KR102467635B1 - System for providing consulting using artificial intelligence - Google Patents

Abstract

The present invention relates to a consulting providing system using artificial intelligence (AI), which comprises: a user terminal used by each user and communicating with a consulting providing server through a wired or wireless communication network; and a consulting providing server communicating with the user terminal, providing consulting-related information according to a request of the user terminal, collecting commercial district data in a specific region requested by the user terminal and personal data of a user using the user terminal, storing the collected commercial district data and personal data in a database, analyzing the commercial district data and personal data converted into big data using AI, and visualizing and sharing the analysis results to provide consulting-related information. According to the present invention, optimal consulting information is quickly provided to each user by providing a consulting service using AI technology.

Description

Consulting providing system using AI {System for providing consulting using artificial intelligence}

The present invention relates to a system for providing consulting, and more particularly, to a system for providing consulting using AI (artificial intelligence).

According to the OECD, the self-employed account for 27.4% of all workers in Korea, ranking fourth in the world. As such, self-employment plays a buffering role in the domestic economy during an economic downturn and also serves as a social safety net. The self-employed, who support the infrastructure of our economy, are relatively vulnerable to economic changes and lack management capabilities. Therefore, as political and social interest in the self-employed increases, the demand for policy management support is also increasing.

According to the 2020 self-employment status of the National Statistical Office, self-employed people who were directly hit by Corona tried to cut costs by sending part-time workers or employees, and as a result, the number of self-employed people with employees decreased by 160,000. Due to the closure of businesses and reduced employment in the self-employment sector, there are concerns about deepening social polarization, and it is expected to become an obstacle to the sustainable growth of the national economy. Therefore, it is absolutely necessary to expand support as a social support policy including self-employment management support at the national level.

According to the Ministry of SMEs and Startups, about KRW 4.2 trillion in self-employment support is expected in 2021, but self-employed and small business digitalization support (KRW 127.6 billion), loans for the underprivileged with low credit (KRW 3.7 trillion), support for closing businesses (KRW 69.1 billion), nurturing 100-year stores (KRW 5.9 billion) and new business start-up academy (KRW 18.95 billion), management support for strengthening self-sufficiency in the mid- to long-term is insufficient. Self-employment management support that checks and supports the management status of the self-employed business owner who is currently running a business, as there is a lack of support policies to improve the management of self-employed business owners who are running their business normally. It seems necessary to expand and diversify the solution.

Changes in the economic environment are accelerating due to the advent of the new normal era and the corona crisis, and continuous loan support is necessary for self-employed people who are vulnerable to changes in the economic environment to strengthen their capabilities, but most self-employed people have no experience in operating a business. As in any TV program, there are many cases where customers lose customers and fall into difficulties due to various and trivial problems such as poor hygiene management and a change in taste that customers do not prefer. Therefore, it is necessary to develop an AI-based self-consulting system that continuously recommends accurate and short-cycle diagnosis and improvement items by learning various data such as customer surveys and feedback, away from management diagnosis based on commercial district analysis data, which has a relatively long update cycle.

Currently, the commercial area analysis system provided by local governments or the Small Business Corporation is provided, but it is difficult for self-diagnosis due to the lack of time and competency of the self-employed. Some self-employed people who want to improve management apply for management support consulting and receive a management diagnosis from a consultant, but since it is a one-time visit, there is little data other than commercial area analysis information, and the consulting content desired by the self-employed does not match the main knowledge base of the visiting consultant. Various problems such as Therefore, it is necessary to provide an easy and convenient AI self-consulting service that anyone can use anytime, anywhere.

Consulting requests from the self-employed are in most cases requested by institutions in the previous stages, such as loans and credit guarantees, and the existing consulting process is difficult to access for livelihood self-employed who lack administrative processing capabilities and time. Therefore, it is necessary to improve the easily accessible and simplified consulting process.

In general, during business hours, the consultant visits the store and conducts on-site consulting, so it is difficult to obtain reliable diagnosis information due to errors in communication with the store owner, errors caused by the store owner's intention to conceal sales information, and lack of time for store diagnosis. It is a state. Therefore, it is necessary to provide an accurate AI-based diagnosis consulting system using various big data such as personal data and customer surveys other than existing commercial district analysis data.

Korean Registered Patent No. 10-1956212

The present invention has been devised to solve the above problems, and an object thereof is to provide a consulting providing system capable of performing self-consulting using AI.

The object of the present invention is not limited to the object mentioned above, and other objects not mentioned will be clearly understood by those skilled in the art from the description below.

To achieve this object, the present invention relates to a consulting providing system using AI (artificial intelligence), which is a terminal used by each user, and communicates with a consulting providing server through a wired or wireless communication network and a user terminal communicating with the user terminal. In addition, consulting-related information is provided according to the request of the user terminal, and commercial district data in a specific area requested by the user terminal and personal data of a user using the user terminal are collected, and the collected commercial district data and personal data are stored in the big data. It includes a consulting providing server that provides consulting-related information by converting into data and storing it in a database, analyzing big data and personal data using AI (artificial intelligence), and visualizing and sharing the analysis results.

The consulting providing server is a commercial business DB, commercial area, sales information, resident population, residential population, office worker population, income quintile, rental market price, building information, road name address, major attracting facilities, foreigner information, shopping mall information, tourism activation index , commercial district vitalization index, business start-up risk, industry overdensity, night lighting information, 3D location analysis, commercial district change information, and commercial district data including automobile registration can be collected.

The consulting providing server collects personal data including user's financial, tax, legal, military service, housing, real estate, health, medical, marriage, parenting, education, entrepreneurship, business, culture, immigration, employment, and work-related data. can

The consulting providing server may provide consulting-related information to the user terminal through a chatbot function.

The consulting providing server is the average business period information required for consulting using the quantitative information of the commercial district through the chatbot function and the qualitative information of the questionnaire linked to hyperresearch, sales per area unit, competition intensity of the same industry, break-even point, residential population information, resident population density It can provide consulting-related information including information, rent, major facilities for attracting customers, opening and closing business information, and trend information on sales change per area unit.

The consulting providing server may provide consulting-related information in the form of a report to the user terminal.

The consulting providing server provides average business period, sales per area unit, competition intensity of the same industry, break-even point, residential population information, permanent population density information, rent, major customer facilities, opening and closing business information, sales change trend information per area unit, hyper research A report including survey information of consumers and store owners can be provided on one screen.

According to the present invention, by providing a consulting service using AI technology, there is an effect that optimal consulting information can be quickly provided to each user. In particular, it is expected that the AI curating service will be able to provide customized services that each self-employed person needs.

And, according to the present invention, there is an effect that can increase the survival period of the self-employed by providing a self-consulting service.

In addition, according to the present invention, there is an effect of providing safe and reliable self-employed consulting to improve the self-employed's confidence in the state and improve the self-employed's business will.

In addition, according to the present invention, various social costs for supporting the self-employed can be expected, and efficient utilization of social, policy, and economic resources provided to the self-employed in the metropolitan area is expected through consulting efficiency.

In addition, psychological stability and social sympathy that the self-employed are protected and supported by the state are strengthened, and inequality with large companies can be resolved by strengthening the self-sufficiency of the self-employed.

1 is a conceptual diagram for explaining the need for AI consulting.
2 illustrates the configuration of a consulting providing system according to an embodiment of the present invention.
3 is a platform configuration diagram of a consulting providing system according to an embodiment of the present invention.
4 illustrates commercial district data according to an embodiment of the present invention.
5 illustrates personal data according to an embodiment of the present invention.
6 illustrates an information providing process through a chatbot according to an embodiment of the present invention.
7 is an example of an information providing screen through a chatbot according to an embodiment of the present invention.
8 is an example of a screen provided with consulting information according to an embodiment of the present invention.

Since the present invention can make various changes and have various embodiments, specific embodiments will be illustrated in the drawings and described in detail. However, this is not intended to limit the present invention to specific embodiments, and should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention.

Terms used in this application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, the terms "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present application, it should not be interpreted in an ideal or excessively formal meaning. don't

In addition, in the description with reference to the accompanying drawings, the same reference numerals are given to the same components regardless of reference numerals, and overlapping descriptions thereof will be omitted. In describing the present invention, if it is determined that a detailed description of related known technologies may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted.

1 is a conceptual diagram for explaining the need for AI consulting.

Referring to Figure 1, consulting 1.0 is a level of oral consulting, consulting 2.0 is to provide knowledge-based consulting services, consulting 3.0 is to provide consulting services based on big data to consulting 2.0.

In the past, consulting quality differed according to the individual competency of consultants, and theory-based, one-size-fits-all consulting services were provided. In addition, due to the high dependence on commercial information, diagnosis by self-employed person was inaccurate, and there was inconvenience for small business owners due to face-to-face consulting.

So, in Consulting 4.0, we propose a consulting service that combines AI technology with the existing Consulting 3.0. According to Consulting 4.0, consulting services can be provided anytime, anywhere using AI technology, and quantitative and qualitative data can be easily utilized throughout life.

The present invention relates to a consulting provision system using AI.

2 illustrates the configuration of a consulting providing system according to an embodiment of the present invention.

Referring to FIG. 2 , the consulting providing system of the present invention includes a consulting providing server 100 and a user terminal 200 .

The consulting providing server 100 communicates with the user terminal 200 and provides consulting-related information according to the request of the user terminal 200, but uses the commercial district data of a specific region requested by the user terminal 200 and the user terminal. Collects personal data of users, converts the collected commercial data and personal data into big data, stores them in the database 110 (DB), analyzes the commercial data and personal data converted into big data using AI (artificial intelligence), , provides consulting-related information by visualizing and sharing analysis results.

The user terminal 200 is a terminal used by each user, and communicates with the consulting providing server 100 through a wired or wireless communication network.

In one embodiment of the present invention, the user terminal 200 refers to any device that can access the server 100 through a wired or wireless communication network, such as a desktop PC, a laptop computer, a tablet PC, a smart phone, a mobile communication terminal, and the server 100 A dedicated application for providing consulting information provided by may be installed.

The consulting providing server 100 is a commercial business DB, commercial area, sales information, resident population, residential population, office worker population, income quintile, rental market price, building information, road name address, major attracting facilities, foreigner information, shop vacancy information, tourism It is possible to collect commercial district data including vitalization index, commercial district vitalization index, entrepreneurial risk, industry overdensity, night lighting information, 3D location analysis, commercial district change information, and vehicle registration.

The consulting providing server 100 is a user's personal data, including finance, tax, law, military service, housing, real estate, health, medical care, marriage, parenting, education, entrepreneurship, business, culture, immigration, employment, and work-related data. can be collected

The consulting providing server 100 may provide consulting-related information to the user terminal 200 through a chat-bot function.

At this time, the consulting providing server 100 is the average business period information required for consulting using the quantitative information of commercial districts through the chatbot function and the qualitative information of the questionnaire linked to hyperresearch, sales per area unit, competitive strength of the same industry, break-even point, residential population information , it can provide consulting-related information including permanent population density information, rent, major customer facilities, opening/closing business information, and sales trend information per area unit.

The consulting providing server 100 may provide consulting-related information to the user terminal 200 in the form of a report.

At this time, the consulting providing server 100 calculates average business period, sales per area unit, competition intensity of the same type of industry, break-even point, residential population information, resident population density information, rent, major customer facilities, opening and closing business information, change in sales per area unit A report containing information, hyperresearch consumers, and store manager survey information can be provided on a single screen.

3 is a platform configuration diagram of a consulting providing system according to an embodiment of the present invention.

Referring to FIG. 3 , the system for providing consulting according to the present invention first collects data and builds storage. Here, it is possible to build a private cloud-based data common infrastructure that can flexibly handle large-capacity traffic and security and connect each data provider, and consists of cloud storage and RDBMS for storing various types of large-volume and large-volume data. can do. In addition, it provides a data collection basis according to data characteristics such as structured, unstructured, and semi-structured data. In addition, for the operation of defined datasets and services, data is periodically collected from data providers and updated to ensure up-to-dateness. In addition, it creates a data pipeline between the data provider and the platform through data collection based on linkage integration middleware. In the present invention, an intuitive data flow is managed through a dashboard-based data collection status board, and an optimal data collection interface is provided in consideration of the operating environment of the data providing institution, data characteristics, and frequency of occurrence.

In the present invention, various preprocessing functions for collected data for data analysis are provided. Deploy the trained model through building a machine learning pipeline and reflect raw data changes. Here, the machine learning pipeline means automating the entire series of processes for utilizing machine learning technology, from data validation, preprocessing and segmentation, model creation and training, verification, and recording and archiving. When raw data is changed or added, a new model is created or a model creation and learning stage machine learning pipeline is built by adding data division nodes and model verification nodes to relearn pre-learned models.

In one embodiment of the present invention, a pipeline in which data preprocessing and regression model learning can be sequentially performed in a Kubernetes environment can be configured. The Argo workflow connects preprocessing algorithms with model training and hyperparameter optimization algorithms. In addition, it provides a management function that can search, delete, and utilize the information of the learned model to infer, and provides an API that can serve the learning model stably and quickly by utilizing kfserving.

In the preprocessing process of the present invention, categorical data and numerical data are automatically recognized. Missing values in categorical data are replaced with the most frequent values, missing values in numeric data are replaced with the median value, and rows with missing values in the prediction target column are deleted because they do not contain information necessary for prediction. In addition, columns with more than a certain percentage of missing values are deleted because they lack information, and categorical data is encoded in a form that can be entered into machine learning. Features of low importance are removed to improve the performance of the machine learning model.

In the present invention, a high-performance decision tree-based regression model algorithm is automatically optimized for the input dataset, model hyperparameters are optimized using a multi-precision based optimization algorithm, and a multi-armed bandit algorithm specialized for budget distribution and Bayesian specialized for optimal solution search are optimized. optimization, quickly explores multiple points in the search space with Bayesian optimization, eliminates budgets at low-precision points with a multi-armed bandit algorithm, and concentrates budgets at high-precision points. Then, learning is repeated to find the optimal hyperparameters for each regression model algorithm, then the best algorithm is selected, and the regression model is automatically optimized for the input data to achieve high prediction accuracy.

In the present invention, the SHAP technique that combines the Shapley value used in game theory and LIME, which is widely used as an XAI technique, is applied, a combination of several features is constructed to determine the importance of a feature, and then the average according to the presence or absence of the feature Seeking positive change In addition, it is possible to easily check how much change in prediction when a specific feature is removed by expressing it as a SHAP value, and it is also possible to express in which direction the feature affects the target to be predicted.

A report can be provided so that the user can easily understand data cleaning and feature engineering individual performance details performed in the preprocessing process of the present invention.

In the present invention, a self-employed self-consulting app can be provided. To this end, from the perspective of the service user, it is divided into a service layer for service users, a management layer for service operation managers, and a linkage layer that provides information linkage with external services using open APIs. The composition of the system framework is divided into application programs, data collection modules from external systems, intelligent analysis modules, and user data handling modules that store and manage information generated from services. It has a circular structure that is reused as a resource.

delete

Through the app, it is possible to provide the self-employed person with suggestions for using My Data, the subject and method of de-identification for storing and utilizing My Data, and the function of using de-identified My Data.

The app provides a diagnosis function for self-employed sales, which analyzes the user's sales information by including the affiliated commercial district, similar commercial district, and industry information, and checks the current business status. In addition, it provides a correlation analysis function with commercial district analysis elements, which analyzes the correlation between individual elements of commercial district analysis and sales, and visualizes the details. The self-diagnosis function for self-employed people displays self-diagnosis questions in an easy-to-see mobile app and provides functions to manage answers (view problems, move to previous/next, save and recall marked contents, etc.).

In the capacity building activity recording function, it provides a function to record the details of carrying out the capacity building activity presented as a result of consulting and a function to attract records.

Customer survey-related functions include a preview of the questionnaire and a function to view survey result trend analysis information, and the diagnosis history management function is a function to check analysis history and analysis results by period and analysis case.

In the present invention, the function of diagnosing the current level of success by combining commercial area information and sales information of self-employed persons. Provides an API that performs the function of providing information corresponding to the location of the applicant as a result of diagnosis by converting information such as sales trends of similar industries in commercial districts, sales trends of the same industries in similar commercial districts, and the average of all detailed industries into the amount per unit area can do.

The self-diagnosis survey function for self-employed workers is a function that extracts and provides survey questions to confirm achievement of key competency elements based on the details of the self-employed business type. have.

The capacity building activity recommendation API is an API that recommends capacity building activities based on self-diagnosis results.

4 illustrates commercial district data according to an embodiment of the present invention.

Referring to FIG. 4, commercial district data includes the number of stores, the number of opening and closing businesses, a commercial business DB including store history check items, a commercial district including tourist commercial districts and development commercial district items, sales information including sales per unit area, Resident population including street/building unit permanent population, residential population including population census registration statistics, office worker population including gender/age registration statistics, income quintile including 10th quintile income estimate, small business rental Rental market price including market information items, building information including address building and building ledger items, road name address including map data road address items, major attracting facilities including population concentration inducing items, sales and living population items Included foreigner information data is exemplified.

In addition, commercial vacancy information including commercial vacancy rate by administrative district, tourism activation index including foreign tourism activation degree, commercial district activation index including activation information by commercial district, and start-up risk including survival probability of start-up companies , Industry overdensity including the degree of overcrowding of industries, night lighting information including satellite night photography (brightness), 3D location analysis including industry/region selection (use of commercial district information), survival and closure areas Vehicle registration book data including changes in commercial districts including period comparison items and consumption expenditure level items are exemplified.

In the present invention, tourism activity information by region and commercial district can be utilized when providing consulting. In the present invention, the process of calculating the foreign tourism activity index and index in the consulting providing server 100 is as follows.

The foreign tourism activity index and indicator is a commercial activity measurement index and indicator of a life-oriented industry, and represents a combination index and indicator of foreign short-term stay and sales information.

Relative comparison at any point in time is not sufficient to indicate the degree of "activity", so a standardized interval (eg, 1 year) is used to implement the relativization index. For example, information on foreign short-term stay population and sales can be used as utilization data.

Individual variable standardization (z-score) = (raw score - mean score (standardization interval)) / standard deviation (standardization interval) (1),

Foreign short-term stay rate = (number of short-term residents in month t, year y)/ (total number of short-term residents in month t, year y) × 100 (2),

Foreign sales rate = (sales amount in month t of year y) / (total amount of sales in month t of year y) × 100 (3), the z-score is as follows.

z-score = (foreign short-term stay rate + foreign sales rate)/2 (4)

The T-score, an index of foreign tourism activity, is as follows.

T-score = z-score × 10 + 50 (5)

In one embodiment of the present invention, the foreign tourism activity index is divided into 4 stages, and the specific details are as follows.

Step 1 (smooth): T (foreign tourism activity index) value is 60 or higher

Step 2 (Slow): T (foreign tourism activity index) value is more than 51 ~ less than 60

Step 3 (delay): T (foreign tourism activity index) value is over 40 to less than 51

Step 4 (Caution): T (foreign tourism activity index) value is 40 or less

In the present invention, the consulting providing server 100 can utilize domestic and foreign tourism activity information by grafting it to a commercial district analysis service, and is consistent as an important variable for estimating the degree of regional activation of tourist commercial district recommendations along with gentrification and commercial district change indicators. Service quality can be improved by securing data. For example, a tourism vitalization map can be created in connection with the shopping district DB.

In the present invention, industry overdensity indicates the degree to which businesses are densely distributed in one commercial district beyond the appropriate level to achieve market equilibrium. It is an index that indicates the degree of risk that can be faced in the commercial environment.

Competitive saturation is the degree to which businesses are densely distributed in one commercial district beyond the appropriate level to achieve market equilibrium, and the appropriate degree of concentration can be measured using the General Method of Moment (GMM) model. Expressing this as a formula is:

(6)

(6)

The risk of business closure indicates the probability that the market environment and business characteristics of a business will lead to actual business closure, and the probability of business closure can be predicted using the survival function (Hazard ratio function) model. Expressing this as a formula is:

(7)

(7)

In the present invention, industry overdensity can be classified into four types, which can be classified into risk, alert, caution, and friendship.

Risk is a level of risk in the regional market due to an increase in the risk of business closure due to increased competition saturation, alert is a level where competition saturation is low but the risk of business closure is high, so regional market vigilance is necessary, and caution is a level where competition saturation is high but the risk of business closure is low The regional market is in a state that requires attention, and the local market is in a relatively favorable state, with both competitive saturation and the risk of business closure being low.

In the present invention, the commercial district change index classifies changes in commercial space based on the average operating period of surviving and closed businesses.

In the present invention, the start-up risk represents the degree of risk obtained by estimating the survival probability of a start-up company in a corresponding region through a business closure rate and survival rate for a recently determined period. For example, the risk of starting a business (100 businesses) close to life can be expressed as a combined index of the closure rate and the 3-year survival rate. Relative comparison at a certain point in time is not sufficient to indicate the degree of “risk” here, so a standardized interval (3 years) is used to implement the relativization index.

5 illustrates personal data according to an embodiment of the present invention.

As shown in FIG. 5, in the present invention, as personal data, the user's finance, tax, law-related information, life, quarantine-related information, housing, real estate-related information, health, medical-related information, marriage, parenting, education-related information, Entrepreneurship, management-related information, leisure, culture, immigration-related information, employment, workplace-related information, etc. are exemplified.

6 illustrates an information providing process through a chatbot according to an embodiment of the present invention.

Referring to FIG. 6, in the present invention, the consulting providing server 100 provides quantitative information through a chatbot function, average business period information required for consulting using qualitative survey information linked to hyperresearch, sales per area unit, and competitive strength of the same industry , break-even point, residential population information, permanent population density information, rent, major customer facilities, opening/closing business information, and consulting-related information including sales trend information per area unit can be provided.

7 is an example of an information providing screen through a chatbot according to an embodiment of the present invention.

Referring to FIG. 7 , when a break-even point is clicked in a state in which a chatbot function is activated, a response requesting input of an industry type is received. In addition, if an industry type is input, a response to input an autonomous district is received, and if an administrative district is clicked, a response to input area, fixed cost, and variable cost is sequentially received. Then, the chatbot guides you through the results based on the entered information.

8 is an example of a screen provided with consulting information according to an embodiment of the present invention.

As shown in FIG. 8, in the present invention, the consulting providing server 100 includes average business period, sales per area unit, competitive strength of the same type of industry, break-even point, residential population information, permanent population density information, rent, major attracting facilities, individual Reports including business closure information, sales trend information per area unit, and survey information of consumers and store owners of Hyper Research can be provided on one screen.

As described above, the user of the consulting provision service provided by the present invention is a self-employed person (a person who manages a business that provides consulting in the service or is entrusted with management), and the management consulting service that the self-employed person can receive is business status analysis and It is a self-diagnosis-based capacity building recommendation service and customer questionnaire recommendation service. Through the consulting service provided by the present invention, the self-employed can set up a marketing plan that responds to external factors that affect their business situation, and can conduct business stably by improving internal factors through capacity building activities. can

In the present invention, by using Hyperdata, an AI analysis platform that provides data insight through big data collection and processing and real-time analysis, it is intended to provide self-employed business owners with insight into the correlation between commercial data and self-employed personal data and sales. Among them, consulting information is provided using AutoML, a technology that helps non-experts to easily perform AI analysis by automating machine learning technologies such as data preprocessing and application of artificial intelligence algorithms. In the service scenario, when commercial data and self-employed personal data exist, a correlation between commercial data and sales is derived through data analysis, and a simple sales prediction artificial intelligence model is created using the data so that self-employed people can use it.

When commercial district data and self-employed individual data exist, in the case of sales forecasting for AutoML services, it is a regression process that predicts a figure called sales rather than a process of classifying a category, so select the regression process required on the AutoML API You can proceed with the analysis by selecting the target column, which is the column you want to predict, as sales. When the AutoML service operates, it is largely divided into a data pre-processing step and an artificial intelligence algorithm application step. In the data preprocessing step, various functions are performed, such as missing value processing, data processing according to the degree of imbalance, data conversion, and key feature selection. While data pre-processing is in progress, information on which tasks have been performed may be provided to the user.

Regression algorithm is an analysis method that measures the degree of fit after obtaining a model between the observed independent variable and the dependent variable. Currently using Random Forest, XGBoost, LightGBM, CatBoost algorithms. These algorithms are algorithms that show excellent performance in the field of structured data regression, and artificial intelligence models are learned through these algorithms. Each predictive model is formed through the corresponding algorithm, and it is stored in the AutoML server created through the kubernetes environment. In addition, through feature importance derived from the process of learning the model, the self-employed can analyze the correlation between commercial district data and sales. In addition, the performance of the trained model can be evaluated through the evaluation index. In addition, when additional data is generated as the self-employed conduct capacity building activities through self-surveys, AutoML is performed by creating a new sales forecasting model based on data that integrates the data and commercial district data. Correlation between empowerment factors and sales can be derived through feature importance from the process of training the model. In addition, the performance of the trained model can be evaluated through the evaluation index. The self-employed can use the learned model to infer future sales based on new data.

In the present invention, data insights related to sales are provided to the self-employed through such a technical process. Hyperdata has the advantage of integrating and analyzing various types of data based on data virtualization technology. In addition, it has the advantage of analyzing data in real time by performing logical data analysis modeling. In particular, by minimizing physical data duplication, the required data can be easily analyzed and utilized at the desired time, maximizing efficiency. Lastly, there is an advantage that self-employed people without AI knowledge can easily perform AI analysis. In the present invention, if the commercial data construction and the self-employed personal data construction are well done, it is expected that the AutoML service can be easily used by linking the data with Hyperdata.

The present invention has been described above using several preferred embodiments, but these embodiments are illustrative and not limiting. Those skilled in the art to which the present invention pertains will understand that various changes and modifications can be made without departing from the spirit of the present invention and the scope of rights set forth in the appended claims.

100 consulting providing server 200 user terminal
110DB

Claims (7)

As a terminal used by each user, the user terminal communicates with the consulting providing server through a wired or wireless communication network; and
Communicates with the user terminal, provides consulting-related information according to the request of the user terminal, collects commercial district data of a specific region requested by the user terminal and personal data of a user using the user terminal, and collects commercial district data A consulting providing server that converts and personal data into big data and stores them in a database, analyzes big data commercial data and personal data using AI (artificial intelligence), and provides consulting-related information by visualizing and sharing the analysis results. including,
The consulting providing server is a commercial business DB, commercial area, sales information, resident population, residential population, office worker population, income quintile, rental market price, building information, road name address, major attracting facilities, foreigner information, shopping mall information, tourism activation index , Collecting commercial district data including commercial district vitalization index, entrepreneurial risk, industry overdensity, night lighting information, 3D location analysis, commercial district change information, and automobile registration,
The consulting providing server collects personal data including data related to finance, tax, law, military service, housing, real estate, health, medical care, marriage, parenting, education, entrepreneurship, management, culture, immigration, employment, and work of the user, and ,
The consulting providing server provides consulting-related information to the user terminal through a chatbot function, and at this time, average business period information required for consulting using quantitative information of commercial districts and qualitative survey information linked to hyper research through a chatbot function, sales per area unit, Provides consulting-related information, including competition intensity, break-even point, residential population information, permanent population density information, rent, major guest facilities, opening and closing business information, and sales change trend information per area unit in the same industry,
The consulting providing server provides consulting-related information in the form of a report to the user terminal, wherein the average business period, sales per area unit, competitive strength of the same type of industry, break-even point, residential population information, permanent population density information, rent, and major attracting customers It provides a report that includes facilities, opening and closing information, sales trend information per area unit, and survey information of consumers and store owners of Hyper Research on one screen,
The commercial district data includes the number of stores, the number of opening and closing businesses, a commercial business DB including store history check items, a commercial district area including tourism commercial areas and developed commercial areas areas, sales information including sales items per area unit, road unit/building unit Resident population including permanent population, residential population including population census registration statistics, office worker population including gender/age registration statistics, income quintile including 10th quintile income estimate, small business rental price information rental market price, building information including address building and building ledger items, road name address including map data road address items, major attracting facilities including facilities that cause population concentration, foreigner information data including sales and living population items , commercial vacancy information including commercial vacancy rate by administrative district, tourism vitalization index including foreign tourism activation degree, commercial district activation index including activation information by commercial district, start-up risk including survival probability of start-up companies, Industry overdensity including degree of industry overcrowding, night lighting information including satellite night photography (brightness), 3D location analysis including industry/region selection (use of commercial district information), survival and closure period Including trade area changes, vehicle registration data, including comparison items,
The consulting providing server is a commercial activity measurement index and indicator of a life-oriented industry, and can calculate and provide a foreign tourism activity index and indicator representing a combined index and indicator of foreign short-term stay and sales information. At this time,
Individual variable standardization (z-score) = (raw score - mean score (standardization interval)) / standard deviation (standardization interval) (1),
Foreign short-term stay rate = (number of short-term residents in month t, year y)/ (total number of short-term residents in month t, year y) × 100 (2),
Foreign sales rate = (sales amount in month t of year y) / (total amount of sales in month t of year y) × 100 (3), z score,
z score = (foreign short-term stay rate + foreign sales rate) / 2 (4),
The T score, which is the foreign tourism activity index,
T score = z score × 10 + 50 (5),
The foreign tourism activity index is,
Step 1 (smooth): T (foreign tourism activity index) value is 60 or higher
Step 2 (Slow): T (foreign tourism activity index) value is more than 51 ~ less than 60
Step 3 (delay): T (foreign tourism activity index) value is over 40 to less than 51
Step 4 (Caution): T (foreign tourism activity index) value is 40 or less
divided into four stages of
The consulting providing server indicates the degree to which businesses are densely distributed in one commercial area beyond the appropriate level to achieve market equilibrium, and the risk that businesses may face in the commercial environment of the local market through the evaluation of the degree of competition saturation and the risk of business closure It is possible to calculate and provide industry overdensity, which is an indicator of
The degree of competition saturation is the degree to which businesses are densely distributed in one commercial district beyond the appropriate degree to achieve market equilibrium, and the appropriate degree of concentration can be measured using the General Method of Moment (GMM) model. ,

(6)
can be expressed as
The risk of closure indicates the probability that the market environment and business characteristics of the business lead to actual closure, and the probability of closure can be predicted using a survival function (Hazard ratio function) model. Expressed as a formula,

(7)
can be expressed as
The above industry overdensity is the risk that the regional market is at a risky level due to the increase in the risk of business closure due to the increase in competition saturation. A consulting provision system characterized in that it can be classified into four types: attention, which is a state requiring attention in the local market, and friendly, in which the local market is in a relatively favorable state with both competition saturation and business closure risk at a low level.
delete delete delete delete delete delete

Images