CN118485444A - Intelligent customer relationship management system based on big data - Google Patents
Intelligent customer relationship management system based on big data Download PDFInfo
- Publication number
- CN118485444A CN118485444A CN202410476085.4A CN202410476085A CN118485444A CN 118485444 A CN118485444 A CN 118485444A CN 202410476085 A CN202410476085 A CN 202410476085A CN 118485444 A CN118485444 A CN 118485444A
- Authority
- CN
- China
- Prior art keywords
- client
- customer
- module
- information
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Game Theory and Decision Science (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Computational Linguistics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an intelligent customer relation management system based on big data, which relates to the technical field of customer relation management and comprises a customer information collection unit, wherein the customer information collection unit is used for acquiring customer related information of an e-commerce company, processing the acquired customer related information and storing the processed customer related information for subsequent use. The invention provides an intelligent customer relation management system based on big data, which takes e-commerce company customer relation management as an example, obtains and stores relevant information of e-commerce company customers through a set customer information collection unit, constructs an evaluation system for evaluating the value of the e-commerce company customers through a set customer comprehensive evaluation unit, determines the value of each customer based on the evaluation system, divides each customer into different echelons, and re-divides the customer echelon conditions according to unit time.
Description
Technical Field
The invention relates to the technical field of customer relationship management, in particular to an intelligent customer relationship management system based on big data.
Background
The invention relates to a customer relationship management, which is characterized in that an enterprise coordinates the interaction between the enterprise and customers in sales, marketing and service by utilizing corresponding information technology and Internet technology so as to promote the management mode, provides innovative personalized customer interaction and service for the customers, and discloses a big data-based intelligent customer relationship management system and method in the Chinese patent with the application number of 202211010969.8, belonging to the technical field of customer relationship management, the virtual clients are eliminated through the registration information and the login information of the clients, the effective clients are further finely divided, different client roles and client use log categories are set in the dividing process, the clients are classified in an all-around mode, deep iteration optimization is conducted through different attributes or characteristics in the client relationship, invalid clients are eliminated, accurate service is conducted on the relationship clients after each iteration, and high-quality and efficient experience can be provided for the clients. ";
The prior art only solves the problems that the existing management mode focuses on single clients, the service mode aiming at the single clients is emphasized and analyzed, service correlation among the clients is often ignored, and quantitative identification methods and means for large client groups are lacked, the requirement of carrying out echelon division based on the value of the clients is not considered, the client echelon situation is required to be divided again according to the change situation of the client value within a certain time, whether the clients have consumption behaviors or not is predicted, and the timely adjustment of an e-commerce marketing strategy and the hierarchical management of different echelon clients are realized by combining the prediction results, early warning is carried out on the client loss situation, and timely saving of the clients is realized.
Disclosure of Invention
The invention aims to provide an intelligent customer relationship management system based on big data so as to solve the problems in the background technology.
In order to achieve the above purpose, the present invention provides the following technical solutions: the intelligent client relation management system based on big data comprises a client information collection unit, wherein the client information collection unit is used for acquiring client related information of an e-commerce company, processing the acquired client related information and storing the processed client related information for subsequent use;
The comprehensive evaluation unit is used for constructing an evaluation system for evaluating the value of the clients of the electronic commerce company, determining the weight of each evaluation index in the constructed evaluation system, calculating the value of each client of the electronic commerce company based on the weight, and classifying the clients of the electronic commerce company based on the calculation result;
The client relation management unit is used for constructing a client behavior prediction model for predicting client consumption behaviors, carrying out model training on the client behavior prediction model through the constructed client consumption behavior data set, carrying out precision verification on the trained client behavior prediction model, and realizing adjustment of marketing strategies of different types of clients based on the output result of the client behavior prediction model so as to realize hierarchical management of the different types of clients;
The client loss management unit is used for constructing a client loss early-warning model for predicting the client loss condition, carrying out model training on the client loss early-warning model through the constructed client loss data set, carrying out precision verification on the trained client loss early-warning model, and saving corresponding clients based on the output result of the client loss early-warning model.
Preferably, the client information collecting unit includes a client information obtaining module, where the client information obtaining module is configured to obtain client related information of an e-commerce company, and specifically includes basic information, transaction information and browsing information of a client, where the basic information of the client is information generated when the client registers a corresponding platform of the e-commerce company, including name, gender, contact information, age, height weight and living area of the client, and the transaction information of the client is information generated when the client performs transaction activities on the e-commerce platform, including transaction amount, transaction times, transaction time, transaction product types and transaction product quantity, and the browsing information of the client is information generated when the client performs browsing operations on the e-commerce platform, including product retrieval information, store information of browsing products and browsing product information.
Preferably, the client information collecting unit further comprises a client information processing module and a client data storage module, wherein the client information processing module is used for performing data cleaning processing on the acquired client related information, specifically, performing data cleaning processing by filling in missing values, deleting abnormal values and smoothing noise data so as to remove invalid and repeated data and ensure the accuracy and the integrity of the acquired client related data, and the client data storage module is used for uploading the client related data processed by the client information processing module into a database built in a system for storage and classifying the data by taking a city as a unit according to the living area of a client.
Preferably, the customer comprehensive evaluation unit includes an evaluation system construction module and an evaluation index weight determination module, wherein the evaluation system construction module is used for constructing an evaluation system for evaluating the customer value of the e-commerce company, specifically selecting the latest consumption interval, consumption period and average single consumption amount of the customer as the evaluation indexes, and the calculation formula of each evaluation index is as follows:
R＝tn-tl
Wherein R represents the last consumption interval of the client, S represents the consumption period, A represents the average single consumption amount, t n represents the current time, t l represents the last consumption time of the client, t f represents the first consumption time of the client, N represents the consumption times of the client in the appointed time, P represents the total consumption amount of the client in the appointed time, and the calculation results of all evaluation indexes of the client are collected, the evaluation index weight determining module is used for determining the weight of all evaluation indexes, specifically, the calculation results of all evaluation indexes of the collected client are subjected to data standardization processing, the standard deviation of all indexes is calculated, and the conflict of all indexes is further calculated through a data correlation algorithm, wherein the specific calculation formula is as follows:
wherein M j represents the conflict value of the jth index, i represents the ith index, n represents the total number of evaluation indexes, r ij represents the correlation coefficient between the ith index and the jth index, the calculation result of each index is multiplied by the standard deviation of the corresponding index to obtain the data quantity of the corresponding index, and the weight of the index is calculated by a weight algorithm based on the data quantity, wherein the specific calculation formula is as follows:
where ω j denotes the weight of the jth index, and C j denotes the data amount of the jth index.
Preferably, the comprehensive evaluation unit of the client further comprises a client value determining module and a client value classifying module, wherein the client value determining module is used for obtaining the value of the client by multiplying each index by the corresponding weight and summing the product result based on the numerical value of each index calculated in the evaluation system constructing module and the weight value of each index calculated in the evaluation index weight determining module, and the client value classifying module is used for classifying the client into four teams according to the size of the client value from high to low based on the quartile of the client value calculated by the client value determining module, sequentially comprising a high-value client, a key client, a general value client and a low-value client, and re-classifying the client of the current month based on the calculation result of the client value of the previous month in unit time.
Preferably, the client relationship management unit includes a client behavior prediction model construction module and a prediction model verification module, where the client behavior prediction model construction module is used to construct a client behavior prediction model for predicting whether a client consumes, specifically based on a BP neural network, and each index in a client value evaluation system is selected as input data of the client behavior prediction model, and the prediction model verification module is used to construct a client consumption behavior data set for model training, and sample data types in the client consumption behavior data set are the same as input data types of the client behavior prediction model, and after normalization processing is performed on the sample data, the client behavior prediction model is trained through the processed client consumption behavior data set, and accuracy verification is performed on the trained client behavior prediction model through a model accuracy algorithm, and a specific calculation formula is as follows:
In the formula, PR represents the precision value of the client behavior prediction model, TP represents the number of sample data with correct prediction, FP represents the number of sample data with incorrect prediction, when the calculation result is larger than 0.95, the precision of the client behavior prediction model meets the requirement, when the calculation result is not larger than 0.95, the precision of the client behavior prediction model is not in accordance with the requirement, the client behavior prediction model needs to be optimized and adjusted, and a cost loss function is specifically selected as a loss function optimized by the client behavior prediction model, and the specific calculation formula is as follows:
Where K represents the total number of sample data in the customer consumption behavior data set, s represents the s-th sample data, y g represents the label of the g-th sample data, the prediction success is 1, the prediction failure is 0, Indicating the probability of success of the prediction of the g-th sample data.
Preferably, the client relationship management unit further includes a marketing strategy adjustment module and a client grading management module, where the marketing strategy adjustment module is configured to obtain, based on an output result of the client behavior prediction model to each client, a client proportion of the predicted result as having a consumption behavior, increase a quantity of issued coupons when the proportion is not greater than eighty percent, increase a discount degree of issued coupons and reduce a use threshold when the proportion is not greater than seventy percent, increase an applicable merchant range of issued coupons when the proportion is not greater than sixty percent, and the client grading management module is configured to obtain, based on an output result of the client behavior prediction model to each client, a client proportion of the predicted result as having a consumption behavior in each echelon client, increase a service budget and a discount degree to high-value clients and key clients when the proportion is not greater than eighty percent in the high-value clients and key clients, and increase a service budget and a discount degree to general-value clients and low-value clients when the proportion is not greater than sixty percent in the general-value clients and low-value clients.
Preferably, the client loss management unit includes a client loss early-warning model construction module, an early-warning model training module and a client maintenance module, wherein the client loss early-warning model construction module is used for constructing a client loss early-warning model for predicting client loss conditions, a XGBoost model and a Logistic model are specifically selected as a first-layer base model, a Logistic regression is a meta model, a client loss early-warning model is constructed in a Stacking fusion mode, various indexes in a client value evaluation system are selected as input data of the client loss early-warning model, the early-warning model training module is used for constructing a client loss data set for model training, the sample data type in the client loss data set is the same as the input data type of the client loss early-warning model, after normalization processing is performed on the sample data, the client loss early-warning model is trained through the processed client loss data set, and the client loss early-warning model is subjected to accuracy verification after completion of the client loss training, and when the accuracy of the client loss model is not in accordance with requirements, the client maintenance module is used for optimizing the client loss model, and the client maintenance module is used for carrying out the corresponding client on output results based on the client loss early-warning models, and the client loss early-warning model, and the priority of the client is the client loss early-warning model is the same as the client queue.
Compared with the prior art, the invention has the beneficial effects that at least: the invention provides an intelligent customer relation management system based on big data, which takes e-commerce company customer relation management as an example, obtains and stores relevant information of e-commerce company customers through a set customer information collection unit, builds an evaluation system for evaluating the value of the e-commerce company customers through a set customer comprehensive evaluation unit, determines the value of each customer based on the evaluation system, divides each customer into different echelons, reclassifies the customer echelon according to unit time, predicts whether the customers have consumption behaviors in the current unit time through the set customer relation management unit, adjusts marketing strategies of the e-commerce company and hierarchical management of different echelon customers based on prediction results, pre-warns customer losses through the set customer loss management unit, and saves the corresponding customers, wherein the priority of the saved customers is the same as the order of the customer echelons.
Drawings
FIG. 1 is a schematic overall flow chart of an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a client information collection unit according to an embodiment of the present invention;
FIG. 3 is a schematic flow chart of a customer comprehensive evaluation unit according to an embodiment of the present invention;
FIG. 4 is a schematic flow chart of a customer relationship management unit according to an embodiment of the present invention;
fig. 5 is a schematic flow chart of a customer churn management unit according to an embodiment of the present invention.
In the figure: 100. a customer information collection unit; 101. a customer information acquisition module; 102. a client information processing module; 103. a customer data storage module; 200. a customer comprehensive evaluation unit; 201. an evaluation system construction module; 202. an evaluation index weight determining module; 203. a customer value determination module; 204. a customer value classification module; 300. a customer relationship management unit; 301. a client behavior prediction model building module; 302. a prediction model verification module; 303. marketing strategy adjustment module; 304. a client grading management module; 400. a customer churn management unit; 401. a customer loss early warning model building module; 402. the early warning model training module; 403. and a client maintenance module.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1-5, the present invention provides a technical solution: the intelligent client relation management system based on big data comprises a client information collection unit 100, wherein the client information collection unit 100 is used for acquiring client related information of an e-commerce company, processing the acquired client related information and storing the processed client related information for subsequent use, and taking e-commerce company client relation management as an example;
The client comprehensive evaluation unit 200 is used for constructing an evaluation system for evaluating the client value of the e-commerce company, determining the weight of each evaluation index in the constructed evaluation system, calculating the value of each client of the e-commerce company based on the weight, and classifying the clients of the e-commerce company based on the calculation result;
The client relationship management unit 300, the client relationship management unit 300 is used for constructing a client behavior prediction model for predicting client consumption behaviors, performing model training on the client behavior prediction model through the constructed client consumption behavior data set, performing accuracy verification on the trained client behavior prediction model, and based on the output result of the client behavior prediction model, realizing adjustment of marketing strategies of different types of clients, and further realizing hierarchical management of different types of clients;
The client loss management unit 400 is used for constructing a client loss early-warning model for predicting the client loss condition, carrying out model training on the client loss early-warning model through the constructed client loss data set, carrying out accuracy verification on the trained client loss early-warning model, and saving corresponding clients based on the output result of the client loss early-warning model.
The client information collection unit 100 includes a client information acquisition module 101, where the client information acquisition module 101 is configured to acquire client related information of an e-commerce company, and specifically includes basic information, transaction information and browsing information of a client, where the basic information of the client is information generated when the client registers a corresponding platform of the e-commerce company, including name, gender, contact manner, age, height weight and living area of the client, and the transaction information of the client is information generated when the client performs transaction activities on the e-commerce platform, including transaction amount, transaction times, transaction time, transaction product types and transaction product numbers, and the browsing information of the client is information generated when the client performs browsing operations on the e-commerce platform, including product retrieval information, store information of browsing products and browsing product information;
The client information collecting unit 100 further includes a client information processing module 102 and a client data storage module 103, where the client information processing module 102 is configured to perform data cleaning processing on the acquired client related information, specifically, implement data cleaning processing by filling in missing values, deleting abnormal values and smoothing noise data, so as to remove invalid and repeated data, ensure accuracy and integrity of the acquired client related data, and the client data storage module 103 is configured to upload the client related data processed by the client information processing module 102 into a database built in the system for storage, and perform data classification according to a living area of the client in units of market;
the customer comprehensive evaluation unit 200 includes an evaluation system construction module 201 and an evaluation index weight determination module 202, wherein the evaluation system construction module 201 is used for constructing an evaluation system for evaluating the customer value of the e-commerce company, specifically selecting the latest consumption interval, the consumption period and the average single consumption amount of the customer as evaluation indexes, and the calculation formula of each evaluation index is as follows:
R＝tn-tl
Wherein R represents the last consumption interval of the client, S represents the consumption period, a represents the average single consumption amount, t n represents the current time, t l represents the last consumption time of the client, t f represents the first consumption time of the client, N represents the number of times of consumption of the client in a specified time, P represents the total consumption amount of the client in the specified time, and the calculation results of all evaluation indexes of the client are collected, the evaluation index weight determining module 202 is used for determining the weight of each evaluation index, specifically operates to perform data standardization processing on the calculation results of all evaluation indexes of the collected client, calculates the standard deviation of each index, and further calculates the conflict of each index through a data correlation algorithm, and the specific calculation formula is as follows:
wherein M j represents the conflict value of the jth index, i represents the ith index, n represents the total number of evaluation indexes, r ij represents the correlation coefficient between the ith index and the jth index, the calculation result of each index is multiplied by the standard deviation of the corresponding index to obtain the data quantity of the corresponding index, and the weight of the index is calculated by a weight algorithm based on the data quantity, wherein the specific calculation formula is as follows:
Wherein ω j represents the weight of the jth index, and C j represents the data amount of the jth index;
The client comprehensive evaluation unit 200 further comprises a client value determining module 203 and a client value classifying module 204, wherein the client value determining module 203 is used for obtaining the value of the client by multiplying each index by the corresponding weight and summing the product result based on the numerical value of each index calculated in the evaluation system constructing module 201 and the weight value of each index calculated in the evaluation index weight determining module 202, the client value classifying module 204 is used for classifying the client into four teams according to the size of the client value from high to low based on the quartile of the client value calculated by the client value determining module 203, sequentially comprising a high-value client, a key client, a general-value client and a low-value client, and re-classifying the client of the month based on the calculation result of the client value of the month in unit time;
The client relationship management unit 300 includes a client behavior prediction model construction module 301 and a prediction model verification module 302, where the client behavior prediction model construction module 301 is configured to construct a client behavior prediction model for predicting whether a client consumes, specifically based on a BP neural network, and select each index in a client value evaluation system as input data of the client behavior prediction model, and the prediction model verification module 302 is configured to construct a client consumption behavior data set for model training, where sample data types in the client consumption behavior data set are the same as input data types of the client behavior prediction model, and after normalization processing is performed on the sample data, the client behavior prediction model is trained through the processed client consumption behavior data set, and the trained client behavior prediction model is accurately verified through a model accuracy algorithm, and a specific calculation formula is as follows:
In the formula, PR represents the precision value of the client behavior prediction model, TP represents the number of sample data with correct prediction, FP represents the number of sample data with incorrect prediction, when the calculation result is larger than 0.95, the precision of the client behavior prediction model meets the requirement, when the calculation result is not larger than 0.95, the precision of the client behavior prediction model is not in accordance with the requirement, the client behavior prediction model needs to be optimized and adjusted, and a cost loss function is specifically selected as a loss function optimized by the client behavior prediction model, and the specific calculation formula is as follows:
Where K represents the total number of sample data in the customer consumption behavior data set, s represents the s-th sample data, y g represents the label of the g-th sample data, the prediction success is 1, the prediction failure is 0, Representing the probability of success of the prediction of the g-th sample data;
the client relationship management unit 300 further includes a marketing strategy adjustment module 303 and a client grading management module 304, the marketing strategy adjustment module 303 is configured to obtain, based on the output result of the client behavior prediction model, a client with consumption behavior as a prediction result, increase the number of issued coupons when the client is not more than eighty percent, increase the discount degree of issued coupons and reduce the use threshold when the client is not more than seventy percent, increase the applicable merchant range of issued coupons when the client is not more than sixty percent, and the client grading management module 304 is configured to obtain, based on the output result of the client behavior prediction model, a client with consumption behavior as a prediction result in each echelon client, increase the service budget and the discount degree for the high-value client and the high-value client when the client is not more than eighty percent, and increase the service budget and the discount degree for the general-value client and the low-value client when the client is not more than sixty percent;
The client loss management unit 400 includes a client loss early-warning model construction module 401, an early-warning model training module 402 and a client maintenance module 403, where the client loss early-warning model construction module 401 is used for constructing a client loss early-warning model for predicting a client loss situation, specifically selecting XGBoost models and Logistic models as a first layer base model, logically regressing the first layer base model as meta models, constructing the client loss early-warning model in a Stacking fusion manner, selecting various indexes in a client value evaluation system as input data of the client loss early-warning model, the early-warning model training module 402 is used for constructing a client loss data set for model training, sample data types in the client loss data set are the same as input data types of the client loss early-warning model, after normalization processing is performed on the sample data, the client loss early-warning model is trained through the processed client loss data set, and when accuracy of the trained client loss early-warning model is not in accordance with requirements, the client loss early-warning model is optimized, and the client maintenance module 403 is used for saving output results of each client based on the client, and priority of the corresponding clients is the client loss early-warning model is the same as that of the client loss early-warning model.
Working principle: acquiring the client related information of the e-commerce company through a client information acquisition module 101, performing data cleaning processing on the acquired client related information through a client information processing module 102, uploading the client related data processed through the client information processing module 102 into a database built in a system through a client data storage module 103 for storage, constructing an evaluation system for evaluating the client value of the e-commerce company through an evaluation system construction module 201, determining the weight of each evaluation index in the evaluation system through an evaluation index weight determination module 202, obtaining the value of each client through a client value determination module 203, classifying the clients into four teams according to the size of the client value based on the quartiles of the client value calculated by the client value determination module 203 through a client value classification module 204, constructing a customer behavior prediction model for predicting whether customers consume by a customer behavior prediction model construction module 301, training the customer behavior prediction model by a prediction model verification module 302, verifying the accuracy of the trained customer behavior prediction model, adjusting marketing strategies by a marketing strategy adjustment module 303 based on the output results of the customers by the customer behavior prediction model, performing hierarchical management on the customers in the echelon by a customer hierarchical management module 304, constructing a customer loss early-warning model for predicting the customer loss situation by a customer loss early-warning model construction module 401, training the customer loss early-warning model by an early-warning model training module 402, verifying the accuracy of the trained customer loss early-warning model, performing accuracy verification on the output results of the customers by a customer maintenance module 403 based on the customer loss early-warning model, the corresponding clients are saved.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (8)
1. An intelligent customer relationship management system based on big data is characterized in that: the system comprises a client information collection unit (100), wherein the client information collection unit (100) is used for acquiring client related information of an e-commerce company, processing the acquired client related information and storing the processed client related information for subsequent use, and is used for managing the client relationship of the e-commerce company;
The comprehensive evaluation unit (200) is used for constructing an evaluation system for evaluating the value of the clients of the electronic commerce company, determining the weight of each evaluation index in the constructed evaluation system, calculating the value of each client of the electronic commerce company based on the weight, and classifying the clients of the electronic commerce company based on the calculation result;
The client relation management unit (300), the said client relation management unit (300) is used for constructing the customer behavior prediction model of predicting customer's consumption behavior, and carry on model training to the customer behavior prediction model through the customer's consumption behavior dataset constructed, and carry on the precision verification to the customer behavior prediction model that is trained, and based on the output result of the customer behavior prediction model, realize the adjustment to marketing tactics of different types of customer, and then realize the hierarchical management to different types of customer;
the client loss management unit (400) is used for constructing a client loss early-warning model for predicting the client loss condition, carrying out model training on the client loss early-warning model through the constructed client loss data set, carrying out precision verification on the trained client loss early-warning model, and saving corresponding clients based on the output result of the client loss early-warning model.
2. The big data based intelligent customer relationship management system of claim 1, wherein: the client information collecting unit (100) comprises a client information obtaining module (101), wherein the client information obtaining module (101) is used for obtaining client related information of an electronic commerce company, and specifically comprises basic information, transaction information and browsing information of the client, wherein the basic information of the client is information generated when the client registers a corresponding platform of the electronic commerce company, the basic information comprises names, sexes, contact ways, ages, height weights and living areas of the client, the transaction information of the client is information generated when the client performs transaction activities on the electronic commerce platform, the transaction information comprises transaction amount, transaction times, transaction time, transaction product types and transaction product quantity, and the browsing information of the client is information generated when the client performs browsing operations on the electronic commerce platform, and the browsing information comprises product retrieval information, store information of browsing products and browsing product information.
3. The big data based intelligent customer relationship management system of claim 1, wherein: the client information collecting unit (100) further comprises a client information processing module (102) and a client data storage module (103), wherein the client information processing module (102) is used for carrying out data cleaning processing on the acquired client related information, specifically, the data cleaning processing is realized by filling in missing values, deleting abnormal values and smoothing noise data so as to remove invalid and repeated data and ensure the accuracy and the integrity of the acquired client related data, and the client data storage module (103) is used for uploading the client related data processed by the client information processing module (102) into a database in a system for storage and carrying out data classification according to the living area of a client by taking a city as a unit.
4. The big data based intelligent customer relationship management system of claim 1, wherein: the customer comprehensive evaluation unit (200) comprises an evaluation system construction module (201) and an evaluation index weight determination module (202), wherein the evaluation system construction module (201) is used for constructing an evaluation system for evaluating the customer value of an electronic commerce company, specifically selecting the latest consumption interval, the latest consumption period and the average single consumption amount of a customer as evaluation indexes, collecting the calculation results of all the evaluation indexes of the customer, and the evaluation index weight determination module (202) is used for determining the weights of all the evaluation indexes, specifically operating to perform data standardization processing on the calculation results of all the evaluation indexes of the collected customer, calculating the standard deviation of all the indexes, further calculating the conflict of all the indexes through a data correlation algorithm, multiplying the calculation results of all the indexes with the standard deviation of corresponding indexes to obtain the data quantity of the corresponding indexes, and calculating the weights of all the indexes through a weight algorithm based on the data quantity.
5. The big data based intelligent customer relationship management system according to claim 4, wherein: the comprehensive customer evaluation unit (200) further comprises a customer value determining module (203) and a customer value classifying module (204), wherein the customer value determining module (203) is used for dividing customers into four teams according to the size of the customer value from high to low, sequentially high-value customers, key customers, general-value customers and low-value customers on the basis of the numerical values of the various indexes calculated in the evaluation system constructing module (201) and the weight values of the various indexes calculated in the evaluation index weight determining module (202), the value of the customers is obtained by multiplying the various indexes by corresponding weights and summing the product results, and the customer value classifying module (204) is used for dividing the customers into teams according to the calculation results of the customer value of the last month in unit time.
6. The big data based intelligent customer relationship management system of claim 1, wherein: the client relationship management unit (300) comprises a client behavior prediction model construction module (301) and a prediction model verification module (302), wherein the client behavior prediction model construction module (301) is used for constructing a client behavior prediction model for predicting whether a client consumes, specifically, constructing the client behavior prediction model based on a BP neural network, selecting various indexes in a client value evaluation system as input data of the client behavior prediction model, the prediction model verification module (302) is used for constructing a client consumption behavior data set for model training, the sample data type in the client consumption behavior data set is the same as the input data type of the client behavior prediction model, after normalization processing is carried out on the sample data, the client behavior prediction model is trained through the processed client consumption behavior data set, and the trained client behavior prediction model is subjected to accuracy verification through a model accuracy algorithm, when a calculation result is larger than 0.95, the accuracy of the client behavior prediction model is indicated to reach the requirement, when the calculation result is not larger than 0.95, the accuracy of the client behavior prediction model is indicated to be not in line with the requirement, the client behavior prediction model is required to be optimized, and a cost function is specifically selected to be used as a loss function of client behavior prediction optimization.
7. The big data based intelligent customer relationship management system of claim 1, wherein: the client relationship management unit (300) further comprises a marketing strategy adjustment module (303) and a client grading management module (304), wherein the marketing strategy adjustment module (303) is used for obtaining the predicted result of each client based on the output result of the client behavior prediction model, obtaining the predicted result of the client with consumption behavior, increasing the quantity of issued coupons when the predicted result of each client is not more than eighty percent, increasing the discount degree of issued coupons and reducing the use threshold when the predicted result of each client is not more than seventy percent, increasing the applicable merchant range of issued coupons when the predicted result of each client is not more than sixty percent, and the client grading management module (304) is used for obtaining the predicted result of each client in a echelon to obtain the client proportion of the client with consumption behavior based on the output result of each client behavior prediction model, increasing the service budget and the discount degree of the client with high value and the key client when the predicted result of each client with high value and the key client is not more than eighty percent, and increasing the service budget and the discount degree of the client with the general value and the service budget and the discount degree of the client when the predicted result of each client with high value and the key client with the high value client is not more than sixty percent.
8. The big data based intelligent customer relationship management system of claim 1, wherein: the customer loss management unit (400) comprises a customer loss early warning model construction module (401), an early warning model training module (402) and a customer maintenance module (403), wherein the customer loss early warning model construction module (401) is used for constructing a customer loss early warning model for predicting customer loss conditions, a XGBoost model and a Logistic model are specifically selected as a first layer base model, logistic regression is carried out as meta models, a customer loss early warning model is constructed in a Stacking fusion mode, various indexes in a customer value evaluation system are selected as input data of the customer loss early warning model, the early warning model training module (402) is used for constructing a customer loss data set for model training, the sample data type in the customer loss data set is the same as the input data type of the customer loss early warning model, after the sample data is normalized, the customer loss early warning model is trained through the processed customer loss data set, the trained customer loss early warning model is subjected to accuracy verification, when the accuracy of the customer loss model is not in accordance with requirements, the customer loss early warning model is saved, and the customer maintenance module (403) is used for carrying out the output early warning on the customers in a cascade mode based on the customer loss early warning mode, and the customer loss early warning model is the same as the priority of the customer queue.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410476085.4A CN118485444A (en) | 2024-04-19 | 2024-04-19 | Intelligent customer relationship management system based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410476085.4A CN118485444A (en) | 2024-04-19 | 2024-04-19 | Intelligent customer relationship management system based on big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118485444A true CN118485444A (en) | 2024-08-13 |
Family
ID=92197763
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410476085.4A Pending CN118485444A (en) | 2024-04-19 | 2024-04-19 | Intelligent customer relationship management system based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118485444A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118674329A (en) * | 2024-08-22 | 2024-09-20 | 浙江迈新科技股份有限公司 | Low-code platform management method and system |
-
2024
- 2024-04-19 CN CN202410476085.4A patent/CN118485444A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118674329A (en) * | 2024-08-22 | 2024-09-20 | 浙江迈新科技股份有限公司 | Low-code platform management method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110704730B (en) | Product data pushing method and system based on big data and computer equipment | |
| US7676521B2 (en) | Keyword search volume seasonality forecasting engine | |
| CN118469405B (en) | Processing method, device, equipment and storage medium for household supply chain data | |
| CN109636482B (en) | Data processing method and system based on similarity model | |
| CN112562863A (en) | Epidemic disease monitoring and early warning method and device and electronic equipment | |
| CN119359357A (en) | A method and system for full-process marketing management of auto insurance channels | |
| CN113869768A (en) | Method, device, equipment and readable medium for evaluating strength of industrial chain | |
| CN114997916A (en) | Prediction method, system, electronic device and storage medium of potential user | |
| CN1430758A (en) | Revenue forecasting and managing sellers using statistical analysis | |
| CN118485444A (en) | Intelligent customer relationship management system based on big data | |
| CN115841345A (en) | Cross-border big data intelligent analysis method, system and storage medium | |
| CN117974321A (en) | Financial product risk management and control method based on rule engine | |
| KR102829094B1 (en) | Inventory management method and system through sales forecasting from launch of new products to the end of life cycle | |
| CN114707733B (en) | Risk indicator prediction method, device, electronic device and storage medium | |
| CN119991193A (en) | Big data driven market forecasting system and system | |
| KR102831812B1 (en) | Prediction system and its control method, and learning method of the prediction system | |
| CN118296389B (en) | Construction and evaluation method of data index model | |
| CN118626959B (en) | Regional alliance merchant data processing method and system | |
| KR102543211B1 (en) | Company's growth potential prediction system using unstructured data | |
| CN117557346B (en) | Full-link intelligent business decision analysis method based on dynamic consumption data | |
| Konda et al. | An In-Depth Evaluation of Machine Learning Techniques for Anticipating Effective Human Health Outcomes | |
| CN119722254B (en) | Recommendation method of financial products | |
| CN119295122A (en) | Method, device, equipment and storage medium for evaluating price elasticity of online car-hailing service | |
| CN119067451A (en) | An integrated business and financial service platform | |
| CN119338494A (en) | A method and system for intelligently collecting POS machine sales data based on big data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |