Enterprise digital operation management system based on artificial intelligence

Abstract

The invention, which relates to the technical field of digital management, discloses an artificial intelligence-based enterprise digital operation management system comprising a data integration module, a behavior prediction module, a feature recommendation module, a clustering subdivision module and a decision support module. The data integration module collects and arranges original data and transmits the original data to the behavior prediction module and the feature recommendation module, the behavior prediction module constructs a prediction model according to an artificial intelligence algorithm, analyzes customer behaviors and purchase modes and predicts customer demands and commercial intentions, and the feature recommendation module performs feature recommendation according to a collaborative filtering algorithm. The service recommendation module is used for providing service recommendation according to customer characteristics, the clustering subdivision module is used for subdividing customer groups according to a clustering analysis algorithm, and the decision support module is used for grading service grades according to customer group types. Transfer development of enterprises in the B2B business field and the B2C business field is met, and the enterprises are prevented from falling into data islands and information isolation.

Description

Technical Field

[0001] The present invention relates to the field of digital management technology, and more specifically, to an enterprise digital operation management system based on artificial intelligence. Background Art

[0002] When conducting enterprise digital management, the demand characteristics of B2B business and B2C business are different. B2B sales are usually complex, involving multiple steps and multiple buyers. The sales cycle may take years. Compared with typical B2C sales, B2B sales have larger transaction scales and lower frequencies. B2B products are not universally needed, so there are few potential customers. The B2C sales environment is not as complex as the B2B sales environment. The entire B2C sales process is usually measured in days, and there are fewer touch points in the sales process. Although the transaction scale is smaller, the number of products purchased each time is large. The audience and customer segmentation of B2C sales are not as narrow as B2B sales. Enterprises need to focus on a wide range of audiences to conduct large-scale and rapid sales. The target audience managed by the customer relationship management system is wider;

[0003] Under pressure from the rapidly changing market environment, companies that were originally oriented towards B2B business may turn to the B2C market, while companies that were originally oriented towards B2C business may have to take into account the B2B market. It is unrealistic to maintain two different customer relationship management systems at the same time, and a priori judgments on business models may not be accurate. Therefore, integrating and optimizing customer relationship management systems, using artificial intelligence technology to solve repetitive basic operations, and directly conducting characteristic analysis of business individuals themselves are effective ways to reduce labor costs.

[0004] In order to solve the above defects, a technical solution is now proposed. Summary of the invention

[0005] The purpose of the present invention is to provide an enterprise digital operation management system based on artificial intelligence to solve the problems in the above-mentioned background technology.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an enterprise digital operation management system based on artificial intelligence, including a data integration module, a behavior prediction module, a feature recommendation module, a clustering and segmentation module, and a decision support module;

[0007] The data integration module is used to collect and organize the original data, and transmit the organized original data to the behavior prediction module and the feature recommendation module;

[0008] The behavior prediction module is used to build prediction models based on artificial intelligence algorithms, analyze customer behavior and purchasing patterns, and predict customer needs and business intentions;

[0009] The feature recommendation module is used to provide service recommendations based on customer characteristics according to collaborative filtering algorithms;

[0010] The clustering segmentation module is used to segment customer groups based on clustering analysis algorithms;

[0011] The decision support module is used to rate the service level according to the type of customer group.

[0012] Preferably, the method for constructing a prediction model according to an artificial intelligence algorithm is:

[0013] Collect customer information, interaction data, sales data, and usage data, remove duplicate data, process missing values ​​and outliers, and normalize or standardize the collected data. Customer information includes the number of employees Nu and the tax amount Ta. Interaction data includes the number of email interactions En, the number of phone records Tn, and the number of meeting records Mn. Sales data includes the number of purchases Pn, the contract amount Ac, and the purchase cycle Bc. Usage data includes the number of product uses Uc and the number of fault feedback Fn.

[0014] Extract features from the collected data, divide the data into training set and test set, use the random forest algorithm to set initial hyperparameters, and use the training set to train the random forest model;

[0015] Use the test set to evaluate model performance, calculate the F1 score of the random forest model, and test the stability of the random forest model through cross-validation;

[0016] Perform hyperparameter tuning to adjust the number of trees and maximum depth parameters;

[0017] The optimized random forest model is deployed to predict customers’ purchase intentions.

[0018] Preferably, in extracting features based on collected data, the feature selection method is:

[0019] The characteristics include scale weight coefficient, interaction coefficient, consumption tendency coefficient, and usage feedback index. The scale weight coefficient is Sc, the interaction coefficient is In, the consumption tendency coefficient is Ct, and the usage feedback coefficient is Tf. The calculation expression of the scale weight coefficient is Sc=Nu*Ta, the calculation expression of the interaction coefficient is In=Mn*(En+Tn), and the calculation expression of the consumption tendency coefficient is The calculation expression using the feedback coefficient is

[0020] Preferably, the method for recommending services using collaborative filtering algorithm is:

[0021] Collect user behavior data on service ratings, browsing times, and purchase times;

[0022] Evaluate service characteristics based on user behavior data on the service;

[0023] Based on the service feature evaluation, similarity algorithms are used to calculate the similarity between users;

[0024] For each user, find several similar users with the highest similarity to the user;

[0025] Recommend services selected by similar users to target users based on their behaviors;

[0026] The definition logic of service feature evaluation is:

[0027] Service feature evaluation = score × (number of purchases / number of views).

[0028] Preferably, based on the service feature evaluation, the logic of using the similarity algorithm to calculate the similarity between users is:

[0029] Let s be the service number, and s = {1, 2, 3…n}, where n is a positive integer. Calculate the Pearson correlation coefficient between users. The calculation expression is: In the formula, A n , B n is the evaluation of service characteristics of service n by user A and user B, is the average service feature evaluation of all purchased services by user A, is the mean service feature evaluation of all purchased services by user B.

[0030] Preferably, the clustering and segmentation module is used to segment the customer groups according to the clustering analysis algorithm;

[0031] Collect the customer's number of employees, annual purchase volume, years of consumption, income, and historical consumption amount. According to the K-means algorithm, determine the K value through the elbow method or silhouette coefficient method. Based on the results of cluster analysis, analyze the characteristics of each cluster and classify the customer type.

[0032] Preferably, the logic for analyzing the cluster analysis results is:

[0033] When the K value of K-means clustering is 2, for B2B business customers, if the number of employees and annual purchasing volume of customers in cluster 1 are greater than those in cluster 2, then the customers in cluster 1 are marked as large customers, and the customers in cluster 2 are marked as small and medium-sized customers; for B2C business customers, if the historical consumption amount and income of customers in cluster 1 are greater than those in cluster 2, and the consumption years of customers in cluster 1 are less than those of customers in cluster 2, then the customers in cluster 1 are marked as high-income customers, and the customers in cluster 2 are marked as medium-income customers.

[0034] Preferably, the logic of the decision support module for grading the service level according to the customer group type is:

[0035] When a B2B business customer is marked as a large customer, the service level is set to V1, providing customized solutions and bulk purchase discounts;

[0036] When B2B business customers are marked as small and medium-sized customers, the service level is set to V2, providing flexible procurement plans and support services;

[0037] When B2C business customers are marked as high-income customers, the service level is set to S1, providing recommendations and discounts on high-end fashion and technology products;

[0038] When a B2C business customer is marked as a middle-income customer, the service level is set to S2, providing discount promotions on daily necessities and home services.

[0039] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0040] This application optimizes and expands the digital operations of enterprises by using artificial intelligence technology in combination with customer relationship management systems to meet the company's shift and development in the B2B business field and the B2C business field, and avoids the company using customer relationship management systems for two different business tracks at the same time, resulting in data silos and information isolation barriers. It analyzes and predicts customers of different business types based on artificial intelligence technology to ensure the stability of customer relationships, accurately control the cooperation status with different types of customers, and reduce the energy cost of managers. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0042] Figure 1 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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 creative work are within the scope of protection of the present invention.

[0044] Example 1: Please refer to Figure 1As shown, the present invention is an enterprise digital operation management system based on artificial intelligence, including a data integration module, a behavior prediction module, a feature recommendation module, a clustering segmentation module, and a decision support module;

[0045] The data integration module is used to collect and organize the original data, and transmit the organized original data to the behavior prediction module and the feature recommendation module;

[0046] The behavior prediction module is used to build prediction models based on artificial intelligence algorithms, analyze customer behavior and purchasing patterns, and predict customer needs and business intentions;

[0047] The feature recommendation module is used to provide service recommendations based on customer characteristics according to collaborative filtering algorithms;

[0048] The clustering segmentation module is used to segment customer groups based on clustering analysis algorithms;

[0049] The decision support module is used to rate the service level according to the type of customer group.

[0050] When conducting digital operation management of enterprises, enterprises have different requirements and conditions for customer relationship management of B2B business and B2C business. For B2B business, customers are usually companies or organizations. Data sources include sales records, contracts, customer visit records, industry reports, enterprise public data, etc. Data types include enterprise size, industry classification, financial status, procurement history, business needs, etc. For B2C business, customers are individual consumers. Data sources include purchase records, browsing records, social media interactions, customer feedback, mobile application usage data, etc. Data types include demographic information, purchase history, interest preferences, consumption behavior, etc. Demographic information usually includes age, Gender, income level, etc. The analysis goals for B2B sales and B2C sales are also different. For B2B business, the stratification logic of the analysis goal is to stratify customers based on factors such as enterprise size, purchase volume, and cooperation time, identify key customers and potential customers, and evaluate the total value of customers throughout the cooperation cycle to formulate targeted customer maintenance strategies. The focus of value analysis is on maintaining customer relationships. Sales funnel analysis is usually used to track and optimize each stage from potential customers to final orders, improve sales conversion rates, and predict customer purchasing behavior, demand changes, renewal possibilities, etc. In B2C business, the segmentation logic of the analysis goal is based on consumers' The goal of value analysis is to evaluate the value of customers throughout their life cycle in order to develop retention and recovery strategies, focusing on the analysis of customers' repeated consumption capacity. Shopping basket analysis is usually used to analyze the combination of different goods purchased by consumers to optimize product recommendations and promotion strategies. At the same time, sentiment analysis is used to deeply explore customer feedback and social media interactions to understand customer emotions and satisfaction. In the field of analysis methods, there are also differences between B2B and B2C businesses. For B2B business, association rule analysis is used to identify the purchase associations between different products or services in order to carry out cross-selling or up-selling. For sales, we analyze the business relationships and networks between enterprises through relationship network analysis to discover potential cooperation opportunities. We use trend analysis to track and analyze industry trends and market dynamics to predict changes in customer needs. For B2C business, we recommend personalized products or services based on the consumer's historical behavior and the behavior of similar users based on the collaborative filtering recommendation algorithm. We use the cluster analysis algorithm to divide customers into different groups for targeted marketing and services. We use sentiment analysis and natural language processing technology to analyze customer comments and feedback to understand customer emotional tendencies and needs. We use time series analysis to capture consumers' purchase time patterns and optimize the timing of marketing activities.

[0051] Although the big data analysis of B2B and B2C customer relationship management differs in data sources, analysis objectives and analysis methods, the ultimate goal is the same, that is, to understand customer needs, improve customer satisfaction, increase sales and business growth. Therefore, B2B business data and B2C business data can be integrated and analyzed according to common goals to adapt to the rapidly changing market environment.

[0052] The data integration module is used to collect and organize the original data, and transmit the organized original data to the behavior prediction module and the feature recommendation module;

[0053] Raw data includes sales data, visitor information, financial information, purchase history information, browsing data, and individual information;

[0054] Use data lakes to store raw data, and use data warehouses to clean, transform, and load raw data. Data cleaning is used to identify, correct, or delete errors, duplicates, and inconsistent data items in the data. Data conversion converts data from the source system format to the format of the data warehouse application. Data conversion includes data type conversion, field remapping, data aggregation or decomposition, etc. The cleaned and transformed data is loaded into the data warehouse. The loading process involves data sorting, indexing, and optimization, which can improve query performance.

[0055] When it comes to the combined management of B2B and B2C businesses, it is crucial for enterprises to integrate and manage data. Integrated data management can ensure that data from all data sources is accurate and reduce errors and redundant data. High-quality data can improve the reliability and effectiveness of business decisions. Through data standardization, the data format and structure of different business departments are consistent, avoiding data silos and information inconsistencies. Through integration and management of data, in-depth data analysis can be carried out to explore potential business opportunities and market trends. Data-driven decisions can improve the scientificity and accuracy of business strategies and reduce decision-making risks. Data integration provides a foundation for business process automation and intelligence, and can improve operational efficiency and reduce human intervention and errors through robotic process automation and artificial intelligence technology. Integrated data can help enterprises better manage and optimize resource allocation and avoid resource waste and duplication. Through data analysis, enterprises can optimize inventory management, supply chain operations and marketing strategies to improve overall operational efficiency.

[0056] The behavior prediction module is used to build prediction models based on artificial intelligence algorithms, analyze customer behavior and purchasing patterns, and predict customer needs and business intentions;

[0057] The steps to build a predictive model using artificial intelligence algorithms include:

[0058] Collect customer information, interaction data, sales data, and usage data, remove duplicate data, process missing values ​​and outliers, and normalize or standardize the collected data. Customer information includes the number of employees Nu and the tax amount Ta. Interaction data includes the number of email interactions En, the number of phone records Tn, and the number of meeting records Mn. Sales data includes the number of purchases Pn, the contract amount Ac, and the purchase cycle Bc. Usage data includes the number of product uses Uc and the number of fault feedback Fn.

[0059] Extract features from the collected data, divide the data into training set and test set, use the random forest algorithm to set initial hyperparameters, and use the training set to train the random forest model;

[0060] Use the test set to evaluate model performance, calculate the F1 score of the random forest model, and test the stability of the random forest model through cross-validation;

[0061] Perform hyperparameter tuning to adjust the number of trees and maximum depth parameters;

[0062] The optimized random forest model is deployed to predict customers’ purchase intentions.

[0063] Random forest makes predictions by integrating multiple decision trees. Each decision tree is trained on different random samples and feature subsets. The final prediction result is the consistent result of all trees, which can effectively improve the accuracy and robustness of the model. Moreover, since random forest uses multiple decision trees for prediction, the overfitting risk of a single tree is dispersed. Even if some decision trees may overfit, the overall prediction result can still maintain good generalization ability. Random forest can provide an importance assessment of each feature and show which features contribute most to the prediction result. By using different feature subsets for training, random forest can still maintain good prediction performance when facing missing values, and is more robust when facing data noise and outliers, and is not easily affected by extreme values.

[0064] In extracting features based on collected data, the feature selection method is:

[0065] The characteristics include scale weight coefficient, interaction coefficient, consumption tendency coefficient, and usage feedback index. The scale weight coefficient is Sc, the interaction coefficient is In, the consumption tendency coefficient is Ct, and the usage feedback coefficient is Tf. The calculation expression of the scale weight coefficient is Sc=Nu*Ta, the calculation expression of the interaction coefficient is In=Mn*(En+Tn), and the calculation expression of the consumption tendency coefficient is The calculation expression using the feedback coefficient is

[0066] The feature recommendation module is used to provide service recommendations based on customer characteristics according to collaborative filtering algorithms;

[0067] The method of using collaborative filtering algorithm for service recommendation is:

[0068] Collect user behavior data on service ratings, browsing times, and purchase times;

[0069] Evaluate service characteristics based on user behavior data on the service;

[0070] Based on the service feature evaluation, similarity algorithms are used to calculate the similarity between users;

[0071] For each user, find several similar users with the highest similarity to the user;

[0072] Based on the behaviors of similar users, recommend services selected by similar users to the target user.

[0073] Similarity algorithms include cosine similarity, Pearson correlation coefficient, etc.

[0074] The definition logic of service feature evaluation is:

[0075] Service feature evaluation = score × (number of purchases / number of views);

[0076] The logic of calculating the similarity between users using the Pearson correlation coefficient is:

[0077] Let s be the service number, and s = {1, 2, 3…n}, where n is a positive integer. Calculate the Pearson correlation coefficient between users. The calculation expression is: In the formula, A n , B n is the evaluation of service characteristics of service n by user A and user B, is the average service feature evaluation of all purchased services by user A, is the mean service feature evaluation of all purchased services by user B.

[0078] Using the Pearson correlation coefficient to calculate the similarity between users can effectively measure the similarity of user rating behaviors, thereby providing personalized service recommendations for users in the collaborative filtering algorithm.

[0079] The clustering segmentation module is used to segment customer groups based on clustering analysis algorithms;

[0080] Collect the customer's number of employees, annual purchase volume, years of consumption, income, and historical consumption amount. According to the K-means algorithm, determine the K value through the elbow method or silhouette coefficient method. Based on the results of cluster analysis, analyze the characteristics of each cluster and classify the customer type.

[0081] The decision support module is used to rate the service level according to the type of customer group;

[0082] When the K-means clustering K value is 2, for B2B business customers, if the number of employees and annual purchase volume of customers in cluster 1 are greater than those in cluster 2, then the customers in cluster 1 are marked as large customers, and the customers in cluster 2 are marked as small and medium-sized customers; for B2C business customers, if the historical consumption amount and income of customers in cluster 1 are greater than those in cluster 2, and the consumption years of customers in cluster 1 are less than those of customers in cluster 2, then the customers in cluster 1 are marked as high-income customers, and the customers in cluster 2 are marked as medium-income customers;

[0083] When a B2B business customer is marked as a large customer, the service level is set to V1, providing customized solutions and bulk purchase discounts;

[0084] When B2B business customers are marked as small and medium-sized customers, the service level is set to V2, providing flexible procurement plans and support services;

[0085] When B2C business customers are marked as high-income customers, the service level is set to S1, providing recommendations and discounts on high-end fashion and technology products;

[0086] When a B2C business customer is marked as a middle-income customer, the service level is set to S2, providing discount promotions on daily necessities and home services.

[0087] This application optimizes and expands the digital operations of enterprises by using artificial intelligence technology in combination with customer relationship management systems to meet the company's shift and development in the B2B business field and the B2C business field, and avoids the company using customer relationship management systems for two different business tracks at the same time, resulting in data silos and information isolation barriers. It analyzes and predicts customers of different business types based on artificial intelligence technology to ensure the stability of customer relationships, accurately control the cooperation status with different types of customers, and reduce the energy cost of managers.

[0088] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0089] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of computer program goods. The computer program goods include one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0090] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0091] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0092] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0093] If the functions are implemented in the form of software functional units and sold or used as independent goods, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can essentially or in other words, the part that contributes to the prior art or the part of the technical solution can be embodied in the form of software goods, and the computer software goods are stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0094] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. The enterprise digital operation management system based on artificial intelligence is characterized by: It includes data integration module, behavior prediction module, feature recommendation module, clustering and segmentation module, and decision support module; The data integration module is used to collect and organize the original data, and transmit the organized original data to the behavior prediction module and the feature recommendation module; The behavior prediction module is used to build prediction models based on artificial intelligence algorithms, analyze customer behavior and purchasing patterns, and predict customer needs and business intentions; The feature recommendation module is used to provide service recommendations based on customer characteristics according to collaborative filtering algorithms; The clustering segmentation module is used to segment customer groups based on clustering analysis algorithms; The decision support module is used to rate the service level according to the type of customer group.

2. The enterprise digital operation management system based on artificial intelligence according to claim 1 is characterized in that: The method of building a prediction model based on artificial intelligence algorithms is: Collect customer information, interaction data, sales data, and usage data, remove duplicate data, process missing values ​​and outliers, and normalize or standardize the collected data. Customer information includes the number of employees Nu and the tax amount Ta. Interaction data includes the number of email interactions En, the number of phone records Tn, and the number of meeting records Mn. Sales data includes the number of purchases Pn, the contract amount Ac, and the purchase cycle Bc. Usage data includes the number of product uses Uc and the number of fault feedback Fn. Extract features from the collected data, divide the data into training set and test set, use the random forest algorithm to set initial hyperparameters, and use the training set to train the random forest model; Use the test set to evaluate model performance, calculate the F1 score of the random forest model, and test the stability of the random forest model through cross-validation; Perform hyperparameter tuning to adjust the number of trees and maximum depth parameters; The optimized random forest model is deployed to predict customers’ purchase intentions.

3. The enterprise digital operation management system based on artificial intelligence according to claim 2 is characterized in that: In extracting features based on collected data, the feature selection method is: The characteristics include scale weight coefficient, interaction coefficient, consumption tendency coefficient, and usage feedback index. The scale weight coefficient is Sc, the interaction coefficient is In, the consumption tendency coefficient is Ct, and the usage feedback coefficient is Tf. The calculation expression of the scale weight coefficient is Sc=Nu*Ta, the calculation expression of the interaction coefficient is In=Mn*(En+Tn), and the calculation expression of the consumption tendency coefficient is The calculation expression using the feedback coefficient is 4. The enterprise digital operation management system based on artificial intelligence according to claim 1 is characterized in that: The method of using collaborative filtering algorithm for service recommendation is: Collect user behavior data on service ratings, browsing times, and purchase times; Evaluate service characteristics based on user behavior data on the service; Based on the service feature evaluation, similarity algorithms are used to calculate the similarity between users; For each user, find several similar users with the highest similarity to the user; Recommend services selected by similar users to target users based on their behaviors; The definition logic of service feature evaluation is: Service feature evaluation = score × (number of purchases / number of views).

5. The enterprise digital operation management system based on artificial intelligence according to claim 4 is characterized in that: Based on the service feature evaluation, the logic of using the similarity algorithm to calculate the similarity between users is: Let s be the service number, and s = {1, 2, 3…n}, where n is a positive integer. Calculate the Pearson correlation coefficient between users. The calculation expression is: In the formula, A n , B n is the evaluation of service characteristics of service n by user A and user B, is the average service feature evaluation of all purchased services by user A, is the mean service feature evaluation of all purchased services by user B.

6. The enterprise digital operation management system based on artificial intelligence according to claim 5 is characterized in that: The clustering segmentation module is used to segment customer groups according to the clustering analysis algorithm; Collect the customer's number of employees, annual purchase volume, years of consumption, income, and historical consumption amount. According to the K-means algorithm, determine the K value through the elbow method or silhouette coefficient method. Based on the results of cluster analysis, analyze the characteristics of each cluster and classify the customer type.

7. The enterprise digital operation management system based on artificial intelligence according to claim 6 is characterized in that: The logic for analyzing the cluster analysis results is: When the K value of K-means clustering is 2, for B2B business customers, if the number of employees and annual purchasing volume of customers in cluster 1 are greater than those in cluster 2, then the customers in cluster 1 are marked as large customers, and the customers in cluster 2 are marked as small and medium-sized customers; for B2C business customers, if the historical consumption amount and income of customers in cluster 1 are greater than those in cluster 2, and the consumption years of customers in cluster 1 are less than those of customers in cluster 2, then the customers in cluster 1 are marked as high-income customers, and the customers in cluster 2 are marked as medium-income customers.

8. The enterprise digital operation management system based on artificial intelligence according to claim 7 is characterized in that: The logic used by the decision support module to rate the service level according to the type of customer group is; When a B2B business customer is marked as a large customer, the service level is set to V1, providing customized solutions and bulk purchase discounts; When B2B business customers are marked as small and medium-sized customers, the service level is set to V2, providing flexible procurement plans and support services; When B2C business customers are marked as high-income customers, the service level is set to S1, providing recommendations and discounts on high-end fashion and technology products; When a B2C business customer is marked as a middle-income customer, the service level is set to S2, providing discount promotions on daily necessities and home services.

Images