externally transmitting, and deleting data, and the time of execution, may be matched and stored as an audit record.

[0126] FIG. 2 is a drawing showing an example of a user screen displayed on a user terminal according to an embodiment of the present invention.

[0127] Figure 2 shows an example of a screen of a user terminal connected to a free drawing emotion observation and story development system (100).

[0128] Referring to FIG. 2, a free drawing display area for displaying a free drawing created by the user may be provided on the screen of the user terminal.

[0129] The freehand drawing display area can display the original freehand drawing along with the production date and the sequence of the production process, and the user can also selectively play the production video.

[0130] Questions generated by the self-explanatory agent (161) may be provided in the question area of ​​the screen.

[0131] As shown in Fig. 2, for example, questions may include "What is the title of this picture?", "Who or what appears in this picture?", "What is happening right now?", "How do you feel at this time?", etc.

[0132] As additional questions, questions such as "What happened before?", "What will happen next?", "What is the most important part?", and "Is there anything you would like to modify?" may be provided.

[0133] Users can respond to questions through keyboard input, voice input, list selection, selection of a location on an image, or selection of a facial expression.

[0134] Multiple story candidates may be provided on the user screen. Each story candidate may be displayed along with visual information representing simple text or elements extracted from freehand drawings.

[0135] The user can select one of multiple story candidates and modify the content of the selected story.

[0136] The story selected or modified by the user can be saved along with the user selection information and subsequently provided to the expert review department (180).

[0138] FIG. 3 is a drawing showing an example of an expert screen displayed on an expert terminal according to an embodiment of the present invention.

[0139] Figure 3 shows an example of a screen provided to an expert terminal through an expert review unit (180).

[0140] Referring to Fig. 3, one or more of case information, original free drawing, objective observation facts, emotional observation hypothesis, user self-description, RAG search basis data, basis verification results, conflicting information, story candidates, and expert review opinions may be displayed separately on the expert screen.

[0141] The original free drawing area can provide functions such as zooming in, zooming out, rotating, or checking the production process.

[0142] In the objective observation fact domain, AI can display facts directly detected from image and production process data.

[0143] In the emotion observation hypothesis area, hypothesis identification information, hypothesis content, and interpretation reliability for each hypothesis may be displayed. For example, by classifying multiple hypotheses as H1, H2, H3, etc., experts can review each hypothesis individually.

[0144] In the user self-description area, user responses obtained from the user screen of FIG. 2 can be displayed by question.

[0145] In the source material area, the type, content, source, and relevance of the data searched by the RAG knowledge search unit (140) can be displayed.

[0146] The evidence verification results area may display the number of supporting grounds, the number of conflicting grounds, key evidence, interpretation reliability, and verification status for each emotion observation hypothesis. For example, the verification status may be displayed as "Supported," "Further verification needed," "Multiple possibilities," or "Judgment reserved."

[0147] In the conflict information area, conflicts between AI-generated hypotheses and user self-explanations can be displayed in parallel. Additionally, to verify the conflict information, follow-up verification questions generated by the AI ​​and the user's additional responses to them can be displayed together.

[0148] The story candidate area may provide multiple story candidates generated by the story structuring agent (170).

[0149] The expert review and opinion area may include an input area where experts can enter review comments and an area for entering processing information such as approval, approval after modification, hold, or deletion.

[0150] The expert review unit (180) can store the final processing result of the expert separately from the AI ​​initial analysis result.

[0152] An example of a freehand drawing depicting a rainy scene in one embodiment of the present invention is as follows. An example is described where a user creates a freehand drawing including a house, a person, a tree, rain clouds, and the sun.

[0153] The multimodal data input unit (110) can collect the coordinates, pressure, production order, and color information of each stroke along with the final image of the free drawing.

[0154] The event division unit (111) can distinguish production events such as the production of a house, the production of a sun, the production of a cloud, repeated overdrawing, and the production of a person.

[0155] The free drawing feature extraction unit (120) can calculate feature values ​​such as black clouds, the relative positions of clouds and the sun, the size of houses and people, and the distribution of colors.

[0156] The emotion observation agent (150) can first generate objective observation facts such as “black clouds are depicted relatively large,” “a scene of rain is depicted,” and “a house is placed in the center of the screen.”

[0157] Next, multiple different emotional observation hypotheses are generated based on each observation. At this time, each hypothesis does not determine the user's actual emotional state but is provided as a possible interpretation that can be expressed in free drawing.

[0158] The self-explanatory inducing agent (161) can ask the user about the title of the picture, the object appearing in the picture, the event that is happening, and the emotions of the characters.

[0159] If the user describes, "It is raining, so my friend is running home without an umbrella," that description is saved as a user-created description.

[0160] The evidence verification agent (160) compares the sentiment observation hypothesis generated by the AI ​​with the user's self-description and the retrieved evidence data. For example, if a hypothesis representing a specific emotional state is generated based only on the black clouds in the picture, and the user describes the clouds as a simple weather scene, the conflict between the two pieces of information can be recorded.

[0161] The free drawing emotion observation and story development system (100) can preserve the AI ​​hypothesis and the user's self-explanation along with their respective grounds instead of deleting the hypothesis.

[0162] The story structuring agent (170) can generate multiple story candidates, such as “a story about a friend going home on a rainy day” and “a story about meeting a friend outside after the rain has stopped.”

[0163] The user can select or modify desired story candidates through the user screen of Fig. 2.

[0164] As shown in Fig. 3, the original free drawing, objective observation facts, multiple emotional observation hypotheses, user self-description, search basis, verification results, conflicting information, and story candidates may be provided to the expert.

[0166] In one embodiment of the present invention, a free drawing including a repetitive shape is exemplified as follows.

[0167] When a user repeatedly expresses multiple identical or similar circular shapes, the free drawing feature extraction unit (120) can calculate the number of repetitions, the size of the circular shapes, the position, the order of production, and the characteristics of the strokes.

[0168] The emotion observation agent (150) does not determine the repetition itself as a specific emotional state, but can generate multiple hypotheses such as shape search, interest in a specific object, repetitive expression techniques, or story elements.

[0169] If the user describes each of the prototypes as a family home, the basis verification agent (160) can update the verification information for the hypothesis that matches the user's self-description.

[0170] A story structuring agent (170) can structure each prototype into different houses and generate story candidates by using the space or connecting line between the houses as a path for the story.

[0172] In one embodiment of the present invention, if there is a conflict between the sentiment observation hypothesis and the user self-description, production process data, or searched evidence data, the evidence verification agent (160) may not merely store the conflict information but may analyze the cause of the conflict.

[0173] For example, the cause of the conflict may be one or more of the following: the applicable age of the retrieved source material differs from that of the user; the production materials or purpose of the activity of the retrieved source material differ from the production conditions of the current free drawing; the meaning included in the user's self-description was not reflected in the initial search query; or the difference in characteristics between the individual baseline and the current work was not reflected in the initial search.

[0174] The evidence verification agent (160) can change the search conditions provided to the RAG knowledge search unit (140) in response to the above conflicting causes. For example, it can add one or more of user age, production materials, activity purpose, user self-description, difference from individual baseline, and conflicting sentiment observation hypotheses to the search query, or change the knowledge domain to be searched.

[0175] The RAG knowledge search unit (140) re-searches the evidence data according to the changed search conditions, and the evidence verification agent (160) can recalculate the reliability of the interpretation of the corresponding sentiment observation hypothesis using the re-searched evidence data.

[0176] Accordingly, even if the initial search results conflict with the user's self-description, the judgment is not terminated based on the initial search results; instead, by performing a subsequent search that reflects the cause of the conflict, the appropriateness of the evidence for the sentiment observation hypothesis can be improved.

[0178] In one embodiment of the present invention, the individual difference normalization unit (121) may generate change information indicating the difference between the individual baseline calculated from the user's previous works and the feature value calculated in the current free drawing. For example, the change information may include one or more of the following: a change in the area ratio by color, a change in average pen pressure, a change in the number of repeated overdrawings, a change in the margin ratio, a change in the size or position of a specific shape, and a change in the frequency of appearance of repeated symbols.

[0179] The context information structuring unit (130) can link the above change information with the production event of the current free drawing, the user's self-description, and the production context.

[0180] The RAG knowledge search unit (140) can use not only the current feature value itself but also change information regarding individual baselines as search conditions. For example, features with a large amount of change compared to individual baselines can be selected as priority search target features, and supporting data related to the change can be searched.

[0181] Accordingly, it is possible to distinguish between the expressive characteristics that users habitually and repeatedly display and the expressive characteristics that have newly changed in the current work, and to provide supporting data that corresponds to individual user changes rather than simple group standards.

[0182] The individual growth portfolio section (190) can check whether the same change is temporary or repeated across multiple works by storing the above change information, the searched evidence and expert review results related to the change in chronological order.

[0184] In one embodiment of the present invention, the story structuring agent (170) can adjust the range of creation of extension elements proposed by artificial intelligence based on one or more of the interpretation reliability of the sentiment observation hypothesis and conflicting information calculated by the evidence verification agent (160). For example, if the degree of agreement between elements directly detected in free drawing, user self-description, and searched evidence is high and the ratio of conflicting information is low, the story structuring agent (170) can generate multiple AI extension elements regarding the development of events, helpers, emotional changes, or endings based on the information.

[0185] On the other hand, if the reliability of interpretation does not meet the pre-set conditions or if there is a large conflict between the user's self-description and the sentiment observation hypothesis, the story structuring agent (170) may limit the generation of AI extension elements and construct story candidates based on elements directly detected from free drawing and elements obtained from the user's self-description.

[0186] In addition, in the above case, the story structuring agent (170) may generate multiple selectable subsequent scenes instead of arbitrarily determining the ending or event, or request additional selection from the user using a self-explanatory inducing agent (161).

[0187] Accordingly, when the evidence for the emotion observation hypothesis is insufficient, it is possible to limit the excessive generation of stories by artificial intelligence that are unrelated to free drawing or the user's intention, and to adjust the scope of story generation in response to the appropriateness of the evidence.

[0188] Although the present invention has been described above using several preferred embodiments, these embodiments are illustrative and not limiting. Those skilled in the art will understand that various changes and modifications can be made without departing from the spirit of the invention and the scope of rights set forth in the appended claims. Explanation of the symbols

[0190] 100: Free Drawing Emotion Observation and Story Development System 110: Multimodal Data Input Section 111: Case Division 120: Free Drawing Feature Extraction Section 121: Individual Differences Normalization Section 130: Contextual Information Structured Section 140: RAG Knowledge Search Department 150: Emotion Observation Agent 160: Evidence Verification Agent 161: Self-explanatory inducing agent 170: Story Structuring Agent 180: Expert Review Department 190: Individual Growth Portfolio Department 195: Safety Confirmation Department 199: Consent Management Department

Claims

Claim 1 A free drawing emotion observation and story development system that supports emotion observation and story development regarding a user's free drawing, comprising: a multimodal data input unit that collects image data of the free drawing and production process data generated during the production process of the free drawing; a free drawing feature extraction unit that calculates feature values ​​regarding one or more of line, color, shape, spatial arrangement, margin, repetition, relationships between shapes, and production sequence from the image data and production process data; a context information structuring unit that generates context information by linking one or more of the user's age, developmental level, production materials, purpose of activity, production context, previous works, and user self-description with the feature values; a RAG knowledge search unit that searches for supporting materials corresponding to the feature values ​​and context information from a knowledge base; an emotion observation agent that generates objective observation facts based on the feature values ​​and generates multiple emotion observation hypotheses based on the objective observation facts, context information, and searched supporting materials; a supporting material verification agent that verifies whether one or more of the multiple emotion observation hypotheses, the user self-description, and searched supporting materials match or conflict; and people, objects, and spaces extracted from the free drawing A free-drawing emotion observation and story development system characterized by comprising: a story structuring agent that generates multiple story candidates including one or more of a protagonist, background, goal, conflict, helper, emotional change, and ending based on one or more of colors, symbols, and the user's self-description; and an expert review unit that allows an expert to review, modify, or approve one or more of the multiple emotion observation hypotheses and story candidates. Claim 2 A free drawing emotion observation and story development system according to claim 1, wherein the multimodal data input unit collects production process data including one or more of the coordinates of a stroke constituting the free drawing, a start time, an end time, pen pressure, inclination, color, thickness, speed, direction of progress, stopping, repeating, erasing, and modification, and one or more of the production video, audio, artwork description, self-report data, expert observation data, gaze information, facial expression information, and non-verbal response information of the free drawing. Claim 3 A free drawing emotion observation and story development system according to claim 2, further comprising an event division unit that aligns two or more of the image data, production process data, voice and non-verbal response information along a common time axis, and divides one or more of the start or end of a stroke, color change, tool change, pen pressure change, pause for a certain period of time or longer, repeated overdrawing of the same area, erasing, modification, utterance and non-verbal response into production events. Claim 4 A free drawing emotion observation and story development system according to claim 1, further comprising an individual difference normalization unit that normalizes the feature value based on one or more of the user's age, developmental level, motor ability, visual ability, communication method, production material, task conditions, and individual baseline derived from previous works. Claim 5 A free drawing emotion observation and story development system according to claim 1, wherein the knowledge base of the RAG knowledge search unit is divided into two or more areas among an official education and development material area, an academic and verification material area, an institution-approved practical material area, an individual portfolio area, and a creative expansion material area, and wherein the RAG knowledge search unit selects or restricts the area to be searched based on one or more of the user's age, activity purpose, analysis purpose, consent status, access rights, and production context. Claim 6 A free-drawing emotion observation and story development system according to claim 1, wherein the emotion observation agent stores objective observation facts generated from the feature values ​​and a plurality of emotion observation hypotheses generated for the objective observation facts separately from each other, and generates two or more different emotion observation hypotheses for the same objective observation facts. Claim 7 A free drawing emotion observation and story development system according to claim 1, further comprising a self-description induction agent that generates questions regarding one or more of the names of shapes included in the free drawing, relationships between shapes, emotions of characters, causes of events, previous scenes, next scenes, important parts, and parts to be modified, and collects the user's self-description in one or more ways among text, voice, picture cards, facial expression cards, color selection, gaze selection, shape indication by finger, gestures, and augmentative and alternative communication symbols. Claim 8 A free drawing emotion observation and story development system according to claim 1, wherein the basis verification agent calculates interpretation reliability for each of the plurality of emotion observation hypotheses based on two or more of visual feature consistency, production process consistency, user self-description consistency, search basis suitability, consistency with previous works, and conflicting data ratio, and assigns one or more state information among additional verification required, existence of multiple possibilities, and judgment pending for emotion observation hypotheses where the interpretation reliability does not satisfy preset conditions or where conflicting data exists. Claim 9 A free drawing emotion observation and story development system according to claim 1, wherein, when the emotion observation hypothesis and the user self-description conflict, the basis verification agent stores the conflict information along with the basis corresponding to each, without deleting either the emotion observation hypothesis or the user self-description, and generates additional verification questions based on the conflict information. Claim 10 A free drawing sentiment observation and story development system according to claim 1, wherein the story structuring agent classifies and manages each story component constituting the story candidate into one of the elements directly detected from the free drawing, elements obtained from the user's self-description, and extension elements proposed by artificial intelligence. Claim 11 A free drawing sentiment observation and story development system according to claim 1, wherein the story structuring agent generates two or more different story candidates or ending candidates for the same free drawing and receives input from the user for selection, modification, combination, or rejection of the story candidates or ending candidates. Claim 12 A free drawing emotion observation and story development system according to claim 1, wherein the expert review unit provides to the expert one or more of an emotion observation hypothesis generated by the emotion observation agent, a verification result generated by the evidence verification agent, the user self-description, searched evidence materials, and story candidates, and stores the initial analysis content generated by the artificial intelligence and the analysis content modified or approved by the expert separately. Claim 13 A free drawing emotion observation and story development system according to claim 1, further comprising a personal growth portfolio section that stores, in chronological order, one or more of artwork images, production process data, feature values, user self-descriptions, emotion observation hypotheses, retrieved evidence, expert review contents, and story candidates regarding the user's free drawing generated at multiple points in time, and analyzes temporal changes in one or more of color, line, spatial composition, margin, repetitive symbol, relationships between shapes, story themes, and user self-descriptions. Claim 14 A free drawing emotion observation and story development system according to claim 1, characterized in that when an expression related to self-harm, harm to others, abuse, violence, or potential crisis is detected from the free drawing, the system does not determine the danger state solely based on said expression, but further includes a safety verification unit that additionally requires one or more of the user's direct statement, whether it is related to actual experience, whether it is related to media experience, whether it is repeated in multiple works, and expert verification. Claim 15 A free drawing emotion observation and story development system according to claim 1, further comprising a consent management unit that sets different access scopes for at least two of a user, guardian, expert, researcher, and administrator for one or more of the image data, production process data, user self-description, searched evidence, emotion observation hypothesis, expert review content, and story candidate, restricts the data included in the search target of the RAG knowledge search unit or the analysis target of the free drawing emotion observation and story development system according to the user's consent status, and stores one or more actions among viewing, modifying, approving, externally transmitting, and deleting the data as audit records.

Images