KR101782521B1 - Marketing promotion system of Knowledge base and the method thereof - Google Patents

Abstract

A knowledge-based marketing promotion system of the present invention comprises: a knowledge storing place; a rule manager; a rule database; a rule matching module; a rule filtering module; and an interface unit. The knowledge storing place stores client source data; a situation information table for extracting features of each client from the client source data; and a summary table for preprocessing the situation information table for client selection rule generation. The rule manager generates a client selection rule based on the summary table of the knowledge storing place. The rule database stores the client selection rule generated in the rule manager. The rule matching module selects a client who meets a promotion condition by matching the summary table information and client selection rule data. The rule filtering module enables the selected client to filter overlapping of the client selection rule. The interface unit transmits promotion information to the filtered client. The present invention generates the client selection rule by combining the knowledge storing place with knowledge of a marketer; and provides a promotion customized to changed client data by applying the client selection rule in real time.

Description

[0001] The present invention relates to a marketing promotion system based on a knowledge system,

The present invention relates to a marketing promotion system based on a knowledge system and a method thereof, and more particularly, it is possible to generate customized rules through a combination of knowledge of a knowledge store and a marketer and to provide customized promotions for changing customer data Based marketing promotion system and method thereof.

In recent years, small shops and small merchants such as convenience stores, franchise restaurants, and the like have been actively researching a system for searching for customers who need the promotion information and selectively providing them in order to promote the head office or the merchant's stock.

Most of the systems are based on Rule-based systems. Rule-based systems are based on expert knowledge and heuristics to form a knowledge base and solve various problems based on them. The knowledge base is called "if quot; -then "rule set.

However, the present promotion system for small merchants has the following limitations.

First, the massive accumulation of data for customers due to the introduction of smart devices (such as smart phones) and wireless information communication technologies (RFID, NFC) in order to update the complexity / As vast amounts of data are stored in a complex data structure, human interpretation is not possible and human interpretation is possible only after preprocessing after data collection. That is, it is difficult to apply a rule based on accumulated data in real time because it is not possible to form a rule set after the expert's post-analysis on the preprocessed data and to insert, modify or delete a new rule is not easy. Therefore, there is a need for a system capable of improving customer and merchandise data accumulation methods to facilitate interpretation of data, and to create and apply rules in real time.

Second, there is a difficulty in solving the overlapping between rules. If the amount of the rule becomes too large, it becomes difficult to search for overlapping between rules due to human judgment. For example, a rule A is a rule that sends a promotion message "1" to a woman (if) of 20 or more, and a rule B sends a promotion message "to a woman under 30 In the case of 20 women, you receive the same message twice, which reduces the promotion effectiveness.

Therefore, a system that can modify, delete, and prioritize rules through appropriate inter-rule evaluation is needed.

The present invention provides a marketing promotion system and a method thereof based on a knowledge system capable of generating customer selection rules through a combination of knowledge of a knowledge store and a marketer and realizing them and providing customized promotions to customer data The purpose.

In addition, the present invention can design an information storage (Knowledge Pool) for collecting, storing and managing information of customers as an information collection system capable of analyzing real time information, and can modify, delete, and prioritize rules, A marketing promoting system based on a knowledge system capable of filtering for overlapping applications, and a method thereof.

According to another aspect of the present invention, there is provided a knowledge promoting system for promoting marketing based on a knowledge system, the system comprising: a customer information database for storing customer information, customer information, A knowledge repository to store summary tables for preprocessing of generation; A rule manager for generating a customer selection rule based on the summary table of the knowledge repository; A rule database storing a customer selection rule generated by the rule manager; A rule matching module for matching the summary table information with the customer selection rule data to select a customer who meets the corresponding promotion condition; A rule filtering module for filtering whether the selected customer is redundantly applied to the customer selection rule; And interface means for transmitting the promotion information to the filtered customer, wherein the customer selection rule generated by the rule manager is composed of a condition-result (If-Then) rule, and the customer who is suitable for the promotion in the summary table A result rule for the delivery method and a response level of the rule executed at least once to the customer, and the rule manager determines, based on the response degree, It is characterized by the ability to modify or delete customer selection rules.

Herein, the rule filtering module filters whether the customer selection rule is redundantly applied according to a priority condition of a predetermined customer selection rule, and the priority condition is a response frequency when there are a plurality of the same conditions, The new rules without any response, the order of the highest response, and in the case that there are several customer selection rules, when the results are different, .

The effects of the present invention are as follows.

First, the present invention can create a customer selection rule through a combination of knowledge of a knowledge store and a marketer, and provide customized promotion to customer data by applying the rule in real time.

Second, the present invention enables customers to be selected according to marketing promotion conditions, thereby improving system satisfaction of customers as well as companies performing marketing promotion.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 schematically illustrates a knowledge promoting marketing promotion system according to an embodiment of the present invention; FIG.
FIG. 2 schematically illustrates the structure of information stored in the knowledge repository of FIG. 1; FIG.
3 is a flowchart illustrating a method of marketing promotion based on a knowledge system according to an embodiment of the present invention.

While the present invention has been described in connection with certain exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, is intended to cover various modifications and similarities. It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the accompanying drawings. In addition, numerals (e.g., first, second, etc.) used in the description of the present invention are merely an identifier for distinguishing one component from another.

Also, in this specification, when an element is referred to as being "connected" or "connected" with another element, the element may be directly connected or directly connected to the other element, It should be understood that, unless an opposite description is present, it may be connected or connected via another element in the middle.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a marketing promotion system and method according to the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a diagram schematically showing a configuration of a marketing promotion system based on a knowledge system according to an embodiment of the present invention.

Referring to FIG. 1, the knowledge promotion system based on the knowledge system of the present invention includes a situation information table extracting characteristics of each customer from customer source data and customer source data, and a summary table for pre- A knowledge store 110 for storing a table; A rule manager 120 for generating a customer selection rule based on the summary table of the knowledge repository 110; A rule database 130 for storing a customer selection rule generated by the rule manager 120; A rule matching module 150 for matching the summary table information with the customer selection rule data to select a customer meeting the corresponding promotion condition; A rule filtering module (160) for filtering whether the selected customer is redundantly applied to the customer selection rule; And interface means 170 for transmitting the promotional information to the filtered customer.

FIG. 2 schematically illustrates the structure of information stored in the knowledge repository of FIG. 1; FIG. As shown in FIG. 2, the knowledge repository 110 includes a situation information table 201 and a summary table 202, which are data bases for accumulating data for transforming source data into an interpretable form, .

The situation information table 201 is a table that is rewritten to extract the characteristics of each customer from the source data (log record, etc.).

More specifically, the situation information table 201 includes a profile, which is unique information of the customer, activity information on the behavior characteristics of the customer, and reaction information on the response characteristics of the customer to the marketing activities.

Here, the profile information includes name, age, sex, occupation, etc., which are unique information of a customer, as shown in Table 1 below.

As shown in the following Table 2, the activity information includes information on a behavior characteristic of a customer, such as a purchased product, a product purchase time, a shop stay time, and the like.

As shown in Table 3, the response information includes information on the response characteristics of a customer to a marketing activity, for example, an event response, a questionnaire response, and the like.

The status information may be changed as the customer information is changed. The status manager 140 continuously monitors the status information table 201 of the knowledge repository 110 to determine whether the status information is changed. If there is a change, 150).

The summary table 202 is a table that summarizes the situation information and preprocessed in an interpretable form so that an expert can create a customer selection rule. That is, the summary table may include a summary of user behavior information, a summary of shop visit characteristics of customers, and a cumulative number of purchases by product category according to the standard classification.

      The summary of the user behavior information may include a stay time, a number of items purchased, an event reception count, an event response count, and the like as a table summarizing information of the customer in the store as shown in Table 4 below.

As shown in Table 5, the customer summary-1 is a table summarizing the store visit characteristics of the customer, and may include the total number of store visits, the total sum of product purchases, and the sum of store stay times.

Table 2 summarizes the cumulative number of purchases according to the commodity classification according to the small-sized commodity standard classification as shown in Table 6, and may include, for example, the number of purchases of foods, the number of purchases of non-foods, and the like.

 Based on these summary tables, marketing experts will create customer selection rules. In addition, the summary table summarizes the customer information required for marketing, and it is possible to variously modify the customer information classified according to the marketing type, but is not limited thereto.

The rule manager 120 is a module in which a specialist creates a customer selection rule based on a summary table in the knowledge repository.

More specifically, the customer selection rule consists of a condition rule and a result rule (If-Then), and a condition in which a specialist in the summary table selects a customer best suited to the promotion message is a condition rule (If).

Further, the result rule (Then) is to select a message to be delivered to the customer (text, picture, etc.) corresponding to the condition rule (If) and a result of which channel (SMS, Push Message, DID,

For example, if you have selected customers who are more than 30 female customers, who have more than 3 purchases, more than 2 event responses, and over 30 minutes in the store, Content and delivery method.

At this time, since there is a degree of response to the customer selection rule applied to the promotion at least once, the customer selection rule is modified or deleted based on the response level. That is, it is possible to modify or delete the number of responses to the event, the content of the message, and the delivery method.

The rule database 130 stores rules generated through the rule manager 120. [

More specifically, the condition rule (If), the result rule (Then), and the degree of reaction generated in the rule manager are constructed and stored in the table of Table 7 below. Here, the degree of reaction stores the degree of response to the customer selection rule executed once by the customer, and the rule for the degree of reaction is calculated as shown in Equation 1 below.

Equation 1

는 임의의 해당 프로모션 reaction: degree of reaction,

Any corresponding promotion

The status manager 140 monitors the status information in the knowledge repository 110 in real time and transmits the information to the rule matching module 150 when the status information of the client is changed.

Here, information to be monitored and transmitted to the rule matching module 150 is transmitted as unique ID information and summary table information corresponding to the corresponding customer, as shown in Table 8 below.

The rule matching module 150 selects a customer satisfying the corresponding condition through the matching between the summary table information transmitted from the situation manager 140 and the rule database 130 and transmits the selected customer to the rule filtering module 160 do.

More specifically, information transmitted from the rule matching module 150 to the rule filtering module 160 includes a customer ID, a condition part If of a matched rule, a result classification result of a matched rule, , And the response of the matched rule.

The rule filtering module 160 solves the overlapping application of the customer selection rule received from the rule matching module 150.

In other words, if a customer has the same condition (If) as the result (Then), only one rule is executed according to the following priority order. In other words, new rules with no response level and higher response levels are executed.

In addition, if a customer has several rules but the results are different, the rule is executed according to the following priority order, but the schedule is adjusted to improve the effectiveness. In other words, the new rules that do not have the degree of reaction are executed in the order of input first, in the order of higher response (in case of the same degree of reaction).

The interface unit 170 is divided into a channel interface 170a and a user interface 170b. Here, the channel interface 170a selects a delivery channel such as SMS, Push Message, DID, and the like as a method of delivering a promotion message to a customer, and the user interface 170b selects a promotion message The contents of the promotional message expressed in the text or picture are transmitted to the customer.

3 is a flowchart illustrating a method of marketing promotion based on a knowledge system according to an embodiment of the present invention. As shown in FIG. 3, the customer source data is first extracted and stored as a situation information table (S301).

More specifically, the process of extracting the customer source data from the characteristics of each customer and re-creating and storing them in the status information table monitors the status information table in real time, and when the status information of the customer is changed, To the matching module. Here, the situation information table includes profile information, activity information, and reaction information of a customer. Referring to Table 1, Table 2, and Table 3, the profile information may include information such as name, age, gender, occupation, etc., which is unique information of a customer, and the activity information may include information Purchased product, purchase time of goods, time of staying in a shop, and the like. In addition, the reaction information may include information on a response characteristic of a customer for a marketing activity, for example, an event response, a questionnaire response, and the like.

The re-created situation information table is created and stored in a summary table for preprocessing the customer management rule generation (S302).

More specifically, the summary table is a table that summarizes the situation information and preprocessed in an interpretable form so that an expert can create a customer selection rule. That is, the summary table may include a summary of user behavior information, a summary of shop visit characteristics of customers, and a cumulative number of purchases by product category according to the standard classification.

The summary of the user behavior information may include a stay time, a number of purchased items, an event reception count, an event response count, etc. as a table summarizing the information that the customer has performed in the store.

A table summarizing the store visit characteristics of the customer, and the total sum of the store visit count, the total sum of product purchase, and the sum of the store stay time.

The table summarizes the cumulative number of purchases according to the commodity classification according to the small-sized commodity standard classification, and may include, for example, the number of purchases of the foods, the number of the purchases of the non-foods, and the like.

Based on these summary tables, marketing experts will create customer selection rules. In addition, the summary table summarizes the customer information required for marketing, and it is possible to variously modify the customer information classified according to the marketing type, but is not limited thereto.

Then, the rule manager 120 creates a customer selection rule based on the summary table and stores it in the rule database 130 (S303).

More specifically, the customer selection rule includes a condition rule for selecting a customer suitable for the promotion in the summary table, a content to be delivered to the customer in accordance with the condition rule, and a delivery method The outcome rule for the customer and the response of the rule to the customer at least once.

The condition rule (If) is a condition for selecting the best customer for the corresponding promotion message in the summary table.

Further, the result rule (Then) is to select a message to be delivered to the customer (text, picture, etc.) corresponding to the condition rule (If) and a result of which channel (SMS, Push Message, DID,

For example, if you have selected customers who are more than 30 female customers, who have more than 3 purchases, more than 2 event responses, and over 30 minutes in the store, Content and delivery method.

At this time, since there is a degree of response to the customer selection rule applied to the promotion at least once, the customer selection rule is modified or deleted based on the response level. That is, it is possible to modify or delete the number of responses to the event, the content of the message, and the delivery method.

Here, the generated customer management rule is stored in the rule database 130. The condition rule (If), result rule (Then), and reaction degree generated in the rule manager are configured and stored in the table of Table 7 above. Here, the degree of the reaction stores the degree of the reaction to the customer selection rule executed even once by the customer.

Then, the summary table information and the stored customer selection rule data are matched with each other to select a customer satisfying the condition (S304).

More specifically, the rule matching module 150 selects a customer meeting the corresponding condition through matching between the summary table information transmitted from the situation manager 140 and the rule database 130. Here, the rule matching module 150 selects the customer ID, the condition part If of the matched rule, the result classification of the matched rule, and the reaction degree of the matched rule as shown in Table 9 .

In operation S305, the selected customer is filtered to determine whether the customer selection rule is redundantly applied. That is, the rule filtering module 160 solves the overlapping application of the customer selection rule received from the rule matching module 150.

In other words, if a customer has the same condition (If) as the result (Then), only one rule is executed according to the following priority order. In other words, new rules with no response level and higher response levels are executed.

In addition, if a customer has several rules but the results are different, the rule is executed according to the following priority order, but the schedule is adjusted to improve the effectiveness. In other words, the new rules that do not have the degree of reaction are executed in the order of input first, in the order of higher response (in case of the same degree of reaction).

Finally, the step of transmitting the promotion information to the filtered customer is performed (S306). Here, the channel interface 170a selects a delivery channel such as SMS, Push Message, DID, etc. as a method of delivering a promotion message to a customer, and the user interface 170b selects a delivery channel The contents of the promotional message expressed in the text or picture are transmitted to the customer.

Therefore, the present invention can create a customer selection rule through a combination of knowledge of a knowledge store and a marketer, and apply it in real time to provide customized promotions to customer data.

The scope of the present invention is not limited to the above-described embodiments, but may be embodied in various forms of embodiments within the scope of the appended claims. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention as defined by the appended claims.

110 --- Knowledge Store 120 --- Rules Manager
130 --- Rules database 140 --- Situation manager
150 --- Rule matching module 160 --- Rule filtering module
170a --- Channel interface 170b --- User interface

Claims (2)

A knowledge storage unit for storing a situation information table in which characteristics of each customer are extracted from customer source data and customer source data, and a summary table for pre-processing of generation of a customer selection rule in the situation information table;
A rule manager for generating a customer selection rule based on the summary table of the knowledge repository;
A rule database storing a customer selection rule generated by the rule manager;
A rule matching module for matching the summary table information with the customer selection rule data to select a customer who meets the corresponding promotion condition;
A rule filtering module for filtering whether the selected customer is redundantly applied to the customer selection rule; And
And interface means for transmitting the promotion information to the filtered customer,
The customer selection rule generated by the rule manager includes a condition-result (If-Then) rule. In the summary table, a condition rule for selecting a suitable customer for the corresponding promotion, contents delivered to the customer according to the condition rule, The resulting rules for the method and the response of the rule to the customer at least once,
And the rule manager can modify or delete the customer selection rule based on the degree of reaction. The method according to claim 1,
Wherein the rule filtering module filters whether or not the customer selection rule is redundantly applied according to a priority condition of a predetermined customer selection rule, A new rule, and a higher degree of reaction, and in the case where there are a plurality of customer selection rules, when the results are different, Based marketing promotion system.

Images