Method and apparatus for operating artificial intelligence platform

Abstract

The disclosure below relates to a method for operating an artificial intelligence platform, and the method may comprise: learning multi-modal personality data in an artificial neural network model to obtain a plurality of personality models; linking the plurality of personality models to a pipeline-based artificial intelligence service; generating a user personality model by combining the plurality of personality models on the basis of a user input to the artificial intelligence service; and updating the artificial intelligence service on the basis of the user personality model.

Description

Artificial intelligence platform operation method and device thereof

[0001] The following disclosure relates to an artificial intelligence platform operation method and device.

[0002] An AI platform facilitates AI model training and service. Typically, an AI platform may include features for development convenience, such as automated development environment setup, dataset management, training, and service delivery, as well as administrative functions such as model management, resource management, and monitoring. AI platforms can manage multiple versions of AI learning results, automatically distribute them for service delivery, and utilize automated load balancing for requests to ensure smoother request processing.

[0003] The embodiments relate to a method of learning a personality model on an artificial intelligence platform and applying it to an artificial neural network model so that users can develop an artificial neural network model that reflects the desired form of personality.

[0004] The problems to be solved by the present invention are not limited to the problems described above, and problems not mentioned can be clearly understood by a person having ordinary skill in the technical field to which the present invention pertains from this specification and the attached drawings.

[0005] A method of operating an artificial intelligence platform according to one embodiment may include a step of acquiring a plurality of personality models by learning multi-modal personality data in an artificial neural network model, a step of linking the plurality of personality models to a pipeline-based artificial intelligence service, a step of generating a user personality model by combining the plurality of personality models based on a user's input to the artificial intelligence service, and a step of updating the artificial intelligence service based on the user personality model.

[0006] The above multimodal personality data may include at least one of verbal, paralinguistic and non-verbal information.

[0007] Each of the above multiple personality models is trained with personality data labeled with the Big 5 (OCEAN) factor, and may include personality models specialized for one personality.

[0008] The above linking step may include a step of providing the plurality of personality models to a user interface of the artificial intelligence service.

[0009] The step of generating the user personality model may include a step of determining priorities and degrees of reflection among the plurality of personality models based on the user's selection input and the user's prompt input for the user interface.

[0010] The step of generating the user personality model may include a step of evaluating an intermediate user personality model that combines the plurality of personality models, and a step of repeatedly performing an operation of combining the plurality of personality models by changing the priority and degree of reflection between the plurality of personality models until the user-inputted setting performance is achieved.

[0011] The above artificial intelligence service can generate an artificial neural network model suitable for the user based on the user's input.

[0012] The above user personality model can preprocess the user input for the language model of the artificial intelligence service or postprocess the output of the language model of the artificial intelligence service.

[0013] The above-mentioned artificial intelligence service, after the update is completed, may further include a step of providing a user interface for distributing an artificial neural network model generated according to the user's input.

[0014] An electronic device according to one embodiment may include a memory including instructions, a processor for learning multi-modal personality data in an artificial neural network model to obtain a plurality of personality models, linking the plurality of personality models to a pipeline-based artificial intelligence service, generating a user personality model by combining the plurality of personality models based on a user input to the artificial intelligence service, and updating the artificial intelligence service based on the user personality model.

[0015] Figure 1 is a flowchart illustrating the operation of an artificial intelligence platform according to one embodiment.

[0016] FIG. 2 is a block diagram schematically illustrating a personality model service according to one embodiment.

[0017] Figure 3 schematically illustrates the use of a personality model in a pipeline format according to one embodiment.

[0018] FIG. 4 is a diagram illustrating an example of a pipeline according to one embodiment.

[0019] FIG. 5 is a block diagram illustrating an electronic device according to one embodiment.

[0020] Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only and may be modified and implemented in various forms. Therefore, the actual implementation is not limited to the specific embodiments disclosed, and the scope of this specification includes modifications, equivalents, or alternatives within the technical concepts described in the embodiments.

[0021] Although terms such as "first" or "second" may be used to describe various components, these terms should be interpreted solely to distinguish one component from another. For example, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component.

[0022] When it is said that a component is "connected" to another component, it should be understood that it may be directly connected or connected to that other component, but there may also be other components in between.

[0023] Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, the terms "comprises" or "has" should be understood to indicate the presence of a described feature, number, step, operation, component, part, or combination thereof, but not to exclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0024] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art. Terms defined in commonly used dictionaries should be interpreted to have a meaning consistent with their meaning in the context of the relevant technology, and will not be interpreted in an idealized or overly formal sense unless explicitly defined herein.

[0025] Hereinafter, embodiments will be described in detail with reference to the attached drawings. In the description with reference to the attached drawings, identical components are assigned the same reference numerals regardless of the drawing numbers, and redundant descriptions thereof will be omitted.

[0026]

[0027] Figure 1 is a flowchart illustrating the operation of an artificial intelligence platform according to one embodiment.

[0028] For convenience of explanation, steps (110 to 140) are described as being performed using the electronic device (400) illustrated in FIG. 4. However, these steps (110 to 140) may be utilized via any other suitable electronic device and within any suitable system.

[0029] Furthermore, the operations of FIG. 1 may be performed in the order and manner illustrated, but the order of some operations may be changed or some operations may be omitted without departing from the spirit and scope of the illustrated embodiment. Multiple operations illustrated in FIG. 1 may be performed in parallel or simultaneously.

[0030] In step (110), a processor according to one embodiment (e.g., processor (520) of FIG. 5) can acquire multiple personality models by learning multi-modal personality data in an artificial neural network model. The processor (520) can control the overall operation of the electronic device (500). In one embodiment, the processor (520) may be implemented as an array of a plurality of logic gates, or may be implemented as a combination of a general-purpose microprocessor and a memory storing a program that can be executed in the microprocessor. In addition, it will be understood by those skilled in the art that the present embodiment may be implemented as other types of hardware.

[0031] Artificial neural network models may include artificial intelligence models such as large language models (LLMs).

[0032] Multimodal personality data can represent data that understands and expresses an individual's personality through various types of information sources. Multimodal personality data can include information collected from various media, including language, voice, images, video, text, social media posts, and other sources. Using multimodal personality data to understand and predict personality can help with sentiment analysis, behavioral prediction, improving user experience, marketing, and developing personalized services.

[0033] For example, multimodal personality data can include linguistic data, voice data, image and video data, text data, and interaction data. Linguistic data expressed in text or speech can be used to identify personality traits and emotions. Linguistic data can be collected from emails, social media posts, chat logs, and other sources. Voice analysis can identify voice tone, speed, intonation, vocal energy, and emotional state. Voice data can be obtained from voice messages, phone calls, and recorded conversations. Images and video data can provide visual information through photos and videos. Facial expressions, posture, environment, and visual content in images and videos can be used to identify personality traits. Text data collected from text messages, social media posts, and web articles can be used to understand personality traits and interests. Interaction data can be a variety of data generated when individuals interact with digital services. Analysis of interaction data can reveal user preferences and behavior by analyzing click patterns, search queries, purchase history, and app usage.

[0034] Multimodal personality data may include at least one of linguistic, paralinguistic, and non-linguistic information. Linguistic information, paralinguistic information, and non-linguistic information are different types of information that play important roles in fields related to communication and information processing.

[0035] Linguistic information can refer to information conveyed through language. Linguistic information is expressed in linguistic units such as words, sentences, and documents, and can include linguistic structure, grammar, meaning, and vocabulary. For example, words and sentences contained in text documents, emails, web pages, novels, and papers can be linguistic information.

[0036] Paralinguistic information can represent nonlinguistic characteristics of speech, such as voice, pronunciation, intensity, intonation, tone, speed, and emotional quality of speech. Paralinguistic information can be used alongside language to convey emotional nuance or emphasis. For example, in a voice message, the intensity, intonation, and emotional tone of speech can be considered paralinguistic information.

[0037] Nonverbal information can refer to information unrelated to language. It can include visual, auditory, tactile, and olfactory information. Nonverbal information is conveyed through senses other than language and can primarily include body language, images, drawings, graphics, movements, and colors. For example, human gestures, photographs, videos, drawings, the positions of chess pieces, musical notes and rhythms, and the structure of a maze can all be considered nonverbal information.

[0038] Linguistic information types are important in communication, natural language processing, sentiment analysis, machine learning, computer vision, and various information processing tasks, and can be used to improve interaction and understanding between humans and machines.

[0039] According to one embodiment, a processor can train an artificial neural network (ANN) model with the aforementioned multi-modal personality data. The plurality of personality models trained with the multi-modal personality data may include personality models specialized for a single personality. For example, assume that the multi-modal personality data is labeled with the Big 5 (OCEAN) factor. The Big 5 (OCEAN) factor means that an individual's personality characteristics are explained in five major dimensions: Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. When the processor trains an artificial neural network model with personality data labeled with the Big 5 (OCEAN) factor, five personality models specialized for each personality characteristic can be obtained.

[0040] For example, if a user types "Please recommend some hobbies to do today" into a personality model specialized in extroversion, it can recommend activities such as exercise that can be done outside.

[0041] However, the personality model is not limited to the described examples, and may be acquired based on personality traits such as a calm personality, an impatient personality, a bright personality, and a cool personality that generally exist.

[0042] In step (120), a processor according to one embodiment can link multiple personality models to a pipeline-based artificial intelligence service.

[0043] A processor according to one embodiment may provide multiple personality models to a user interface of an artificial intelligence service.

[0044] An AI service according to one embodiment may be a service that generates an artificial neural network model suitable for the user based on user input. Furthermore, a pipeline-based AI service may refer to a service that provides an infrastructure for efficiently developing, deploying, and managing various AI technologies and services. Such AI services can help businesses, organizations, and developers develop diverse applications and services by leveraging AI technology.

[0045] The user interface can be provided in a block-based or chat-based format, making it easy for developers and users to use. For example, a pipeline-based AI service can be provided in the form of an artificial neural network model transformed into blocks (e.g., the pipeline (400) in FIG. 4). Multiple personality models are provided in the AI ​​service in block form, allowing users to easily select, insert, move, and delete them within the pipeline.

[0046] In step (130), a processor according to one embodiment may generate a user personality model by combining a plurality of personality models based on a user's input to an artificial intelligence service.

[0047] In one embodiment, a processor may determine priorities and degrees of reflection among multiple personality models based on user inputs for selection and prompt inputs on a user interface. The processor may adjust the priorities and degrees of reflection of a personality model or group of personality models in a pipeline.

[0048] For example, suppose a user wants to create an artificial neural network model for an AI service with extroverted and impatient personalities. The processor can combine the extroverted personality model and the impatient personality model to create a user personality model with the desired impatient personality.

[0049] In one embodiment, a processor may evaluate an intermediate user personality model that combines multiple personality models. The processor may repeatedly perform operations to combine multiple personality models by changing priorities and degrees of reflection among the multiple personality models until the user-inputted set performance is achieved.

[0050] For example, as described above, suppose a user wants to create an artificial neural network model with extroverted and impatient personalities, but with a focus on extroversion. In this case, the processor can combine the extroverted and impatient personality models in any ratio and generate test outputs. If the test outputs are not what the user wants or are evaluated as falling short of the set performance, the priority and degree of reflection of the extroverted and impatient personality models can be changed. A method for testing performance in this case could be to input that the test outputs are not what the user wants. Alternatively, the processor could perform self-supervised learning. Self-supervised learning is a subfield of machine learning and a form of supervised learning. This method involves training a model without data labels or external supervision provided by a human. Instead, the model can extract and learn features from the data itself.

[0051] In step (140), a processor according to one embodiment may update an artificial intelligence service based on a user personality model.

[0052] A user personality model according to one embodiment may be a model that preprocesses user input for a language model of an artificial intelligence service or postprocesses output of a language model of an artificial intelligence service.

[0053] A processor according to one embodiment may provide a user interface for deploying an artificial neural network model generated based on user input in an updated artificial intelligence service.

[0054] For example, an AI service with an updated personality model for an extroverted and open-minded user could deploy an artificial neural network model that produces outputs like the following: If a user inputs, "Recommend a place for a company dinner today," the output could be a place like, "A barbecue restaurant where you grill meat outdoors."

[0055]

[0056] FIG. 2 is a block diagram schematically illustrating a personality model service according to one embodiment.

[0057] The description referring to Fig. 1 can be equally applied to Fig. 2, and overlapping content can be omitted. One or more blocks and combinations of blocks of Fig. 2 can be implemented by a special-purpose hardware-based computer performing a specific function, or a combination of special-purpose hardware and computer instructions.

[0058] An artificial intelligence service platform (200) according to one embodiment may perform personality model learning (230) based on personality data (210) to provide a personality model service (250). The personality data (210) may refer to data related to personality characteristics such as images, voices, and texts. The artificial intelligence service platform (200) may select a model to be learned through automated learning model selection (231) in personality model learning, and may perform model learning (232) and model performance evaluation (233) based on the selected model to generate a user personality model. The artificial intelligence service platform (200) may update the artificial intelligence service based on the user personality model generated through multi-modal personality model learning (230) to provide a personality model service (250).

[0059]

[0060] Figure 3 schematically illustrates the use of a personality model in a pipeline format according to one embodiment.

[0061] The descriptions made with reference to FIGS. 1 and 2 may be equally applied to FIG. 3, and any overlapping content may be omitted. One or more blocks and combinations of blocks in FIG. 3 may be implemented by a special-purpose hardware-based computer performing a specific function, or a combination of special-purpose hardware and computer instructions.

[0062] According to one embodiment, a processor may select a target personality model to be combined from personality model A and personality model B based on user input. The processor may combine personality models A and personality model B based on priority and degree of reflection to generate an integrated user personality model C. The processor may update the personality model C in the artificial intelligence service (200). The processor may distribute an artificial neural network model reflecting personality in the artificial intelligence service (200) reflecting personality model C through a user interface. The user may obtain the artificial neural network model reflecting personality through the user interface.

[0063]

[0064] FIG. 4 is a diagram illustrating an example of a pipeline according to one embodiment.

[0065] The description referring to FIGS. 1 to 3 can be equally applied to FIG. 4, and overlapping content can be omitted.

[0066] Referring to FIG. 4, an artificial intelligence service platform (200) according to one embodiment can build an artificial neural network into a pipeline-based UI and provide it to a user.

[0067] According to one embodiment, the pipeline UI may include a resource preparation UI, a data cleansing UI, a model training UI, and a model deployment UI. However, Figure 3 is merely an example illustrating a pipeline-based UI and is not necessarily limited thereto.

[0068] The AI ​​service platform (200) can perform model combination and integration. The pipeline-based platform can integrate various AI models. These models perform various functions, including language processing, image analysis, speech recognition, prediction, and recommendation systems.

[0069] The AI ​​service platform (200) can perform personality combinations. For example, the AI ​​service platform (200) can help create customized services by integrating multiple personality models. Users can connect various personality models in a pipeline and effectively combine them.

[0070] The AI ​​service platform (200) can set priorities and the degree of reflection in the combination of personality models. Users can set the priority and degree of reflection for each model or group of models in the pipeline. For example, the AI ​​service platform (200) can control how certain personality models are executed before others or activated under certain conditions.

[0071] The AI ​​service platform (200) can perform data management. For example, the AI ​​service platform (200) can provide tools and interfaces for data management. Users can connect various data to the platform and feed it to models to utilize in services.

[0072] The AI ​​service platform (200) can support real-time batch processing. The AI ​​service platform (200) can mean that the service can generate responses to real-time data streams or batch-processed data.

[0073] The artificial intelligence service platform (200) can integrate models developed by users or add new functions.

[0074]

[0075] FIG. 5 is a block diagram illustrating an electronic device according to one embodiment.

[0076] One or more blocks and combinations of blocks of FIG. 5 may be implemented by a special-purpose hardware-based computer performing a specific function, or by a combination of special-purpose hardware and computer instructions. The descriptions made with reference to FIGS. 1 through 4 may be equally applicable to FIG. 5 . For example, an electronic device (500) according to one embodiment may include an artificial intelligence service platform (200).

[0077] As shown in FIG. 5, the electronic device (500) may include a memory (510) and a processor (520). The electronic device (500) may further include a communication module, and the communication module may include a transmitter and a receiver.

[0078] An electronic device (500) according to one embodiment may include a memory (510) and a processor (520) connected to the memory (510) via a system bus or other suitable circuitry.

[0079] The electronic device (500) may store program code in memory (510). In one embodiment, the memory (510) may include one or more physical memory devices, such as local memory or one or more bulk storage devices. In this case, the local memory may include random access memory (RAM) or other volatile memory devices commonly used while actually executing the program code. The bulk storage device may be implemented as a hard disk drive (HDD), a solid state drive (SSD), or other non-volatile memory device.

[0080] As the executable program code stored in the memory (510) is executed by the electronic device (500), the processor (520) may perform various operations described in the present disclosure. For example, the memory (510) may store program code for causing the processor (520) to perform one or more operations described in FIGS. 1 to 4.

[0081] Depending on the specific type of device being implemented, the electronic device (500) may include fewer components than those illustrated or additional components not illustrated in FIG. 5. Additionally, one or more of the components may be incorporated into, or otherwise form part of, another component.

[0082] A processor (520) according to one embodiment is a hardware configuration that performs overall control functions for controlling the operations of an electronic device (500). For example, the processor (520) may control the electronic device (500) overall by executing programs stored in a memory (510) within the electronic device (500). The processor (520) may be implemented as a central processing unit (CPU), a graphics processing unit (GPU), an application processor (AP), a neural processing unit (NPU), or the like, provided within the electronic device (500), but is not limited thereto.

[0083] The processor (520) learns multi-modal personality data in an artificial neural network model, acquires multiple personality models, links the multiple personality models to a pipeline-based artificial intelligence service, combines the multiple personality models based on a user's input to the artificial intelligence service to create a user personality model, and updates the artificial intelligence service based on the user personality model.

[0084]

[0085] The embodiments described above may be implemented using hardware components, software components, and / or a combination of hardware components and software components. For example, the devices, methods, and components described in the embodiments may be implemented using a general-purpose computer or a special-purpose computer, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing instructions and responding to them. The processing device may execute an operating system (OS) and software applications running on the operating system. Furthermore, the processing device may access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processing device is sometimes described as being used alone; however, one of ordinary skill in the art will recognize that the processing device may include multiple processing elements and / or multiple types of processing elements. For example, a processing unit may include multiple processors, or a processor and a controller. Other processing configurations, such as parallel processors, are also possible.

[0086] Software may include a computer program, code, instructions, or a combination of one or more of these, which may configure a processing device to perform a desired operation or may, independently or collectively, command the processing device. The software and / or data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage medium or device, or transmitted signal wave, for interpretation by the processing device or for providing instructions or data to the processing device. The software may also be distributed over networked computer systems and stored or executed in a distributed manner. The software and data may be stored on a computer-readable recording medium.

[0087] The method according to the embodiment may be implemented in the form of program commands that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program commands, data files, data structures, etc., alone or in combination, and the program commands recorded on the medium may be those specially designed and configured for the embodiment or may be known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute program commands such as ROMs, RAMs, and flash memories. Examples of program commands include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter, etc.

[0088] The hardware device described above may be configured to operate as one or more software modules to perform the operations of the embodiment, and vice versa.

[0089] Although the embodiments described above have been described with limited drawings, those skilled in the art will appreciate that various technical modifications and variations can be applied based on the described embodiments. For example, appropriate results can still be achieved even if the described techniques are performed in a different order than described, and / or components of the described systems, structures, devices, circuits, etc. are combined or combined in a different manner than described, or are replaced or substituted with other components or equivalents.

[0090] Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the claims described below.

Claims

1. A step of acquiring multiple personality models by learning multi-modal personality data in an artificial neural network model; A step of linking the above multiple personality models to a pipeline-based artificial intelligence service; A step of generating a user personality model by combining the plurality of personality models based on the user's input for the artificial intelligence service; and A step of updating the artificial intelligence service based on the above user personality model. A method of operating an artificial intelligence platform, comprising:

2. In paragraph 1, The above multi-modal personality data A method of operating an artificial intelligence platform, comprising at least one of linguistic, paralinguistic and non-linguistic information.

3. In paragraph 1, Each of the above multiple personality models An operation method of an artificial intelligence platform that includes personality models specialized for a single personality, trained with personality data labeled with the Big 5 (OCEAN) factors.

4. In paragraph 1, The above linking steps are A step of providing the above multiple personality models to the user interface of the above artificial intelligence service. A method of operating an artificial intelligence platform, comprising:

5. In paragraph 4, The steps for creating the above user personality model are: A step of determining the priority and degree of reflection among the plurality of personality models based on the user's selection input and the user's prompt input for the user interface. A method of operating an artificial intelligence platform, comprising:

6. In paragraph 5, The steps for creating the above user personality model are: A step of evaluating an intermediate user personality model that combines the above multiple personality models; and A step of repeatedly performing an operation of combining the plurality of personality models by changing the priority and reflection level among the plurality of personality models until the user-input setting performance is achieved. A method of operating an artificial intelligence platform, comprising:

7. In paragraph 1, The above artificial intelligence service A method of operating an artificial intelligence platform, which generates an artificial neural network model suitable for the user based on the input of the user.

8. In paragraph 1, The above user personality model is An operation method of an artificial intelligence platform for preprocessing user input for a language model of the artificial intelligence service or postprocessing output of a language model of the artificial intelligence service.

9. In paragraph 1, A step for providing a user interface for distributing an artificial neural network model generated according to the user's input in the above artificial intelligence service where the update is completed. A method of operating an artificial intelligence platform, further comprising:

10. A computer program stored on a computer-readable recording medium to execute the first clause by being combined with hardware.

11. Memory containing instructions; By learning multi-modal personality data in an artificial neural network model, multiple personality models are obtained. Link the above multiple personality models to a pipeline-based artificial intelligence service, Based on the user's input to the above artificial intelligence service, a user personality model is created by combining the above multiple personality models, A processor that updates the artificial intelligence service based on the above user personality model. An electronic device comprising:

Images