System for consumer propensity analysis and typing through small data extraction model and method thereof

Abstract

The present invention relates to a system for consumer propensity analysis and typing through a small data extraction model, the system including a typing result analysis unit that analyzes what needs classified consumers of each type have, and provides a customized solution for each consumer type. The present invention makes it possible to get an understanding of customers of a client company by classifying the customers into 256 types according to individual propensities, and enables the client company to establish customized customer services and customized marketing strategies for each consumer type, and provide product recommendations and personalization services.

Description

A consumer tendency analysis and typology system and method using a small data extraction model.

[0001] The present invention relates to a system and method for analyzing and categorizing consumer tendencies using a small data extraction model, and more particularly, to a system and method for analyzing and categorizing consumer tendencies using a small data extraction model, which utilizes consumer-related data collected from a client company to categorize the client company's consumers by tendencies and enable customized services for each customer type.

[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 understanding an individual's inner tendencies, preferences, preferences, and psychological characteristics. Small data focuses on understanding consumer characteristics 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' in-depth behavioral patterns and personalized needs.

[0004] Patents related to analyzing consumer personal tendencies include the following:

[0005] 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.

[0006] Publication Patent No. 10-2024-0062613 relates to an interior brokerage platform system that provides interior brokerage services to enable consumers to achieve desired results by identifying interior tendencies. The tendencies analysis unit builds an interior tendencies model for various interior styles recommended by a group of experts for each living space in advance, classifies the consumer's test answers received through the tendencies analysis process into preset answer types, learns a tendency classification method based on the matching of the interior tendencies model and answer types using a machine learning model, and classifies the interior tendencies models into multiple groups based on the tendencies classification method.

[0007] 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.

[0008] Publication No. 10-2024-0062613 is somewhat useful in predicting consumer preferences for interior design, but has limitations in that it cannot analyze and categorize in detail the patterns of diverse consumers of companies with diverse characteristics.

[0009] Therefore, by integrating and analyzing various small-scale data such as surveys, feedback, and purchase history, companies need a method to identify and categorize customer psychological characteristics, enabling them to better understand their customers and, based on this, improve services and products and manage customer management.

[0010] The present invention was invented to improve the above problems, and aims to provide a consumer tendency analysis and classification system through a small data extraction model that can accurately and in-depth analyze the consumer tendencies of a client company by applying data collected from the client company to a small data model and classify them by customer characteristics.

[0011] The present invention aims to provide a consumer tendency analysis and classification system through a small data extraction model that can accurately classify the types of consumers of a client company by classifying the consumers of the client company into 256 types based on four tendency themes of relationship attitude, way of thinking, decision-making, and lifestyle.

[0012] The present invention aims to provide a consumer tendency analysis and classification system using a small data extraction model that classifies and identifies the consumer of a client company into 256 types and enables the client company to provide customized customer services and customized marketing strategies for each consumer type.

[0013] The present invention was invented to improve the above problems, and aims to provide a consumer tendency analysis and classification system through a small data extraction model that can accurately and in-depth analyze the consumer tendencies of a client company by applying data collected from the client company to a small data model and classify them by customer characteristics.

[0014] The present invention aims to provide a consumer tendency analysis and classification system through a small data extraction model that can accurately classify the types of consumers of a client company by classifying the consumers of the client company into 256 types based on four tendency themes of relationship attitude, way of thinking, decision-making, and lifestyle.

[0015] The present invention aims to provide a consumer tendency analysis and classification system using a small data extraction model that classifies and identifies the consumer of a client company into 256 types and enables the client company to provide customized customer services and customized marketing strategies for each consumer type.

[0016] According to the present invention having the above configuration, the following effects are achieved.

[0017] By utilizing zero-party data collected directly from consumers, such as visit records, search records, purchase history, website activity, review data, inquiry and consultation history, and surveys, we can analyze individuals' inner tendencies more accurately and in-depth, and categorize customers into a total of 256 types based on their characteristics, enabling customized services for each customer type.

[0018] 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.

[0019] Additionally, companies can reduce costs by leveraging small-scale data rather than big data.

[0020] Figure 1 is a functional block diagram of a consumer tendency analysis and typology system using a small data extraction model according to the present invention.

[0021] Figure 2 is a detailed functional block diagram of the main components included in the consumer tendency analysis and typology system using the small data extraction model according to the present invention.

[0022] Figure 3 is a conceptual diagram of a typology module algorithm based on four tendencies.

[0023] Figure 4 is a structural diagram of the first and second indicators for the four tendencies.

[0024] Figure 5 is a conceptual diagram of inferring primary and secondary indicators using artificial intelligence.

[0025] Figure 6 is a flowchart of the consumer tendency analysis and typology method using a small data extraction model.

[0026] In order to achieve the above purpose, the consumer tendency analysis and classification system using a small data extraction model according to the present invention is configured to include a small data collection unit that collects consumer-related data of a client company, a small data preprocessing unit that corrects errors in the collected consumer-related data, removes duplication, and converts it into numerical data, a pattern analysis processing unit that analyzes the relationships between data using a statistical method on the preprocessed data to identify patterns, and algorithmizes the identified patterns to first classify the data, a consumer tendency detailed classification unit that secondarily classifies the consumer groups firstly classified by the pattern analysis processing unit using a cluster analysis algorithm and classifies the secondarily classified consumer groups into types based on four tendency themes of relationship attitude, thinking style, decision-making, and lifestyle, and a classification result interpretation unit that analyzes what needs each type of consumer classified based on the four tendency themes has and provides customized solutions for each consumer type.

[0027] The above pattern analysis processing unit,

[0028] It includes a pattern correlation analysis module that analyzes correlations between data through correlation analysis and regression analysis to identify patterns, and a pattern algorithmization module that formalizes the identified patterns into an algorithm to first type consumer-related data.

[0029] The above consumer tendency detailed typology section is:

[0030] It includes a cluster analysis-based consumer classification unit that uses a cluster analysis algorithm to segment the data classified into primary types by the pattern analysis processing unit into secondary types of data, and a tendency topic-based consumer classification unit that classifies the secondary types of data into primary indicators based on four tendency themes: relationship attitude, thinking style, decision-making, and lifestyle, and then classifies them again into secondary indicators to classify consumer groups into a total of 256 types.

[0031] The above cluster analysis-based consumer typology section is characterized by using hierarchical cluster analysis and density-based cluster analysis algorithms to subdivide groups within the same pattern of the first-typed data into second-typed data.

[0032] The above-mentioned consumer typology based on tendency themes is characterized by the fact that it can classify customers into a total of 256 types by applying a primary indicator that divides each of the four tendency themes of relationship attitude, way of thinking, decision-making, and lifestyle into two types, and a secondary indicator that divides each of the primary indicators into two other types, thereby classifying them into up to 16 types through the primary indicators and up to 16 types through the secondary indicators.

[0033] Specifically, the consumer typology based on the above tendencies is:

[0034] 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 the second classification data segmented by the cluster analysis-based consumer classification unit into the first indicator module to determine the first indicator for each of the four tendency topics, a second indicator determination module that inputs the second classification data segmented by the cluster analysis-based consumer classification unit into the second indicator module to determine the second indicator for each of the four tendency topics, and a final classification module that reflects the decisions of the first and second indicators and outputs the final classification results.

[0035] In addition, the consumer typology based on the above-mentioned tendency topic,

[0036] The second classification data segmented by the consumer classification department based on cluster analysis and customer information, one pre-written question composed of content related to the second classification data and the first indicator matched to the answer, and the first indicator determined by the first indicator determination module are learned by the first artificial intelligence indicator inference engine, and when the second classification data segmented by the consumer classification department based on cluster analysis and customer information are input, the first indicator learning module infers the first indicator for each of the four tendency topics, and the second indicator learning module infers the second indicator for each of the four tendency topics, when the second classification data segmented by the consumer classification department based on cluster analysis and customer information, three pre-written questions composed of content related to the second classification data and the second indicator matched to the answer, and the second indicator determined by the second indicator determination module are input, and the final classification module is The final typology result can be configured to be output by reflecting the decisions of the primary indicator inferred by the primary indicator learning module and the secondary indicator inferred by the secondary indicator learning module.

[0037] The above typology result interpretation section is,

[0038] It includes a consumer needs analysis module that analyzes the needs of consumers by type classified by the consumer tendency detailed typology department, and a customized solution provision module that provides customized solutions that meet the needs of consumers by type analyzed by the consumer needs analysis module.

[0039]

[0040] Meanwhile, in order to achieve the above purpose, the method for analyzing and classifying consumer tendencies using a small data extraction model according to the present invention comprises the following steps: a first step of collecting consumer-related data from a client company; a second step of correcting errors in the collected consumer-related data, removing duplication, and converting it into numerical data; a third step of analyzing correlations between data through correlation analysis and regression analysis to identify patterns; a fourth step of formalizing the identified patterns into an algorithm to first classify consumer-related data; a fifth step of segmenting the first-classified data into second-classified data using a cluster analysis algorithm; a sixth step of classifying the second-classified data into a total of 256 types by classifying them according to the first indicators based on four tendencies themes of relationship attitude, way of thinking, decision-making, and lifestyle, and then classifying them again according to the second indicators; and a seventh step of analyzing what needs each type of consumer classified based on the four tendencies themes has and providing customized solutions for each consumer type.

[0041] Specifically, step 6 is:

[0042] It can be configured to include a step of inputting the secondary typology data into the primary indicator module to determine the primary indicator for each of the four tendency topics, a step of inputting the secondary typology data into the secondary indicator module to determine the secondary indicator for each of the four tendency topics, and a step of outputting the final typology results by reflecting the decisions made in the primary and secondary indicators.

[0043] 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.

[0044] However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms.

[0045] 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.

[0046] And the present invention is defined only by the scope of the claims.

[0047] Accordingly, in some embodiments, well-known components, well-known operations, and well-known techniques are not specifically described to avoid obscuring the present invention.

[0048] 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.

[0049] 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.

[0050] 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.

[0051] Also, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless they are defined otherwise.

[0052] Hereinafter, a preferred embodiment of the present invention will be described with reference to the attached drawings.

[0053] The present invention will be described in detail with reference to FIGS. 1 to 6.

[0054] Consumer Propensity Analysis and Classification System Using Small Data Extraction Models

[0055] The consumer tendency analysis and typology system using a small data extraction model according to the present invention is configured to include a small data collection unit (100), a small data preprocessing unit (200), a pattern analysis processing unit (300), a consumer tendency detailed typology unit (400), and a typology result interpretation unit (500).

[0056] The small data collection unit (100) collects consumer-related data from the client company. Examples of consumer-related data are as follows.

[0057] We collect information about which pages consumers visit and which products they view on websites and mobile apps. This allows us to analyze consumer interests and behavioral patterns.

[0058] We collect information about what products consumers purchase and what payment methods they use. This allows us to understand consumers' purchasing tendencies and preferred payment methods.

[0059] We collect information about when customers made purchases and which shipping method they selected. This allows us to analyze purchase time patterns and identify shipping preferences.

[0060] Collect customer reviews of products they've purchased. This is useful for understanding product quality, customer satisfaction, and areas for improvement.

[0061] Customer inquiries and consultations are converted into data and stored. This allows for analysis of customer issues, needs, and areas for service improvement.

[0062] If existing data is insufficient, conduct a consumer survey to collect additional data. This allows for more detailed consumer opinions and feedback.

[0063] The small data collection unit (100) collects consumer-related data from the client company as described above through website and mobile app log analysis, transaction data collection, review and feedback collection, customer consultation data collection, and survey data collection.

[0064]

[0065] The small data preprocessing unit (200) corrects errors in the collected consumer-related data, removes duplication, and converts it into numerical data.

[0066] Specifically, the small data preprocessing unit (200) identifies and corrects incorrect, unrealistic, or missing values ​​in the collected data. It verifies the consistency of data for the same consumer across different databases or sources and corrects any inconsistencies. It identifies duplicate data items for the same consumer. If duplicate data is found, it consolidates the duplicate records into a single, consistent data item. It converts categorical data into numerical data suitable for analysis.

[0067]

[0068] The pattern analysis processing unit (300) analyzes the relationship between data using statistical methods on preprocessed data to identify patterns, and then algorithmizes the identified patterns to primarily type the data.

[0069] Specifically, the pattern analysis processing unit (300) is configured to include a pattern correlation analysis module (310) and a pattern algorithmization module (330).

[0070] The pattern correlation analysis module (310) analyzes the correlation between data through correlation analysis and regression analysis to identify patterns.

[0071] For example, by analyzing correlations between data, we can identify patterns in which customers search for a specific brand or model name of a high-priced product using keywords and then purchase that product.

[0072] Identify the purchasing patterns of customers who only click on the top menu of the site and those who click intensively on specific category pages.

[0073] Analyze the path a customer takes until they complete a purchase to identify patterns.

[0074] We analyze customer purchase patterns by checking reviews before purchasing a product. We also analyze customer purchase patterns by conducting keyword searches, checking product specifications, and then making a purchase.

[0075] The pattern algorithmization module (330) formalizes identified patterns into algorithms to initially categorize consumer-related data. This enables automatic data classification and analysis. Data is categorized using classification algorithms such as decision trees, random forests, and support vector machines (SVMs).

[0076]

[0077] The consumer tendency detailed typology unit (400) uses a cluster analysis algorithm to secondarily type the consumer group that the pattern analysis processing unit (300) has first-typed, and classifies the second-typed consumer group into types based on four tendency themes: relationship attitude, way of thinking, decision-making, and lifestyle.

[0078] Specifically, the consumer tendency detailed typology unit (400) is composed of a cluster analysis-based consumer typology unit (410) and a tendency topic-based consumer typology unit (420).

[0079] The cluster analysis-based consumer typology unit (410) uses a cluster analysis algorithm to subdivide the data that has been first typified by the pattern analysis processing unit into second typology data.

[0080] The cluster analysis-based consumer typology unit (410) uses hierarchical cluster analysis and density-based cluster analysis algorithms to subdivide groups within the same pattern of the first-typed data into second-typed data.

[0081] For example, hierarchical clustering analysis analyzes specific patterns in data categorized into primary categories, resulting in two groups, such as "high-frequency buyers" and "low-frequency buyers," which are then categorized into secondary categories. Density-based clustering analyzes specific patterns in data categorized into primary categories, further subdividing the groups within the patterns into "active explorers" and "passive explorers," resulting in secondary categorization.

[0082] In summary, the pattern analysis processing unit (300) identifies fundamental patterns within consumer-related data and performs primary categorization, while the cluster analysis-based consumer categorization unit (410) identifies deeper-level structures within the primary categorization. Even among consumers with identical purchasing patterns (primary categorization), groups with distinct characteristics (secondary categorization) are identified.

[0083] For example, let's say you've obtained data with purchase patterns of Types A, B, and C through pattern analysis. Cluster analysis can segment consumers with Type A into groups such as "high-frequency buyers," "low-frequency buyers," and "new buyers." This can help you develop a marketing strategy more tailored to Type A consumers.

[0084] By performing these two steps (first typification, second typification) sequentially, we can understand the structure of the data more clearly and establish a more sophisticated and effective strategy based on this.

[0085]

[0086] The consumer typology based on tendency themes (420) classifies the secondary typology data according to the primary indicators based on the four tendency themes of relationship attitude, way of thinking, decision-making, and lifestyle, and then classifies the consumer group into a total of 256 types by classifying it again according to the secondary indicators.

[0087] The consumer typology unit (420) based on the tendency theme applies the secondary typology data, i.e., the consumer's behavior, purchase pattern, purchase frequency, time taken for purchase, payment style, questioning, purchase time zone, etc. corresponding to the secondary typology, to the typology module based on the four tendency themes of relationship attitude, thinking style, decision-making, and lifestyle to finally type the consumer.

[0088] Referring to Figure 4, it can be seen that the present invention is composed of four tendencies (relationship attitude, way of thinking, decision-making, and lifestyle), a primary indicator of each tendencies, and a secondary indicator belonging to the primary indicator.

[0089] The consumer typology based on the propensity topic (420) applies a primary indicator that divides each of the four propensity topics of relationship attitude, way of thinking, decision-making, and lifestyle into two types, and a secondary indicator that divides each of the primary indicators into two more types, thereby classifying into up to 16 types through the primary indicators and up to 16 types through the secondary indicators, for a total of 256 types.

[0090] The primary indicators, multiplied by two relationship attitudes, two thought patterns, two decision-making styles, and two lifestyle attitudes, yield a total of 16 possible scenarios. Multiplying this by the 16 secondary indicators allows for the classification of consumer groups into a total of 256 types. This is a key feature of the present invention.

[0091] Referring to Figure 1, we can see an example where three consumers are typed as CgIdLpPg, AsEpSyFu, and CtIiLrFa, respectively, as a result of consumer typing.

[0092] The CgIdLpPg type means that the final result indicator of relationship attitude is Cg (primary indicator C, secondary indicator g), the final result indicator of thinking style is Id (primary indicator I, secondary indicator d), the final result indicator of decision-making is Lp (primary indicator L, secondary indicator p), and the final result indicator of lifestyle attitude is Pg (primary indicator P, secondary indicator g).

[0093] That is, when the consumer typology unit (420) based on the tendency topic applies the consumer's behavior, purchase pattern, purchase frequency, time taken to purchase, payment style, whether or not to ask questions, purchase time zone, etc. corresponding to the secondary typology to the typology module based on the four tendency topics of relationship attitude, thinking style, decision-making, and lifestyle, the consumer's tendency is finally typified as CgIdLpPg. The CgIdLpPg consumer type is explained as follows.

[0094] Cg: A tendency to be influenced by others and to want to help others.

[0095] Id: A tendency to be imaginative and think clearly and concisely.

[0096] Lp: A tendency to make rational decisions and prioritize the process over the result.

[0097] Pg: A person who values ​​principles and likes to socialize.

[0098] The present invention is characterized by the fact that it finally types consumer behavior data corresponding to the second typology by matching the first and second indicators based on the four tendencies of relationship attitude, thinking style, decision-making, and lifestyle.

[0099] For example, the consumer typology unit (420) based on the propensity topic applies the consumer's behavior, purchase pattern, purchase frequency, time taken to purchase, payment style, questioning, purchase time zone, etc. corresponding to the second typology to the typology module based on the four propensity topics of relationship attitude, thinking style, decision-making, and lifestyle, and the following insights can be derived through the final typology index.

[0100] A consumer is a consumer who is greatly influenced by others when purchasing furniture and has an intuitive tendency. These consumers prefer unique and trendy designs.

[0101] Consumer B is a practical consumer with a clear subjective viewpoint, and prefers products with good functionality and convenience.

[0102] This method allows consumers to be categorized into 256 categories. Companies can then use these findings to implement customized marketing strategies for each consumer type. For example, for consumers who prefer a minimalist style, furniture with a simple, modern design might be recommended, while for those who prefer a more individualistic style, furniture with a sophisticated, sophisticated design might be recommended.

[0103] And recommendation algorithms and UX can be optimized to enable easy online shopping for each customer type.

[0104]

[0105] A detailed look at the consumer typology based on tendency topics (420) is as follows.

[0106] The consumer typology unit (420) based on the tendency topic is composed of a primary indicator module (430), a secondary indicator module (440), a primary indicator determination module (450), a secondary indicator determination module (460), and a final typology module (470).

[0107] Referring to Figure 4, the primary indicator module (430) categorizes consumer tendencies into two primary indicators for each of the four tendencies. The secondary indicator module (440) categorizes the primary indicators into two secondary indicators based on consumer tendencies.

[0108] Referring to Figure 3, the primary indicator determination module (450) inputs the secondary typology data segmented by the cluster analysis-based consumer typology unit (410) into the primary indicator module (430) to determine the primary indicator for each of the four tendency topics. In other words, the primary indicator is determined based on the consumer behavioral data corresponding to the secondary typology.

[0109] Consumer behavioral data refers to the style of products purchased by consumers, purchase frequency, time taken to purchase, payment style, whether they check reviews, whether they ask questions, and when they usually make purchases.

[0110] Referring to Figure 3, the client company presents consumers with pre-written questions (Q1) that identify their tendencies on four topics related to the client company, such as finance, consumption, learning, health, and food. When the consumer responds (e.g., choosing one or the other), the client company determines the primary indicator matching the consumer's response as the consumer's primary indicator based on a preset primary indicator for each response. In other words, the client company conducts a survey of consumers to collect primary indicators on the four consumer tendencies.

[0111] Here, the pre-written question (Q1) is composed of content related to the client's topic and consumer behavioral data corresponding to the secondary typology. In other words, when conducting a survey targeting consumers of a furniture company, questions (Q1) are asked about furniture, space, interior design, etc. However, the question (Q1) is composed of consumer behavioral data corresponding to the secondary typology, and is related to the client's personality.

[0112] Consumer behavior data is diverse, including the style of products purchased by consumers, purchase frequency, time taken for purchase, payment style, whether or not they check reviews, whether or not they ask questions, and the main time of purchase. In addition, the types of client companies are diverse, such as interior design companies, insurance companies, health food companies, and hospitals. Therefore, the question (Q1) is composed of content related to consumer behavior data that is relevant to the client company and that matches the characteristics of the client company.

[0113] The primary indicator determination module (450) inputs the secondary classification data segmented by the cluster analysis-based consumer classification unit (410) into the primary indicator module (430) to determine the primary indicator for each of the four tendency topics. The primary indicator is determined by referring to the primary indicators matched to the consumer responses collected by the client company through a survey targeting consumers for the four tendency topics.

[0114]

[0115] Referring to Figure 3, the secondary indicator determination module (460) inputs the secondary typology data segmented by the cluster analysis-based consumer typology unit (410) into the secondary indicator module (440) to determine the secondary indicator for each of the four tendency topics. In other words, the secondary indicator is determined based on the consumer behavioral data corresponding to the secondary typology.

[0116] Referring to Figure 3, just like the primary indicator, the secondary indicator also uses a survey of the client's consumers.

[0117] The client company presents consumers with pre-written questions (Q2-1 to Q2-3) that identify their tendencies across four topics related to the client company, such as finance, consumption, learning, health, and food. When the consumer responds (e.g., choosing one or the other), the client company determines the secondary indicator that matches the consumer's answer based on a pre-established secondary indicator for each response as the consumer's secondary indicator. In other words, the client company conducts a survey of consumers and collects secondary indicators on the four tendencies.

[0118] However, unlike the primary indicator, the secondary indicator consists of three question-and-answer methods (Q2-1 to Q2-3) (the primary indicator consists of one question (Q1) for each tendency topic - relationship attitude / thinking style / decision-making / lifestyle). This was designed to enable checking the consistency of consumers' answers to questions about the secondary indicator.

[0119] The secondary indicator is determined by the indicator that exceeds the majority of answers to the three questions (Q2-1 to Q2-3) (for example, if the primary indicator is C, one of the answers to the three questions for the secondary indicator is g and the remaining two are t).

[0120] Here, the pre-written questions (Q2-1 to Q2-3) are composed of topics related to the client company and content related to consumer behavioral data corresponding to the secondary typology. In other words, when conducting a survey targeting consumers of a furniture company, questions (Q2-1 to Q2-3) are asked about furniture, space, interior design, etc. However, the questions (Q2-1 to Q2-3) are composed of content related to consumer behavioral data corresponding to the secondary typology that matches the personality of the client company.

[0121] Consumer behavior data is diverse, including the style of products purchased by consumers, purchase frequency, time taken for purchase, payment style, whether or not they check reviews, whether or not they ask questions, and the main time of purchase. In addition, the types of client companies are diverse, such as interior design companies, insurance companies, health food companies, and hospitals. Therefore, the questions (Q2-1~Q2-3) are composed of topics related to the client companies and content related to consumer behavior data that fits the characteristics of the client companies.

[0122] The secondary indicator determination module (460) inputs the secondary classification data segmented by the cluster analysis-based consumer classification unit (410) into the secondary indicator module (440) to determine the secondary indicators for each of the four tendency topics. The secondary indicators are determined by referring to the secondary indicators matched to the consumer responses collected by the client company through a survey targeting consumers for the four tendency topics.

[0123]

[0124] By examining consumer behavior data, it is possible to match the primary and secondary indicators for each of the four tendencies.

[0125] The final typing module (470) outputs the final typing result by reflecting the decisions of the primary and secondary indices.

[0126]

[0127] The consumer typology unit (420) based on the tendency topic includes a primary indicator learning module (480) and a secondary indicator learning module (490).

[0128] Referring to FIG. 5, the first indicator learning module (480) learns the second classification data segmented by the customer information and the cluster analysis-based consumer classification unit (410), one pre-written question (Q1) composed of content related to the second classification data, the first indicator matched to the answer, and the first indicator determined content of the first indicator determination module (450) using the first artificial intelligence indicator inference engine, and when the customer information and the second classification data segmented by the cluster analysis-based consumer classification unit (410) are input, the first indicator is inferred for each of the four tendency topics.

[0129]

[0130] By learning secondary typology data and corresponding customer information (finance, shopping malls, interior design, education, etc.) through machine learning or deep learning, we can determine the primary indicator for each of the four tendencies.

[0131] The secondary indicator learning module (490) learns the secondary classification data segmented by the customer information and the cluster analysis-based consumer classification unit (410), three pre-written questions (Q2-1 to Q2-3) composed of contents related to the secondary classification data, the secondary indicators matched to the answers, and the secondary indicator determination contents of the secondary indicator determination module (460) using the second artificial intelligence indicator inference engine, and when the customer information and the secondary classification data segmented by the cluster analysis-based consumer classification unit (410) are input, the secondary indicators are inferred for each of the four tendency topics.

[0132] By learning secondary typology data and corresponding customer information (finance, shopping malls, interior design, education, etc.) through machine learning or deep learning, we can determine the secondary indicators for each of the four tendencies.

[0133] Of course, sophisticated labeling of secondary typology data is required for learning.

[0134]

[0135] The typology result interpretation unit (500) analyzes the needs of each type of consumer classified based on four tendencies and provides customized solutions for each consumer type.

[0136] The typology result interpretation unit (500) is configured to include a consumer needs analysis module (510) and a customized solution provision module (530).

[0137] The consumer needs analysis module (510) analyzes consumer needs by type, as classified by the consumer tendency classification unit (400). This module helps understand the characteristics of each consumer group and clarifies the services they require.

[0138] Referring to Figure 4, if the client is an insurance company, for example, if the consumer's primary indicator is C and the secondary indicator is t, the type of consumer who has a tendency to value relationships with others (the primary indicator is C) and also has a tendency to influence others (the secondary indicator is t) is Ct, which is the sum of the primary and secondary indicators. In the case of insurance consumption, it also means 'a person who can influence people around him or her about the product he or she has subscribed to, or in other words, a person who can introduce it well.' The consumer needs analysis module (510) analyzes the needs of consumers in this way.

[0139]

[0140] The customized solution provision module (530) provides customized solutions tailored to the needs of each consumer type analyzed by the consumer needs analysis module (510). Based on the identified consumer needs, personalized content and solutions are provided. For example, customized product recommendations, personalized marketing messages, or customized discounts can be provided to specific consumer groups.

[0141]

[0142] Finally, we conduct a survey of the typified consumer groups and individuals. This allows us to identify gaps between the model's predicted consumer needs and actual consumer experiences.

[0143] Based on the survey results, we analyze the gap between the model's predictions and actual consumer feedback. This allows us to verify the model's accuracy and, if necessary, improve it to provide more effective, personalized services.

[0144]

[0145] [Consumer Propensity Analysis and Classification Methods Using Small Data Extraction Models]

[0146] Referring to Figure 6, we describe a consumer propensity analysis and typology method using a small data extraction model. Detailed explanations of areas that overlap with the above explanations have been omitted.

[0147] Step 1 (S610): Collect customer-related data.

[0148] Step 2 (S620): Correct errors, remove duplicates, and convert collected consumer-related data into numerical data.

[0149] Step 3 (S630): Analyze the correlation between data through correlation analysis and regression analysis to identify patterns.

[0150] Step 4 (S640): The identified patterns are formalized into an algorithm to form a primary type of consumer-related data.

[0151] Step 5 (S650): The first-typed data is subdivided into second-typed data using a cluster analysis algorithm.

[0152] Step 6 (S660): The secondary typology data is classified according to the primary indicators based on the four tendencies of relationship attitude, thinking style, decision-making, and lifestyle, and then classified again according to the secondary indicators to classify the consumer group into a total of 256 types.

[0153] Step 7 (S670): Analyze the needs of each type of consumer classified based on four tendencies and provide customized solutions for each consumer type.

[0154]

[0155] Step 6 can be structured as follows:

[0156] The secondary typology data is input into the primary indicator module to determine the primary indicator for each of the four tendencies.

[0157] Then, the secondary typology data is input into the secondary indicator module to determine the secondary indicator for each of the four tendency topics. At this time, the secondary indicator is selected from among the primary indicators determined above.

[0158] The final typing results are output by reflecting the decisions made in the primary and secondary indices.

[0159]

[0160] According to the present invention, which has the above-described structure, zero-party data collected directly from consumers, such as visit records, search records, purchase history, website activity, review data, inquiry and consultation records, and surveys, can be utilized to more accurately and in-depth analyze individual tendencies and categorize customers into a total of 256 categories based on their characteristics. This allows for the development and implementation of customized marketing strategies for each consumer type, thereby increasing customer loyalty.

[0161] 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.

[0162] 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. Small data collection department that collects consumer-related data from clients; A small data preprocessing unit that corrects errors, removes duplication, and converts collected consumer-related data into numerical data; Pattern analysis processing section that analyzes the relationships between data using statistical methods to identify patterns in preprocessed data and then algorithmically types the data into primary types; The consumer tendency detailed typology section uses a cluster analysis algorithm to secondarily type the consumer group that the above pattern analysis processing section has first typed, and classifies the second typed consumer group into types based on four tendency themes of relationship attitude, thinking style, decision-making, and lifestyle; and A consumer tendency analysis and typology system using a small data extraction model, characterized by including a typology result interpretation unit that analyzes the needs of each type of consumer classified based on four tendency themes and provides customized solutions for each consumer type.

2. In claim 1, The above pattern analysis processing unit, A pattern correlation analysis module that analyzes correlations between data through correlation analysis and regression analysis to identify patterns; and A consumer tendency analysis and classification system using a small data extraction model, characterized by including a pattern algorithmization module that formalizes identified patterns into algorithms to first classify consumer-related data.

3. In claim 1, The above consumer tendency detailed typology section is: A cluster analysis-based consumer classification unit that uses a cluster analysis algorithm to segment the data classified into primary types by the pattern analysis processing unit into secondary type data; and A consumer tendency analysis and typology system using a small data extraction model, characterized by including a tendency-theme-based consumer typology section that classifies secondary typology data into 256 types by classifying them according to primary indicators based on four tendency themes: relationship attitude, way of thinking, decision-making, and lifestyle, and then classifying them again according to secondary indicators.

4. In claim 3, A consumer tendency analysis and classification system using a small data extraction model characterized in that the cluster analysis-based consumer classification unit uses hierarchical cluster analysis and density-based cluster analysis algorithms to subdivide groups within the same pattern of the first-classified data and perform second-class classification.

5. In claim 3, The consumer typology based on the above tendencies applies a primary indicator that divides each of the four tendencies of relationship attitude, way of thinking, decision-making, and lifestyle into two types, and a secondary indicator that divides each of the primary indicators into two other types. A consumer tendency analysis and typology system using a small data extraction model that classifies consumers into up to 16 types through primary indicators and up to 16 types through secondary indicators, resulting in a total of 256 types.

6. In claim 3, The consumer typology based on the above tendencies is as follows: 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 secondary typology data segmented by the consumer typology department based on the cluster analysis into the primary indicator module to determine the primary indicator for each of the four tendency topics; A secondary indicator determination module that inputs the secondary typology data segmented by the cluster analysis-based consumer typology department into the secondary indicator module to determine secondary indicators for each of the four tendency topics; and A consumer tendency analysis and typing system using a small data extraction model, characterized by including a final typing module that outputs the final typing results by reflecting the decisions of the primary and secondary indices.

7. In claim 6, The consumer typology based on the above tendencies is as follows: A first artificial intelligence indicator inference engine learns the second classification data segmented by the consumer classification department based on customer information and cluster analysis, one pre-written question composed of content related to the second classification data and the first indicator matched to the answer, and the first indicator decision content of the first indicator decision module, and when the second classification data segmented by the consumer classification department based on customer information and cluster analysis is input, a first indicator learning module that infers the first indicator for each of the four tendency topics; and It further includes a second artificial intelligence index inference engine that learns the second index determination contents of the second index determination module, the second index classification data segmented by the consumer classification department based on customer information and cluster analysis, three pre-written questions composed of contents related to the second index classification data and the second index matched to the answers, and the second index classification data of the second index determination module, and when the second index classification data segmented by the consumer classification department based on customer information and cluster analysis is input, a second index learning module that infers the second index for each of the four tendency topics; A consumer tendency analysis and typing system using a small data extraction model, characterized in that the final typing module outputs the final typing result by reflecting the decisions of the first indicator inferred by the first indicator learning module and the second indicator inferred by the second indicator learning module.

8. In claim 1, The above typology result interpretation section is, A consumer needs analysis module that analyzes the needs of consumers by type classified by the consumer tendency detailed typology department; and A consumer tendency analysis and classification system using a small data extraction model, characterized by including a customized solution provision module that provides customized solutions tailored to the needs of each type of consumer analyzed by the consumer needs analysis module.

9. The first step is to collect consumer-related data from the client company; The second step is to correct errors, remove duplicates, and convert the collected consumer-related data into numerical data; The third step is to identify patterns by analyzing the correlation between data through correlation analysis and regression analysis; The fourth step is to formalize the identified patterns into an algorithm to first classify consumer-related data; The fifth step is to subdivide the first-typed data into second-typed data using a cluster analysis algorithm; The sixth step is to classify the secondary typology data into a total of 256 types by classifying them according to the primary indicators based on the four tendencies of relationship attitude, thinking style, decision-making, and lifestyle, and then classifying them again according to the secondary indicators; and A method for analyzing and categorizing consumer tendencies using a small data extraction model, characterized by including a seventh step of analyzing the needs of each type of consumer classified based on four tendencies and providing customized solutions for each type of consumer.

10. In claim 9, The above 6th step is, A step of inputting the secondary typology data into the primary indicator module to determine the primary indicator for each of the four tendency topics; A step of inputting the secondary typology data into the secondary indicator module to determine the secondary indicator for each of the four tendency topics; and A method for analyzing and classifying consumer tendencies using a small data extraction model, characterized by including a step of outputting a final classification result by reflecting the decisions of the primary and secondary indices.

Images