Consumer inner disposition typing analysis model production system using generative ai and method thereof

Abstract

The present invention relates to a system for producing a consumer inner disposition typing analysis model by using generative AI. The present invention may allow companies to customize production to be suitable for various customers by generating a final typing module including a question and answer set for consumer type classification according to disposition topics using generative AI and allow customer companies to establish customized customer services and customized marketing strategies for each consumer type and provide product recommendation and personalization services through a consumer typing module.

Description

A system and method for creating a consumer tendency typology analysis model using generative AI.

[0001] The present invention relates to a system for producing a consumer internal tendency typology analysis model using generative AI and a method therefor, and more particularly, to a system for producing a consumer internal tendency typology analysis model using generative AI and a method therefor, which can automatically produce a customized consumer typology module that analyzes and typifies the internal tendencies of consumers by inputting only topics that are suitable for the consumers of a company.

[0002] To differentiate their services and secure a competitive edge in the market, companies want to accurately understand consumer needs and develop customized strategies. Especially in the rapidly changing and competitive digital age, meeting diverse consumer needs and expectations is essential. Conventional big data analytics are useful for identifying general consumer trends and patterns from vast amounts of data. While this allows companies to develop large-scale marketing strategies, it limits their ability to deeply understand individual consumer tendencies and psychological characteristics. Because big data primarily focuses on identifying characteristics of large groups through statistical correlations, it lacks a personalized approach.

[0003] Small data is more useful for understanding and identifying individuals' inner tendencies, preferences, preferences, and psychological characteristics. Small data focuses on understanding consumers' personalities by providing data from a human literacy perspective. For example, while big data can tell us that "many people buy a particular product," small data can help us understand "why a particular consumer buys that product." This is useful for deriving individual characteristics through causal relationships in data. Small data can be used to analyze consumers' inner tendencies and identify personalized needs.

[0004] The technology for creating a consumer internal tendency typology model is an effective tool for analyzing and typifying consumers' internal tendencies, but it requires specialized personnel and has the following problems.

[0005] First, small-data models are vulnerable to data shortages. In particular, data related to extracting consumer tendencies is often difficult to collect and insufficient. Data shortages can lead to poor model learning performance and reduced classification accuracy.

[0006] Moreover, the model training and improvement process is complex and time-consuming. Small-data model training requires significant time and effort, and the model improvement process also requires iterative work. Difficulties in model training and improvement can lead to increased costs and reduced development efficiency.

[0007] Furthermore, developing customized data models based on consumer small data requires the intervention of small data experts. Because consumer tendencies vary from person to person, existing small data models may not adequately reflect individual differences. This leads to limitations in marketing and customer experience delivery.

[0008] Patents relating to consumer typology include:

[0009] Patent No. 10-1355832 is an invention regarding an automatic creation database of customer information in which, when an advertiser inputs customer purchase information into the system online, the system's consumer information classification program automatically classifies customers based on data such as information on the day the customer made a purchase or visited, purchase motivation, and purchase pattern.

[0010] The consumer information classification program is a program that sets items that are thought to influence purchases, such as weather, day of the week, anniversary, purchase motivation, and repurchase period, and is a customized direct advertising system that customizes and classifies each customer and sends advertisements only to the customer in cases where a similar situation occurs, such as the purchase motivation and pattern of a specific customer.

[0011] Publication No. 10-2022-0137233 relates to a product recommendation system based on personal information, which performs statistical analysis using the personal information and purchase history of multiple customers to create various types of customer groups and product groups belonging to the customer groups, and when a product recommendation request is received from a user terminal device, obtains a customer group matching the user's personal information of the user terminal device and outputs recommended product information. Publication No. 10-2022-0137233 classifies and accumulates the personal information and purchase history of multiple customers according to various criteria, and recommends products to customers with similar personal information, thereby providing product information suitable for the customer's lifestyle.

[0012] Patent No. 10-1355832 does not categorize consumers based on their inner tendencies, but rather categorizes them based on factors such as purchase season, purchase date, day of the week, purchase week, whether or not they purchased after an advertisement, repurchase period, and purchased products, making it difficult to accurately analyze consumers' needs.

[0013] Publication Patent No. 10-2022-0137233 recommends products based on personal information, including age, gender, occupation, hobbies, and marital status. While personal information-based customer analysis is suitable for big data analysis, it suffers from inaccuracy in small data analysis.

[0014] Therefore, there is an urgent need to introduce a system that can analyze the internal tendencies of customers' consumers regardless of the customer's data shortage, minimize the customer's cost burden by investing less time in creating a consumer internal tendencies typology analysis model, and produce customized models for various customer topics.

[0015] The present invention was invented to improve the above problems, and provides a system for creating a consumer internal tendency typology analysis model using generative AI that can create a consumer internal tendency typology analysis model tailored to a client company by simply inputting the subject of the typology model tailored to the client company's consumers.

[0016] The present invention aims to provide a system for creating a consumer internal tendency typology analysis model using generative AI that can accurately classify the types of consumers of a client company by classifying the client company's consumers into 256 types based on four tendencies: relationship attitude, way of thinking, decision-making, and lifestyle.

[0017] The present invention aims to provide a system for producing a consumer internal tendency typology analysis model using generative AI that can be customized to suit various consumers of a company by producing a final typology module consisting of a set of questions and answers for classifying consumer types according to tendency topics using generative AI.

[0018] The present invention was invented to improve the above problems, and provides a system for creating a consumer internal tendency typology analysis model using generative AI that can create a consumer internal tendency typology analysis model tailored to a client company by simply inputting the subject of the typology model tailored to the client company's consumers.

[0019] The present invention aims to provide a system for creating a consumer internal tendency typology analysis model using generative AI that can accurately classify the types of consumers of a client company by classifying the client company's consumers into 256 types based on four tendencies: relationship attitude, way of thinking, decision-making, and lifestyle.

[0020] The present invention aims to provide a system for producing a consumer internal tendency typology analysis model using generative AI that can be customized to suit various consumers of a company by producing a final typology module consisting of a set of questions and answers for classifying consumer types according to tendency topics using generative AI.

[0021] According to the present invention having the above configuration, the following effects are achieved.

[0022] By simply entering the topic of the typology model that suits the client's consumers, the prompts are optimized, data is automatically collected through the optimized prompts, and the collected data is combined with the typology model that classifies it into 256 categories based on the four tendencies of relationship attitude, thinking style, decision-making, and lifestyle, and applied to generative AI, a customized consumer typology module that suits the client can be automatically and quickly created.

[0023] The customized consumer typology module ultimately provided to clients allows for a more accurate and in-depth analysis of an individual's inner tendencies and allows for a total of 256 typologies based on consumer characteristics, enabling customized services for each customer type.

[0024] Companies can better understand their customers through consumer propensity analysis and categorization based on small data. Based on this analysis, they can improve services and products, establish and implement customized marketing strategies for each consumer type, and thereby increase customer loyalty.

[0025] Additionally, companies can reduce costs by leveraging small-scale data rather than big data.

[0026] Figure 1 is a functional block diagram of a system for producing a consumer internal tendency typology analysis model using generative AI according to the present invention.

[0027] Figure 2 is a detailed functional block diagram of the basic typing module.

[0028] Figure 3 is a structural diagram of the first and second indicators for the four tendencies.

[0029] Figure 4 is a conceptual diagram of a typology module algorithm that classifies consumers into a total of 256 types based on four tendencies.

[0030] Figure 5 is a flowchart of a method for creating a consumer internal tendency typology analysis model using generative AI.

[0031] In order to achieve the above purpose, the system for producing a consumer internal tendency typology analysis model using generative AI according to the present invention comprises: a consumer typology topic input unit for inputting a topic of a typology model suitable for a client company's consumer; a small data combination unit for collecting and preprocessing data corresponding to the topic of the input typology model, combining the data with a typology module that classifies the consumer's internal tendency for the topic of the typology model into types based on four propensity topics of relationship attitude, way of thinking, decision-making, and attitude toward life, and preparing to input the data into the generative AI; a type data generation unit for applying the data received from the small data combination unit to the generative AI to create and verify a set of text and images for consumer type classification; and a customized consumer typology module generation unit for generating a final typology module consisting of a set of questions and answers that classifies the consumer's internal tendency corresponding to the input topic into four propensity topics of relationship attitude, way of thinking, decision-making, and attitude toward life using the verified set of text and images.

[0032] The above small data binding unit is,

[0033] It includes a prompt optimization module that generates prompts related to four tendencies themes of relationship attitude, way of thinking, decision-making, and lifestyle that can identify the inner tendencies of consumers for an input topic, a data preprocessing module that collects data using the prompts generated by the prompt optimization module and tags and labels the collected data, a basic typing module that combines the tagged and labeled data with a typing module that classifies the inner tendencies of consumers into a total of 256 types based on the four tendencies themes of relationship attitude, way of thinking, decision-making, and lifestyle, and an encoding / token conversion module that handles LLM encoding and token conversion for communication with the generative AI.

[0034] In addition, the above prompt optimization module can apply a method of analyzing the topics and prompt patterns of an existing generated prompt database using an artificial intelligence neural network-based algorithm and automatically generating a prompt that matches the input topic.

[0035] The above basic typing module is,

[0036] It is characterized by classifying the tagged and labeled data into up to 16 types based on the primary indicators based on four tendencies: relationship attitude, way of thinking, decision-making, and lifestyle, and then classifying them into up to 16 types based on the secondary indicators, thereby classifying the consumer group into a total of 256 types.

[0037] Specifically, the basic typing module,

[0038] It can be configured to include a first indicator module that classifies consumer tendencies into two first indicators for each of the four tendency topics, a second indicator module that classifies the first indicator into two second indicators according to the consumer tendencies, a first indicator determination module that inputs data tagged and labeled by the data preprocessing module into the first indicator module to determine a first indicator for each of the four tendency topics, a second indicator determination module that inputs data tagged and labeled by the data preprocessing module into the second indicator module to determine a second indicator for each of the four tendency topics, and a basic typification result output module that reflects the determinations of the first and second indicators and outputs a basic typification result.

[0039] The above basic typing module is,

[0040] The present invention further includes a first indicator learning module that trains a first indicator for each of four tendency topics by applying tagged and labeled training data to a supervised learning-based machine learning engine, and a second indicator learning module that trains a second indicator for each of four tendency topics by applying the tagged and labeled training data and the first indicator to the supervised learning-based machine learning engine, wherein the first indicator determination module determines a first indicator for each of four tendency topics by applying the data tagged and labeled by the data preprocessing module to the first indicator learning module, and the second indicator determination module determines a second indicator for each of four tendency topics by applying the data tagged and labeled by the data preprocessing module and the first indicator determined by the first indicator determination module to the second indicator learning module.

[0041] The above type data generation unit,

[0042] It is composed of a generative AI application module that expands the dataset of the typification module by additionally generating or transforming text and images that classify the consumer's inner tendencies into a total of 256 types based on four tendencies themes of relationship attitude, thinking style, decision-making, and lifestyle by applying the data received from the small data combination unit to the generative AI, a natural language processing module that removes noise from the text generated through the generative AI application module, an image classification module that classifies appropriate images among the images generated through the generative AI application module, and a data verification module that verifies the text with noise removed and the classified images.

[0043] The above generative AI application module is,

[0044] It is characterized by applying the data received from the small data combination unit to the generative AI to create multiple questions to determine the primary indicators and multiple answers matching each primary indicator for each of the four tendencies of relationship attitude, way of thinking, decision-making, and lifestyle, and multiple questions to determine the secondary indicators and multiple answers matching each secondary indicator.

[0045] The above customized consumer typology module generation unit is:

[0046] The verified data received from the type data generation department is used to create one question and a matching answer for each of the four tendencies of relationship attitude, way of thinking, decision-making, and lifestyle to determine the primary indicator, and multiple questions and a matching answer for each of the secondary indicators to determine the secondary indicators. This creates a final typology module that classifies consumers into a total of 256 types and provides it to the client.

[0047] Clients are provided with a final typology module and a survey UI to understand consumers' inner tendencies.

[0048]

[0049] Meanwhile, in order to achieve the above purpose, the method for producing a consumer internal tendency typology analysis model using generative AI according to the present invention comprises the following steps: a first step of inputting a topic of a typology model suitable for a client company's consumer; a second step of generating a prompt related to four tendency topics of relationship attitude, way of thinking, decision-making, and lifestyle that can identify the consumer's internal tendency based on the input topic; a third step of collecting data using the prompt generated in the prompt optimization module and tagging and labeling the collected data; a fourth step of classifying the tagged and labeled data into up to 16 types based on the four tendency topics of relationship attitude, way of thinking, decision-making, and lifestyle, and further classifying the consumer group into a total of 256 types based on the first indicator; a fifth step of processing LLM encoding and token conversion for communication with the generative AI; and a fifth step of applying the received token to the generative AI to classify text and images into a total of 256 types based on the four tendency topics of relationship attitude, way of thinking, decision-making, and lifestyle. It consists of a 6th step of expanding the dataset of the typology module by generating or transforming it, a 7th step of removing noise from text generated through generative AI, an 8th step of classifying suitable images from among the images generated through generative AI, a 9th step of verifying the text with the noise removed and the classified images, and a 10th step of creating a final typology module that classifies consumers into a total of 256 types by generating one question and a matched answer for each of the four tendencies of relationship attitude, thinking style, decision-making, and lifestyle for each of the verified data, and multiple questions and a matched answer for each of the second indicators.

[0050] Specifically, the fourth step is:

[0051] It can be configured in such a way that the tagged and labeled data is applied to the primary indicator learning module to determine the primary indicator for each of the four tendency topics, and the tagged and labeled data and the primary indicator determined by the primary indicator determination module are applied to the secondary indicator learning module to determine the secondary indicator for each of the four tendency topics.

[0052] The advantages and features of the present invention and the method for achieving them will become clear with reference to the embodiments described in detail below together with the attached drawings.

[0053] However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms.

[0054] The embodiments in this specification are provided to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention.

[0055] And the present invention is defined only by the scope of the claims.

[0056] Accordingly, in some embodiments, well-known components, well-known operations, and well-known techniques are not specifically described to avoid obscuring the present invention.

[0057] Additionally, like reference numerals throughout the specification refer to like elements, and the terms used (referred to) herein are for the purpose of describing embodiments and are not intended to limit the present invention.

[0058] In this specification, the singular includes the plural unless specifically stated otherwise in the phrase, and the reference to an element or action as “including (or comprising)” does not exclude the presence or addition of one or more other elements or actions.

[0059] Unless otherwise defined, all terms (including technical and scientific terms) used herein may be used in the sense commonly understood by a person of ordinary skill in the art to which the present invention belongs.

[0060] Also, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless they are defined otherwise.

[0061] Hereinafter, a preferred embodiment of the present invention will be described with reference to the attached drawings.

[0062] The present invention will be described in detail with reference to FIGS. 1 to 5.

[0063] [A system for creating a consumer personality type analysis model using generative AI]

[0064] A system for producing a consumer internal tendency typology analysis model using generative AI according to the present invention comprises a consumer typology topic input unit (100), a small data combination unit (200), a type data generation unit (300), and a customized consumer typology module generation unit (400).

[0065] The customized consumer typology module creation unit (400) creates a final typology module (500) that classifies the client's consumers into a total of 256 types, and creates a customer survey (text, image) applied to the final typology module (500) and provides it to the client.

[0066] The consumer typology topic input section (100) inputs topics of a typology model suited to the client's consumers. For example, topics such as interior design, investment, finance, consumption, learning, health, and food are input.

[0067] The small data combination unit (200) collects and preprocesses data corresponding to the subject of the input typology model, and prepares to input it into the generative AI by combining it with a typology module that classifies the consumer's inner tendencies for the subject of the typology model into types based on four tendencies: relationship attitude, way of thinking, decision-making, and lifestyle.

[0068] The small data combination unit (200) is configured to include a prompt optimization module (210), a data preprocessing module (220), a basic typing module (230), and an encoding / token conversion module (240).

[0069] The prompt optimization module (210) generates prompts related to four tendencies topics of relationship attitude, thinking style, decision-making, and lifestyle that can identify the inner tendencies of consumers for the input topic.

[0070] For example, the prompt optimization module (210) can generate a prompt such as, "I want to categorize consumers' internal tendencies toward interior design based on four tendencies: relationship attitude, way of thinking, decision-making, and lifestyle. Please investigate the results according to various tendencies and tastes toward interior design."

[0071] The prompt optimization module (210) can analyze the topics and prompt patterns of an existing generated prompt database (not shown) using an artificial intelligence neural network-based algorithm and automatically generate a prompt that matches the input topic.

[0072] The prompt database contains a variety of prompts covering a wide range of topics and subjects, including interior design, investing, finance, consumption, learning, health, and food.

[0073] The data preprocessing module (220) collects data in real time using various websites, databases, APIs, etc. using the prompts generated by the prompt optimization module (210), and tags and labels the collected data.

[0074] An example of data collected by the data preprocessing module (220) is as follows.

[0075] In the case of the relationship attitude tendency topic,

[0076] Independent: They value personal space and prefer comfortable and practical interiors. They utilize simple, functional furniture and accessories to create a warm and comfortable atmosphere.

[0077] Social: They value spaces where people interact and prefer flashy, trendy interiors. They utilize a variety of colors and patterns, and use accessories that express their individuality.

[0078] Collaborative: They value harmonious spaces and prefer soft, comfortable interiors. They utilize natural materials and warm colors to create a comfortable and cozy atmosphere.

[0079] In the case of the topic of thinking style and tendencies,

[0080] Logical: You value clear and rational spaces and prefer simple, modern interiors. Using straight lines and geometric shapes, you create a clean, sophisticated atmosphere.

[0081] Sensitive: They value sensual and beautiful spaces and prefer elegant and sensual interiors. They utilize soft colors and natural materials to create emotional and atmospheric spaces.

[0082] Creative: They value original and unique spaces and prefer unique and artistic interiors. They utilize a variety of colors and patterns, and use accessories that express their individuality.

[0083] Users can label collected data or use a trained tagging model leveraging existing labeled data. Furthermore, errors, missing values, and unnecessary information can be removed from the collected data.

[0084] The basic typology module (230) combines the tagged and labeled data with a typology module that classifies the consumer's inner tendencies into a total of 256 types based on four tendencies themes: relationship attitude, way of thinking, decision-making, and lifestyle.

[0085] Referring to Figure 3, the meanings of the primary and secondary indicators for the four propensity themes are illustrated. Each of the four propensity themes—relationship attitudes, thinking styles, decision-making, and lifestyle—contains two primary indicators. Once the primary indicator corresponding to the consumer's propensity is determined for each of the four propensity themes, the final result indicator is output by selecting the corresponding secondary indicator among the two primary indicators.

[0086] When the final result indicators of Cg in relationship attitude, Id in thinking style, Lp in decision-making, and Pg in lifestyle are output, the final typology result of the consumer is CgIdLpPg.

[0087] The CgIdLpPg consumer type is described below.

[0088] Cg: A tendency to be influenced by others and to want to help others.

[0089] Id: A tendency to be imaginative and think clearly and concisely.

[0090] Lp: A tendency to make rational decisions and prioritize the process over the result.

[0091] Pg: A person who values ​​principles and likes to socialize.

[0092]

[0093] Referring to FIG. 4, the basic typing module (230) prepares to input the data collected, tagged, and labeled by the data preprocessing module (220) into a generative AI model by combining it with a typing module algorithm that classifies consumers into a total of 256 types based on four propensity topics.

[0094] The data collected, tagged, and labeled by the data preprocessing module (220) may be insufficient to classify consumers into a total of 256 types based on four tendency topics. With this as a reference, the dataset is expanded so that the generative AI can classify consumers into a total of 256 types.

[0095] The basic typology module (230) is a typology module that classifies the consumer's inner tendencies into a total of 256 types based on four tendencies themes of relationship attitude, way of thinking, decision-making, and lifestyle, and prepares to input the data collected by the data preprocessing module (220) into a generative AI model by matching the tendencies themes used in the typology module, the meaning of the primary indicators, the meaning of the secondary indicators, and the primary and secondary indicators by tendencies theme.

[0096] The typology module, which classifies a total of 256 types based on four tendencies, includes logic that determines the primary indicator through questions for each tendency topic and determines the secondary indicator belonging to the primary indicator through questions.

[0097] The encoding / token conversion module (240) processes LLM encoding and token conversion for communication with the generative AI. LLM encoding is a necessary step for the generative AI model to effectively process text data, and token conversion is the process of breaking down text data into tokens for individual words or sentences.

[0098] Referring to FIG. 3, the basic typology module (230) classifies the tagged and labeled data into up to 16 types based on the four tendencies of relationship attitude, way of thinking, decision-making, and lifestyle, and then classifies them into up to 16 types based on the secondary indicators, thereby classifying the consumer group into a total of 256 types.

[0099] If you multiply the first indicator by two relationship attitudes, two ways of thinking, two decision-making, and two lifestyle attitudes, you get a total of 16 cases. If you multiply it by the 16 second indicators, you can classify consumer groups into a total of 256 types.

[0100]

[0101] The basic typing module (230) is configured to include a primary indicator module (231), a secondary indicator module (232), a primary indicator determination module (233), a secondary indicator determination module (234), and a basic typing result output module (235).

[0102] The primary indicator module (231) categorizes consumer tendencies into two primary indicators for each of the four tendencies.

[0103] The secondary indicator module (232) classifies the primary indicator into two types of secondary indicators according to consumer tendencies.

[0104] The primary indicator determination module (233) inputs the data tagged and labeled by the data preprocessing module (220) into the primary indicator module (231) to determine the primary indicator for each of the four tendency topics.

[0105] The secondary indicator determination module (234) inputs the data tagged and labeled by the data preprocessing module (220) into the secondary indicator module (232) to determine the secondary indicator for each of the four tendency topics.

[0106] The basic typology result output module (235) outputs the basic typology result by reflecting the decisions regarding the primary and secondary indicators. Based on the prompt, the basic typology task is performed by determining the primary and secondary indicators for the collected, tagged, and labeled data.

[0107]

[0108] The basic typology module (230) may include a primary indicator learning module (236) and a secondary indicator learning module (237).

[0109] The primary indicator learning module (236) applies the tagged and labeled training data to a supervised learning-based machine learning engine to train the primary indicator for each of the four tendency topics.

[0110] The secondary indicator learning module (237) applies the tagged and labeled training data and the primary indicator to a supervised learning-based machine learning engine to train the secondary indicator for each of the four tendency topics.

[0111] The primary indicator determination module (233) applies the data tagged and labeled by the data preprocessing module (220) to the primary indicator learning module (236) to determine the primary indicator for each of the four tendency topics.

[0112] The secondary indicator determination module (234) applies the data tagged and labeled by the data preprocessing module (220) and the primary indicator determined by the primary indicator determination module (233) to the secondary indicator learning module (237) to determine the secondary indicator for each of the four tendency topics.

[0113] For example, if the data tagged and labeled by the data preprocessing module (220) is 'I value personal space, prefer comfortable and practical interiors, use simple and functional furniture and props, and create a warm and comfortable atmosphere,' the primary indicator determination module (233) can be applied to the primary indicator learning module (236) to determine the primary indicator as A for the relationship attitude tendency subject, and the secondary indicator determination module (234) can be applied to the secondary indicator learning module (237) to determine the secondary indicator as s for the relationship attitude tendency subject.

[0114]

[0115] The type data generation unit (300) applies the data received from the small data combination unit (200) to generative AI to generate and verify a set of text and images for consumer type classification. Generative AI refers to a large language model (LLM) such as ChatGPT. The text and image set for consumer type classification refers to the content included in a survey that classifies the client's consumers into 256 types.

[0116] The type data generation unit (300) is configured to include a generative AI application module (310), a natural language processing module (320), an image classification module (330), and a data verification module (34).

[0117] The generative AI application module (310) applies the data received from the small data combination unit (200) to the generative AI to additionally generate or modify text and images that classify the consumer's inner tendencies into a total of 256 types based on four tendencies themes of relationship attitude, thinking style, decision-making, and lifestyle, thereby expanding the dataset of the typification module.

[0118] The generated text and images are then validated using accuracy and F1 scores. Accuracy evaluates how well the generated text and images match the actual data, while the F1 score assesses the quality of the generated data using metrics that consider precision and recall.

[0119] Relationship attitudes ask about a person's tendencies toward forming and maintaining relationships with others. Thinking style questions problem-solving and decision-making methods. Decision-making questions the process of gathering, analyzing, judging, and selecting information. Lifestyle questions the person's values ​​and behavioral patterns displayed in daily life.

[0120] For each tendency topic, the tendency is divided into two through the first indicator, and then divided into two again through the second indicator.

[0121] The generative AI application module (310) applies the data received from the small data combination unit (200) to the generative AI to generate multiple questions for determining the primary indicators and multiple answers matching each primary indicator for each of the four tendencies of relationship attitude, thinking style, decision-making, and lifestyle, and generate multiple questions for determining the secondary indicators and multiple answers matching each secondary indicator.

[0122] Referring to Figure 4, there is one question (Q1) to determine the primary indicator in the final consumer survey, and three questions (Q2-1 to Q2-3) to determine the secondary indicator. The generative AI application module (310) generates multiple questions for determining the primary indicator and multiple answers matching each primary indicator, and multiple questions for determining the secondary indicator and multiple answers matching each secondary indicator, and then selects an appropriate one among them.

[0123] The natural language processing module (320) removes noise from text generated by the generative AI application module (310). The natural language processing module (320) removes unnecessary spaces, special characters, spacing errors, etc. from the text data, and consistently converts expressions within the text data.

[0124] The image classification module (330) classifies suitable images among the images generated by the generative AI application module. The image classification module (330) removes images with a suitability of 80% or less. A suitability of 80% is an example.

[0125] The data verification module (340) verifies denoised text and classified images. The data verification module (340) checks for missing values ​​in text generated through generative AI and supplements missing information. It also checks for and corrects contextual errors in the text data.

[0126] The data verification module (340) verifies the suitability of images generated through generative AI to the standards set, evaluates the quality of the images, such as resolution, clarity, and noise, and checks how closely the image data is related to the text data, and removes images whose suitability falls below the standard.

[0127]

[0128] The customized consumer typology module generation unit (400) generates a final typology module (500) consisting of a set of questions and answers that classify the inner tendencies of consumers corresponding to the input topic into four tendencies themes of relationship attitude, thinking style, decision-making, and lifestyle, using a set of verified texts and images.

[0129] The customized consumer typology module generation unit (400) generates a final typology module (500) that classifies consumers into a total of 256 types by generating one question and a matching answer for each of the four tendencies of relationship attitude, thinking style, decision-making, and lifestyle based on the verified data received from the typology data generation unit (300), and generating multiple questions and a matching answer for each of the second indicators to determine the second indicator. The final survey dataset (text, image) generated through the final typology module (500) is output to the client's user UI.

[0130]

[0131] [How to Create a Consumer Inner Tendency Typing Analysis Model Using Generative AI]

[0132] Referring to Figure 5, we describe a method for creating a consumer propensity typology analysis model using generative AI. Detailed explanations of areas that overlap with the above description are omitted.

[0133] Step 1 (S510): Enter the subject of the typology model that suits the customer's consumers.

[0134] Step 2 (S520): Generate prompts related to four tendencies themes: relationship attitude, thinking style, decision-making, and lifestyle, which can help understand the consumer's inner tendencies regarding the input topic.

[0135] Step 3 (S530): Collect data using the prompt generated by the prompt optimization module, and tag and label the collected data.

[0136] Step 4 (S540): Tagged and labeled data are classified into up to 16 types based on the four tendencies of relationship attitude, thinking style, decision-making, and lifestyle, and then further classified into up to 16 types based on the secondary indicators, thereby fundamentally classifying consumer groups into a total of 256 types.

[0137] Step 5 (S550): Process LLM encoding and token conversion for communication with generative AI.

[0138] Step 6 (S560): Using the received tokens, apply them to the generative AI to expand the dataset of the typology module by additionally generating or modifying text and images that are classified into 256 types based on the four tendencies of relationship attitude, thinking style, decision-making, and lifestyle.

[0139] Step 7 (S570): Remove noise from text generated through generative AI.

[0140] Step 8 (S580): Classify suitable images among the images generated through generative AI.

[0141] Step 9 (S590): Verify the text with noise removed and the classified image.

[0142] Step 10 (S600): The verified data is used to create a final typology module that classifies consumers into a total of 256 types by generating one question and matching answers for each of the four tendencies of relationship attitude, way of thinking, decision-making, and lifestyle, and multiple questions and matching answers for each of the four tendencies of relationship attitude, way of thinking, decision-making, and lifestyle, to determine the primary indicators.

[0143]

[0144] Step 4 can be structured as follows:

[0145] The tagged and labeled data is applied to the primary indicator learning module to determine the primary indicator for each of the four tendency topics.

[0146] Then, the tagged and labeled data and the primary indicators determined by the primary indicator determination module are applied to the secondary indicator learning module to determine the secondary indicators for each of the four tendency topics.

[0147]

[0148] To summarize the above, by incorporating generative AI into a typology model capable of analyzing consumer behavior data and internal tendencies, it can be quickly adopted in various industries, including the following.

[0149] [E-commerce / Travel]

[0150] 1. Customized product and content recommendations

[0151] 2. Personalized product and content recommendations based on past purchase history, search history, browsing behavior, and personal preferences.

[0152] 3. Segment customers based on gender, age, income, lifestyle, etc. and develop customized marketing strategies for each segment.

[0153] [Medical and Learning Services]

[0154] 1. Providing customized health management programs tailored to individual preferences by combining personal health data.

[0155] 2. Provide personalized learning content by identifying the learner's learning style and level.

[0156] 3. Provide customized job recommendations and career exploration solutions by understanding the student's or user's inner tendencies.

[0157] [Human Resource Management]

[0158] 1. Identify the capabilities and tendencies of the candidates and select suitable talent.

[0159] 2. Identify employees' strengths and weaknesses and provide customized training and programs.

[0160] [Financial Services]

[0161] 1. Customized financial product recommendations based on individual financial circumstances and investment tendencies.

[0162] [Government and Non-profit Organizations]

[0163] 1. Establish customized policies and provide social services through type classification.

[0164]

[0165] According to the present invention having the above configuration, by inputting only the subject of the typology model that suits the consumer of the client company, the prompt is optimized, data is automatically collected through the optimized prompt, and the collected data is combined with the typology model that classifies the data into a total of 256 types based on the four tendencies of relationship attitude, thinking style, decision-making, and lifestyle, and applied to the generative AI, thereby automatically and quickly creating a customized consumer typology module that suits the client company.

[0166] Although the present invention has been described in detail with reference to preferred embodiments, it should be understood that the embodiments described above are exemplary in all respects and not restrictive, as those skilled in the art can implement the present invention in other specific forms without changing the technical idea or essential characteristics thereof.

[0167] In addition, the scope of the present invention is defined by the claims described below rather than the detailed description above, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention.

Claims

1. Consumer typology topic input section for entering the topic of the typology model that suits the customer's consumers; A small data combination unit that collects and preprocesses data corresponding to the subject of the input typology model, and prepares for input into the generative AI by combining it with a typology module that classifies the consumer's inner tendencies for the subject of the typology model into types based on four tendencies: relationship attitude, way of thinking, decision-making, and lifestyle. A type data generation unit that applies data received from the small data combination unit to the generative AI to generate and verify a set of text and images for consumer type classification; and A system for producing a consumer internal tendency typology analysis model using generative AI, characterized by including a customized consumer typology module generation unit that generates a final typology module consisting of a set of questions and answers that classify the internal tendencies of consumers corresponding to an input topic into four tendencies themes of relationship attitude, thinking style, decision-making, and lifestyle, using a set of verified texts and images.

2. In claim 1, The above small data binding unit is, A prompt optimization module that generates prompts related to four tendencies: relationship attitude, thinking style, decision-making, and lifestyle, which can identify the inner tendencies of consumers for the input topic; A data preprocessing module that collects data using prompts generated from the prompt optimization module and performs tagging and labeling on the collected data; A basic typology module that combines the tagged and labeled data with a typology module that classifies the consumer's inner tendencies into a total of 256 types based on four tendencies: relationship attitude, way of thinking, decision-making, and lifestyle; and A system for producing a consumer internal tendency typology analysis model using generative AI, characterized in that it includes an encoding / token conversion module that processes LLM encoding and token conversion for communication with generative AI.

3. In claim 2, The above prompt optimization module is a system for creating a consumer internal tendency typology analysis model using generative AI, characterized in that it analyzes the topics and prompt patterns of an existing generated prompt database using an artificial intelligence neural network-based algorithm and automatically generates prompts that match the input topics.

4. In claim 2, The above basic typing module is, A consumer internal tendency typology analysis model production system utilizing generative AI that categorizes tagged and labeled data into up to 16 types based on the four tendencies of relationship attitude, thinking style, decision-making, and lifestyle, and then further categorizes them into up to 16 types based on the secondary indicators, thereby categorizing consumer groups into a total of 256 types.

5. In claim 2, The above basic typing module is, A primary indicator module that categorizes consumer tendencies into two primary indicators for each of the four tendencies; A secondary indicator module that categorizes primary indicators into two types based on consumer preferences; A primary indicator determination module that inputs the data tagged and labeled by the above data preprocessing module into the primary indicator module and determines the primary indicator for each of the four tendency topics; A secondary indicator determination module that inputs the data tagged and labeled by the above data preprocessing module into the secondary indicator module to determine a secondary indicator for each of the four tendency topics; and A system for producing a consumer internal tendency typing analysis model using generative AI, characterized by including a basic typing result output module that outputs basic typing results by reflecting the decisions of the primary and secondary indices.

6. In claim 5, The above basic typing module is, A primary indicator learning module that trains primary indicators for each of the four tendency topics by applying tagged and labeled training data to a supervised learning-based machine learning engine; and It further includes a secondary indicator learning module that trains secondary indicators for each of the four tendency topics by applying the tagged and labeled training data and primary indicators to a machine learning engine based on supervised learning; The above first indicator determination module applies the data tagged and labeled by the above data preprocessing module to the first indicator learning module to determine the first indicator for each of the four tendency topics. A system for producing a consumer internal tendency typology analysis model using generative AI, characterized in that the secondary indicator determination module applies the data tagged and labeled by the data preprocessing module and the primary indicator determined by the primary indicator determination module to the secondary indicator learning module to determine the secondary indicator for each of the four tendency topics.

7. In claim 1, The above type data generation unit, A generative AI application module that applies data received from the small data combination unit to the generative AI to expand the dataset of the typification module by additionally generating or modifying text and images that classify the consumer's inner tendencies into 256 types based on four tendencies: relationship attitude, thinking style, decision-making, and lifestyle. A natural language processing module that removes noise from text generated through a generative AI application module; An image classification module that classifies suitable images among the images generated through the generative AI application module; and A system for producing a consumer internal tendency typology analysis model using generative AI, characterized by including a data verification module that verifies text from which noise has been removed and classified images.

8. In claim 7, The above generative AI application module is, A consumer internal tendency typology analysis model production system utilizing generative AI, characterized by applying data received from the small data combination unit to generative AI to generate multiple questions and matching answers for each primary indicator for determining the primary indicator for each of the four tendencies of relationship attitude, way of thinking, decision-making, and lifestyle, and generating multiple questions and matching answers for each secondary indicator for determining the secondary indicator.

9. In claim 1, The above customized consumer typology module generation unit is: A consumer internal tendency typology analysis model production system utilizing generative AI characterized by generating a final typology module that classifies consumers into a total of 256 types by generating one question and a matching answer for each of the four tendencies topics of relationship attitude, thinking style, decision-making, and lifestyle from the verified data received from the type data generation unit, and generating multiple questions and a matching answer for each of the secondary indicators to determine the secondary indicators.

10. The first step is to input the subject of the typology model that suits the customer's consumers; The second step is to generate prompts related to the four tendencies of relationship attitude, thinking style, decision-making, and lifestyle that can identify the inner tendencies of consumers regarding the input topic; The third step is to collect data using the prompts generated from the prompt optimization module and to tag and label the collected data; The fourth stage is to classify the tagged and labeled data into up to 16 types based on the four tendencies of relationship attitude, way of thinking, decision-making, and lifestyle, and then classify them into up to 16 types based on the secondary indicators, thereby classifying the consumer group into a total of 256 types; Step 5: Processing LLM encoding and token conversion for communication with generative AI; Step 6: Using the received tokens, apply them to the generative AI to expand the dataset of the typology module by additionally generating or transforming text and images that are classified into 256 types based on the four tendencies of relationship attitude, thinking style, decision-making, and lifestyle; Step 7: Removing noise from text generated by generative AI; Step 8: Classifying suitable images from among the images generated through generative AI; Step 9: Verification of the denoised text and classified images; and A method for producing a consumer internal tendency typology analysis model using generative AI, characterized by including a 10th step of generating a final typology module that classifies consumers into a total of 256 types by generating one question and a matched answer for each of the four tendencies of relationship attitude, way of thinking, decision-making, and lifestyle for each of the verified data, and multiple questions and a matched answer for each of the second indicators for determining the second indicator.

11. In claim 10, The fourth step above is, Apply the tagged and labeled data to the primary indicator learning module to determine the primary indicator for each of the four tendency topics. A method for creating a consumer internal tendency typology analysis model using generative AI, characterized by applying tagged and labeled data and the primary indicators determined by the primary indicator determination module to the secondary indicator learning module to determine secondary indicators for each of the four tendency topics.

Images