Laundry customer marketing system based on big data

Abstract

The invention relates to the technical field of marketing, and discloses a laundry customer marketing system based on big data, which comprises the following steps: firstly, collecting customer information, specifically collecting customer name, contact information, address, gender and age information in a customer registration filling mode, and collecting evaluation of customers on laundry service, environment and price; third-party data acquisition is carried out to obtain customer data provided by a third-party platform, and the customer data comprises calculation of the average consumption amount, consumption frequency and evaluation scores of customers; carrying out data mining, specifically mining and discovering a hidden mode and a rule in customer data; according to requirements, marketing target customer crowds are analyzed from the angles of transaction channels, transaction amounts, transaction frequencies and member identities, and information of target customers is accurately returned and exported. Through the accurate customer portrait and the self-service analysis function, the laundry can accurately identify the target customer crowd and formulate the personalized marketing strategy, the marketing accuracy is improved, and the marketing effect is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of marketing systems, and in particular to a laundry customer marketing system based on big data. Background Art

[0002] Existing laundry customer marketing mainly relies on traditional methods, such as distributing flyers and sending text messages, which lack specificity and cannot effectively reach target customers, resulting in poor marketing results. These methods have the following limitations:

[0003] Flyer marketing: Flyers are expensive and difficult to accurately distribute, which can easily lead to a waste of resources. In addition, flyers often have a monotonous content, making it difficult to meet the needs of different customers and failing to effectively attract customers.

[0004] SMS marketing: SMS marketing can easily cause information overload, and the content of SMS messages cannot be personalized, making it difficult to arouse customer interest.

[0005] Membership management system: Some businesses try to use membership management systems to manage customers, but the functions are relatively simple. They can only record basic customer information and consumption records, and cannot conduct in-depth data analysis and precision marketing.

[0006] With the development of internet technology, some laundries have begun to leverage online platforms for customer marketing, such as establishing WeChat official accounts and developing apps. These methods can expand the laundry's reach and facilitate online reservations and payment. However, these methods still present the following problems:

[0007] Single marketing approach: Traditional marketing methods lack personalization and cannot meet the needs of different customers. For example, frequent customers need to be offered more discounts and services; price-sensitive customers need to be offered more competitive prices; and customers with high service requirements need to be provided with better services.

[0008] High marketing costs and low efficiency: Traditional marketing methods are costly, inefficient, and difficult to evaluate. For example, distributing flyers requires printing and labor costs, and it's difficult to measure distribution volume and return rates. Bulk text messaging requires SMS fees and can easily lead to information overload, making it difficult to capture customer interest.

[0009] Inability to accurately identify customer characteristics: Precision marketing is difficult due to the inability to accurately identify customer characteristics. For example, it is impossible to distinguish between high-value customers, potential churn customers, price-sensitive customers, etc., making it impossible to develop personalized marketing plans based on the needs of different customers.

[0010] Limited functions of member management system: the existing member management system has limited functions and cannot meet the needs of in-depth data analysis and precise marketing. For example, it cannot conduct customer portrait construction, customer behavior analysis, and personalized marketing. Therefore, the existing laundry customer marketing method cannot meet the needs of modern customers, and laundries need a more precise and efficient marketing method to improve customer satisfaction and loyalty.

[0011] Therefore, in view of the above problems, a laundry customer marketing system based on big data is needed. SUMMARY

[0012] The laundry customer marketing system based on big data is provided. Through precise customer portrait and self-service analysis function, the laundry can accurately identify target customer groups and develop personalized marketing strategies to improve marketing accuracy and effectiveness. Reduce marketing costs: through precise marketing, laundries can avoid blind advertising and reduce marketing costs to improve marketing efficiency.

[0013] The laundry customer marketing system based on big data is provided. Through precise customer portrait and self-service analysis function, the laundry can accurately identify target customer groups and develop personalized marketing strategies to improve marketing accuracy and effectiveness. Reduce marketing costs: through precise marketing, laundries can avoid blind advertising and reduce marketing costs to improve marketing efficiency.

[0014] The laundry customer marketing method based on big data is provided. The method is executed according to the following steps:

[0015] S1: Collect customer information, specifically collect customer name, contact information, address, gender, and age information through customer registration and filling;

[0016] S2: Collect consumption records, record customer consumption time, consumption amount, consumption items, and consumption frequency information; understand customer consumption habits and preferences.

[0017] S3: Collect evaluation data, collect customer evaluation of laundry service, environment, and price; understand customer satisfaction and needs;

[0018] S4: Collect third-party data, obtain customer data provided by third-party platforms, including social media data and geographic location data, and enrich customer portrait;

[0019] S5: Statistically analyze customer data, including calculating customer average consumption amount, consumption frequency, and evaluation score; understand customer consumption behavior and preferences;

[0020] Further data mining is performed to discover hidden patterns and rules in customer data; for example, discover which customers are more likely to purchase which goods and which customers are more likely to lose;

[0021] The method is executed according to the following steps:

[0022] S5.1: First, clean the collected data, including name, contact information, address, gender, age, consumption time, consumption amount, consumption items, consumption frequency, customer reviews of the laundry service, environment, and price, social media data, and geographic location data;

[0023] Use the average method to delete records containing invalid values ​​or retain but mark them as invalid data, use the Z score to identify and process outliers, and choose to retain, correct or delete outliers; as shown below;

[0024]

[0025] Where X is the data point, μ is the mean of the data, and σ is the standard deviation of the data;

[0026] If the outliers represent real business conditions, they are retained; for example, large purchases by high-end consumers; if the outliers are caused by data entry errors, they are manually corrected or deleted.

[0027] Data cleaning specifically includes using data processing tools to perform deduplication and remove duplicate records based on fields;

[0028] Process invalid values ​​and identify invalid values, including NULL, missing NaN, and values ​​that do not conform to the format;

[0029] S5.2: Calculate the average of the processed data. For the consumption amount, use the arithmetic mean to calculate the average consumption amount of all customers; for the evaluation score, also use the arithmetic mean to calculate the average evaluation score; as shown below:

[0030]

[0031] Where xi is the consumption amount or evaluation score of the i-th customer, and n is the total number of customers;

[0032] S5.3: Calculate consumption frequency. Count the number of times each customer consumes within a certain period of time and calculate the consumption frequency as follows;

[0033]

[0034] Among them, fi is the consumption frequency of the i-th customer, and ci is the number of times the i-th customer consumes within a certain period of time T.

[0035] Conduct data mining to uncover hidden patterns and regularities in customer data, including:

[0036] First, we set the support and confidence thresholds. The support includes the frequency of an item set appearing in the total transactions; the confidence is the probability that a transaction containing item set X also contains item set Y.

[0037] Then generate frequent itemsets, use Apriori algorithm to generate frequent itemsets, and use Apriori algorithm to iteratively generate candidate itemsets and prune infrequent itemsets;

[0038] Then, the FP-Growth algorithm is used to mine frequent item sets by constructing a frequent pattern tree FP-Tree. When generating frequent item sets, the data set is first scanned, the number of occurrences of each item or service in the laundry is counted, and all items with support greater than or equal to the threshold are found to form a frequent 1-item set. Frequent k-item sets (k>1) are generated iteratively: the frequent k-1 item set is used to generate a candidate k-item set, and then the data set is scanned to calculate the support of the candidate item set. The support is calculated as follows:

[0039]

[0040] And prune the non-frequent itemsets, retain the itemsets whose support is greater than or equal to the threshold, form frequent k-itemsets, and repeat the above steps until no new frequent itemsets can be generated.

[0041] Generate association rules. First, generate candidate association rules from frequent item sets. For each frequent item set X, divide it into two non-empty subsets A and B (A∪B=X), forming candidate association rules A→B.

[0042] Then calculate the confidence level. For each candidate association rule A→B, calculate its confidence level as follows:

[0043]

[0044] Filter association rules, specifically according to the set confidence threshold, filter out meaningful association rules, and the rules with confidence greater than or equal to the threshold are considered to be valid association rules.

[0045] Collect transaction records of laundry customers, including customer ID, transaction time, and purchased goods or services (dry cleaning services, ironing services, specific types of clothing, etc.).

[0046] Set support and confidence thresholds: Set reasonable support and confidence thresholds based on the laundry's marketing goals and business needs.

[0047] Apply Apriori or FP-Growth algorithm: Based on the collected data and the set threshold, use Apriori or FP-Growth algorithm to generate frequent itemsets and association rules.

[0048] Analyze association rules: Analyze the generated association rules to find patterns and trends in customer purchasing behavior and provide a basis for formulating marketing strategies.

[0049] S6: Based on demand, analyze the target customer groups from the perspectives of transaction channels, transaction amounts, transaction frequencies, and membership status, and accurately return and export the target customer information to facilitate precision marketing for customers.

[0050] To analyze customer identity and behavior, we use the K-means algorithm to iteratively cluster customer data until the cluster center no longer changes significantly or the preset number of iterations is reached. Ultimately, customers are divided into K different categories, and customers within each category have similar consumption behaviors and preferences. The specific formula is as follows:

[0051]

[0052] Where n is the dimension of the sample point (i.e., the number of features), xik and xjk are the values ​​of sample points xi and xj in the kth dimension respectively;

[0053]

[0054] Among them, Sm is the set of sample points in the mth cluster, |Sm| is the number of sample points in the set Sm, and xi is the sample point in the set Sm;

[0055]

[0056] Where K is the number of clusters, Cm is the center of the mth cluster, and d((xi,Cm)) is the distance from the sample point xi to the cluster center Cm.

[0057] Furthermore, the frequent item sets are mined by constructing a frequent pattern tree FP-Tree through the FP-Growth algorithm;

[0058] First, set the support threshold and then build the FP-Tree; scan the data set, count the number of occurrences of each item, find the frequent 1-item sets, sort the frequent 1-item sets in descending order of support, and build the root node of the FP-Tree;

[0059] For each transaction in the data set, the FP-Tree is updated in the sorted order of the itemsets; frequent itemsets are mined recursively starting from the root node of the FP-Tree;

[0060] For each node, if its support is greater than or equal to the threshold, it is a frequent item; and continue to recursively mine larger frequent item sets in the conditional pattern base of the node.

[0061] Furthermore, the present invention provides a laundry customer marketing system based on big data, including a data collection and integration module for collecting transaction records, membership information, and feedback data of laundry customers; and integrating and processing them to form a data set that can be used for analysis. The data sources include: POS system, membership management system, online reservation platform, etc.

[0062] Customer segmentation module: segment customers based on transaction channels, transaction amounts, transaction frequencies, and membership characteristics;

[0063] Output: Form multiple customer groups with similar consumption behaviors and preferences.

[0064] Application: Develop targeted marketing strategies for different customer groups.

[0065] Marketing strategy formulation module: Develop personalized marketing strategies based on customer segmentation results, combined with the laundry store's business goals and market environment;

[0066] Strategy types: membership discounts, channel promotions, new product promotions, seasonal promotions, etc. Output: Specific marketing plans, including event themes, discount content, promotion channels, etc.

[0067] Marketing execution and monitoring module, which executes marketing strategies and tracks the effectiveness of marketing activities in real time through monitoring tools;

[0068] Implementation method: Promotion through multiple channels such as email, SMS, social media, in-store advertising, etc.

[0069] Customer feedback and analysis module, which collects customer feedback on marketing activities and conducts sentiment analysis and satisfaction evaluation;

[0070] Feedback channels: online surveys, social media reviews, in-store feedback, etc.

[0071] Output: Customer feedback report to improve marketing strategies and service quality.

[0072] The data visualization and reporting module intuitively displays key data and analysis results in the form of charts and reports.

[0073] Display content: Customer segmentation results, marketing strategy effectiveness, customer feedback trends, etc. Application: Provide decision support for management and optimize business processes.

[0074] The system security and authority management module ensures the security and integrity of system data and prevents unauthorized access and modification.

[0075] Security measures: data encryption, access control, logging, etc.

[0076] Permission management: Assign different system access permissions to different user roles.

[0077] Furthermore, the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of a laundry customer marketing method based on big data as described in any one of the above items are performed.

[0078] Furthermore, the present invention provides a computer-storable medium, which includes an embedded processing system and a stored program, and controls any one of the above-mentioned laundry customer marketing methods based on big data when the embedded system control program runs.

[0079] Compared with the prior art, the present invention has the following beneficial effects:

[0080] 1. Improve marketing accuracy: Through accurate customer portraits and self-service analysis functions, laundries can accurately identify target customer groups and formulate personalized marketing strategies to improve marketing accuracy and enhance marketing effectiveness.

[0081] 2. Reduce marketing costs: Through precision marketing, laundries can avoid blind advertising, reduce marketing costs and improve marketing efficiency.

[0082] 3. Improve customer satisfaction: Through personalized services and precision marketing, laundries can improve customer satisfaction and loyalty, increase customer stickiness, and increase customer repurchase rates.

[0083] 4. Improve operational efficiency: Through customer data analysis and self-service analysis functions, laundries can optimize operational processes, improve operational efficiency, and reduce operating costs.

[0084] 5. Enhance brand image: Through precision marketing and high-quality services, laundries can enhance their brand image and strengthen their market competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. It is understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0086] Figure 1 is a flow chart of the method of the present invention;

[0087] Figure 2 It is a screenshot of the system operation interface of the present invention;

[0088] Figure 3 It is a screenshot of the customer recommendation result of the present invention;

[0089] Figure 4 The code snippet of the present invention Figure 1 ;

[0090] Figure 5 The code snippet of the present invention Figure 2 ;

[0091] Figure 6 The code snippet of the present invention Figure 3 ;

[0092] Figure 7 The code snippet of the present invention Figure 4 ;

[0093] Figure 8 The code snippet of the present invention Figure 5 . DETAILED DESCRIPTION

[0094] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention for which protection is sought, but is merely for selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0095] See also Figures 1-8 ,The present invention provides a laundry customer marketing method based on big data;

[0096] Follow these steps:

[0097] S1: Collect customer information, specifically by collecting customer name, contact information, address, gender, and age information through customer registration;

[0098] S2: Collect consumption records, record customer consumption time, consumption amount, consumption items, consumption frequency information; understand customer consumption habits and preferences.

[0099] S3: Collect evaluation data to collect customers' comments on the laundry service, environment, and price; understand customers' satisfaction and needs;

[0100] S4: Collect third-party data, including social media data and geographic location data, to enrich the customer profile;

[0101] S5: Perform statistical analysis on customer data, including calculating the average consumption amount, consumption frequency, and evaluation score of customers, to understand their consumption behavior and preferences;

[0102] Further data mining is performed to discover hidden patterns and rules in customer data, such as identifying which customers are more likely to purchase certain products or which customers are more likely to churn;

[0103] The specific steps are as follows:

[0104] S5.1: First, perform data cleaning on the collected data, including name, contact information, address, gender, age information, consumption time, consumption amount, consumption items, consumption frequency information, customer evaluation of laundry service, environment, price, social media data, and geographic location data;

[0105] Use the mean method to delete records containing invalid values or retain but mark as invalid data, use Z-score to identify and handle outliers, and choose to retain, correct, or delete outliers; as follows:

[0106]

[0107] where X is the data point, μ is the mean of the data, and σ is the standard deviation of the data;

[0108] If the outlier represents a real business situation, retain the outlier; for example, high-end consumers with large consumption; if the outlier is caused by data entry errors, manually correct or delete.

[0109] Data cleaning specifically includes using data processing tools to remove duplicates, and removing duplicate records according to fields;

[0110] Handle invalid values, identify invalid values, including NULL, NaN, and values that do not conform to the format;

[0111] S5.2: Calculate the mean of the processed data, calculate the average consumption amount of all customers using the arithmetic mean for consumption amount, and calculate the average evaluation score using the arithmetic mean for evaluation score; as follows:

[0112]

[0113] where xi is the consumption amount or evaluation score of the i-th customer, and n is the total number of customers;

[0114] S5.3: Calculate consumption frequency. Count the number of times each customer consumes within a certain period of time and calculate the consumption frequency as follows;

[0115]

[0116] Among them, fi is the consumption frequency of the i-th customer, and ci is the number of times the i-th customer consumes within a certain period of time T.

[0117] Conduct data mining to uncover hidden patterns and regularities in customer data, including:

[0118] First, we set the support and confidence thresholds. The support includes the frequency of an item set appearing in the total transactions; the confidence is the probability that a transaction containing item set X also contains item set Y.

[0119] Then generate frequent itemsets, use Apriori algorithm to generate frequent itemsets, and use Apriori algorithm to iteratively generate candidate itemsets and prune infrequent itemsets;

[0120] Then, the FP-Growth algorithm is used to mine frequent item sets by constructing a frequent pattern tree FP-Tree. When generating frequent item sets, the data set is first scanned, the number of occurrences of each item or service in the laundry is counted, and all items with support greater than or equal to the threshold are found to form a frequent 1-item set. Frequent k-item sets (k>1) are generated iteratively: the frequent k-1 item set is used to generate a candidate k-item set, and then the data set is scanned to calculate the support of the candidate item set. The support is calculated as follows:

[0121]

[0122] And prune the non-frequent itemsets, retain the itemsets whose support is greater than or equal to the threshold, form frequent k-itemsets, and repeat the above steps until no new frequent itemsets can be generated.

[0123] Generate association rules. First, generate candidate association rules from frequent item sets. For each frequent item set X, divide it into two non-empty subsets A and B (A∪B=X), forming candidate association rules A→B.

[0124] Then calculate the confidence level. For each candidate association rule A→B, calculate its confidence level as follows:

[0125]

[0126] Filter association rules, specifically according to the set confidence threshold, filter out meaningful association rules, and the rules with confidence greater than or equal to the threshold are considered to be valid association rules.

[0127] Collect transaction records of laundry customers, including customer ID, transaction time, and purchased goods or services (dry cleaning services, ironing services, specific types of clothing, etc.).

[0128] Set support and confidence thresholds: Set reasonable support and confidence thresholds based on the laundry's marketing goals and business needs.

[0129] Apply Apriori or FP-Growth algorithm: Based on the collected data and the set threshold, use Apriori or FP-Growth algorithm to generate frequent itemsets and association rules.

[0130] Analyze association rules: Analyze the generated association rules to find patterns and trends in customer purchasing behavior and provide a basis for formulating marketing strategies.

[0131] S6: Based on demand, analyze the target customer groups from the perspectives of transaction channels, transaction amounts, transaction frequencies, and membership status, and accurately return and export the target customer information to facilitate precision marketing for customers.

[0132] To analyze customer identity and behavior, we use the K-means algorithm to iteratively cluster customer data until the cluster center no longer changes significantly or the preset number of iterations is reached. Ultimately, customers are divided into K different categories, and customers within each category have similar consumption behaviors and preferences. The specific formula is as follows:

[0133]

[0134] Where n is the dimension of the sample point (i.e., the number of features), xik and xjk are the values ​​of the sample points xi and xj in the kth dimension respectively;

[0135]

[0136] Among them, Sm is the set of sample points in the mth cluster, |Sm| is the number of sample points in the set Sm, and xi is the sample point in the set Sm;

[0137]

[0138] Where K is the number of clusters, Cm is the center of the mth cluster, and d((xi,Cm)) is the distance from the sample point xi to the cluster center Cm.

[0139] In this embodiment, frequent item sets are mined by constructing a frequent pattern tree FP-Tree through the FP-Growth algorithm;

[0140] First, set the support threshold and then build the FP-Tree; scan the data set, count the number of occurrences of each item, find the frequent 1-item sets, sort the frequent 1-item sets in descending order of support, and build the root node of the FP-Tree;

[0141] For each transaction in the data set, the FP-Tree is updated in the sorted order of the itemsets; frequent itemsets are mined recursively starting from the root node of the FP-Tree;

[0142] For each node, if its support is greater than or equal to the threshold, it is a frequent item; and continue to recursively mine larger frequent item sets in the conditional pattern base of the node.

[0143] In this embodiment, the present invention provides a laundry customer marketing system based on big data, including a data collection and integration module for collecting transaction records, membership information, and feedback data of laundry customers; and integrating and processing them to form a data set that can be used for analysis. The data sources include: POS system, membership management system, online reservation platform, etc.

[0144] Customer segmentation module: segment customers based on transaction channels, transaction amounts, transaction frequencies, and membership characteristics;

[0145] Output: Form multiple customer groups with similar consumption behaviors and preferences.

[0146] Application: Develop targeted marketing strategies for different customer groups.

[0147] Marketing strategy formulation module: Develop personalized marketing strategies based on customer segmentation results, combined with the laundry store's business goals and market environment;

[0148] Strategy types: membership discounts, channel promotions, new product promotions, seasonal promotions, etc. Output: Specific marketing plans, including event themes, discount content, promotion channels, etc.

[0149] Marketing execution and monitoring module, which executes marketing strategies and tracks the effectiveness of marketing activities in real time through monitoring tools;

[0150] Implementation method: Promotion through multiple channels such as email, SMS, social media, in-store advertising, etc.

[0151] Customer feedback and analysis module, which collects customer feedback on marketing activities and conducts sentiment analysis and satisfaction evaluation;

[0152] Feedback channels: online surveys, social media reviews, in-store feedback, etc.

[0153] Output: Customer feedback report to improve marketing strategies and service quality.

[0154] The data visualization and reporting module intuitively displays key data and analysis results in the form of charts and reports.

[0155] Display content: Customer segmentation results, marketing strategy effectiveness, customer feedback trends, etc. Application: Provide decision support for management and optimize business processes.

[0156] The system security and authority management module ensures the security and integrity of system data and prevents unauthorized access and modification.

[0157] Security measures: data encryption, access control, logging, etc.

[0158] Permission management: Assign different system access permissions to different user roles.

[0159] In the present embodiment, Example 1:

[0160] Customers of Store A often use the shoe cleaning service in the summer.

[0161] Based on their consumption patterns, the system used self-service tools to accurately screen customers who made purchases last summer but hadn't made any purchases this year, and then sent them coupons. After receiving the coupons, the customers went to the laundry to have their shoes cleaned and shared their experiences, boosting customer satisfaction.

[0162] Example 2:

[0163] Customer B has not visited the laundry in the past month.

[0164] Based on the RFM model analysis of customer B, the system determined that he was a potential churned customer and sent a coupon to customer B.

[0165] After receiving the coupon, customer B went to the laundry again to shop, thus avoiding customer loss.

[0166] In this embodiment, the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of a laundry customer marketing method based on big data as described in any one of the above items are performed.

[0167] In this embodiment, the present invention provides a computer-storable medium, which includes an embedded processing system and a stored program, and controls any one of the above-mentioned laundry customer marketing methods based on big data when the embedded system control program runs.

[0168] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A laundry customer marketing method based on big data, characterized by: Follow these steps: S1: Collect customer information, specifically by collecting customer name, contact information, address, gender, and age information through customer registration; S2: Collect consumption records, record the customer's consumption time, consumption amount, consumption items, and consumption frequency information; S3: Collect evaluation data and collect customers' evaluations of the laundry service, environment, and price; S4: Conduct third-party data collection to obtain customer data provided by third-party platforms, including social media data and geographic location data, to enrich customer portraits; S5: Conduct statistical analysis on customer data, including calculating the average spending amount, consumption frequency, and evaluation scores of customers; Then conduct data mining to discover hidden patterns and regularities in customer data; S6: Based on demand, analyze the target customer groups from the perspectives of transaction channels, transaction amounts, transaction frequencies, and membership status, and accurately return and export the target customer information.

2. The laundry customer marketing method based on big data according to claim 1, characterized in that: In step S5, the following steps are specifically performed: S5.1: First, clean the collected data, including name, contact information, address, gender, age, consumption time, consumption amount, consumption items, consumption frequency, customer reviews of the laundry service, environment, and price, social media data, and geographic location data; Data cleaning specifically includes using data processing tools to perform deduplication and remove duplicate records based on fields; Process invalid values ​​and identify invalid values, including NULL, missing NaN, and values ​​that do not conform to the format; S5.2: Calculate the average of the processed data. For the consumption amount, use the arithmetic mean to calculate the average consumption amount of all customers; for the evaluation score, also use the arithmetic mean to calculate the average evaluation score; as shown below: Where xi is the consumption amount or evaluation score of the i-th customer, and n is the total number of customers; S5.3: Calculate consumption frequency. Count the number of times each customer consumes within a certain period of time and calculate the consumption frequency as follows; Among them, fi is the consumption frequency of the i-th customer, and ci is the number of times the i-th customer consumes within a certain period of time T.

3. The laundry customer marketing method based on big data according to claim 1, characterized in that: Conduct data mining to uncover hidden patterns and regularities in customer data, including: First, we set the support and confidence thresholds. The support includes the frequency of an item set appearing in the total transactions; the confidence is the probability that a transaction containing item set X also contains item set Y. Then generate frequent item sets, use the Apriori algorithm to generate frequent item sets, and use the Apriori algorithm to iteratively generate candidate item sets and prune infrequent item sets; then use the FP-Growth algorithm to mine frequent item sets by constructing a frequent pattern tree FP-Tree; Generate association rules. First, generate candidate association rules from frequent item sets. For each frequent item set X, divide it into two non-empty subsets A and B (A∪B=X), forming candidate association rules A→B. Then calculate the confidence level. For each candidate association rule A→B, calculate its confidence level as follows: Filter association rules, specifically according to the set confidence threshold, filter out meaningful association rules, and the rules with confidence greater than or equal to the threshold are considered to be valid association rules.

4. The laundry customer marketing method based on big data according to claim 3, characterized in that: When generating frequent itemsets, first scan the data set, count the number of occurrences of each item or service in the laundry, find all items with support greater than or equal to the threshold, and form a frequent 1-item set; iteratively generate frequent k-item sets (k>1): use the frequent k-1 item set to generate the candidate k-item set, then scan the data set again and calculate the support of the candidate item set; the support is calculated as follows; And prune the non-frequent itemsets, retain the itemsets whose support is greater than or equal to the threshold, form frequent k-itemsets, and repeat the above steps until no new frequent itemsets can be generated.

5. The laundry customer marketing method based on big data according to claim 3, characterized in that: Mining frequent item sets by constructing a frequent pattern tree FP-Tree through FP-Growth algorithm; First, set the support threshold and then build the FP-Tree; scan the data set, count the number of occurrences of each item, find the frequent 1-item sets, sort the frequent 1-item sets in descending order of support, and build the root node of the FP-Tree; For each transaction in the data set, the FP-Tree is updated in the sorted order of the itemsets; frequent itemsets are mined recursively starting from the root node of the FP-Tree; For each node, if its support is greater than or equal to the threshold, it is a frequent item; and continue to recursively mine larger frequent item sets in the conditional pattern base of the node.

6. The laundry customer marketing system based on big data according to claim 1, characterized in that: In step S5.1, use the average method to delete records containing invalid values ​​or retain but mark them as invalid data, use the Z score to identify and process outliers, and choose to retain, correct, or delete outliers; as shown below; Where X is the data point, μ is the mean of the data, and σ is the standard deviation of the data; If the outlier represents the actual business situation, it is retained; if the outlier is caused by a data entry error, it is manually corrected or deleted.

7. The laundry customer marketing system based on big data according to claim 1, characterized in that: In step S6, customer identity and customer behavior are analyzed by iteratively clustering the customer data using the K-means algorithm until the cluster center no longer changes significantly or a preset number of iterations is reached. Ultimately, the customers are divided into K different categories, and customers within each category have similar consumption behaviors and preferences. Specifically, the K-means algorithm is used as follows: Where n is the dimension of the sample point (i.e., the number of features), xik and xjk are the values ​​of sample points xi and xj in the kth dimension respectively; Among them, Sm is the set of sample points in the mth cluster, |Sm| is the number of sample points in the set Sm, and xi is the sample point in the set Sm; Where K is the number of clusters, Cm is the center of the mth cluster, and d((xi,Cm)) is the distance from the sample point xi to the cluster center Cm.

8. A laundry customer marketing system based on big data, characterized by: It includes a data collection and integration module to collect transaction records, membership information, and feedback data of laundry customers; Customer segmentation module: segment customers based on transaction channels, transaction amounts, transaction frequencies, and membership characteristics; Marketing strategy formulation module: Develop personalized marketing strategies based on customer segmentation results, combined with the laundry store's business goals and market environment; Marketing execution and monitoring module, which executes marketing strategies and tracks the effectiveness of marketing activities in real time through monitoring tools; Customer feedback and analysis module, which collects customer feedback on marketing activities and conducts sentiment analysis and satisfaction evaluation; Data visualization and reporting module, which intuitively displays key data and analysis results in the form of charts and reports; The system security and authority management module ensures the security and integrity of system data and prevents unauthorized access and modification.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the laundry customer marketing method based on big data described in any one of claims 1 to 7 are implemented.

10. A computer storable medium, characterized in that: The computer-readable storage medium includes an embedded processing system and a stored program, and when the embedded system control program runs, it controls the execution of the laundry customer marketing method based on big data according to any one of claims 1 to 7.

Images