CN111160992A - Marketing system based on user portrait system - Google Patents
Marketing system based on user portrait system Download PDFInfo
- Publication number
- CN111160992A CN111160992A CN202010000618.3A CN202010000618A CN111160992A CN 111160992 A CN111160992 A CN 111160992A CN 202010000618 A CN202010000618 A CN 202010000618A CN 111160992 A CN111160992 A CN 111160992A
- Authority
- CN
- China
- Prior art keywords
- user
- data
- label
- labels
- marketing
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Abstract
A marketing method based on a user portrait system is a marketing system based on a data source module, a formulated user label system module, a label development module and an application module; the data source module is used for collecting multi-source data and storing the multi-source data in the most original data source layer in the data warehouse; the user label system making module is used for making a user label system suitable for the characteristics of the current core service; the label development module comprises three sub-modules of establishing a user ONEID, integrating user full data and performing statistics and data modeling; 1) collecting website customer data and storing the data in a data source layer of a data warehouse; 2) formulating a user label system suitable for the current service characteristics, and determining label levels and categories; 3) establishing ONEID for user identity identification; according to the logic rule of the label system formulated in the step 2), calculating various labels by adopting a model algorithm, and outputting the labels to a data warehouse market layer; and developing a label management page and a user grouping detail page, and directly outputting a target user to transmit to operation and maintenance personnel.
Description
Technical Field
The invention relates to the field of user portrait construction, in particular to a marketing system and a marketing method based on a user portrait system.
Background
In the era of mobile internet, as services and products of electronic commerce enterprises are enriched day by day, refined marketing and personalized services become important competitiveness of the enterprises, the concept of user portrayal is born at the same time, various labels are abstracted from mass data by means of data accumulation of the enterprises, and the user image is embodied to form the portrayal. Today's electronic commerce has already stepped into the big data era, can produce a large amount of customer access action data and consumption record every day, and long-term accumulation, the customer data of each source no longer is only island data, through user rule multisource combination, can describe the website user visualization of computer end, has formed the personage prototype, helps the enterprise fully to know the user, in time masters information such as its current state, action preference and potential demand, and is especially important to enterprise's whole business development.
Existing machine prediction algorithms and models are commonly used, such as gbdt (gradient Boosting decision tree) gradient Boosting iterative decision trees. GBDT is also one of the Boosting algorithms, but is different from the AdaBoost algorithm; the differences are as follows: the AdaBoost algorithm updates the sample weight value by using the error of the weak learner in the previous round, and then iterates one round; GBDT is also iterative, but GBDT requires that the weak learner be a CART model, and GBDT requires that the model predict as little sample loss as possible during model training. The GBDT mechanism is: and residual errors exist between each round of prediction and the actual value, the next round of prediction is carried out according to the residual errors, and finally all predictions are added to obtain a result. The support vector machine method SVM analyzes linear divisible conditions, and for linear inseparable conditions, linear divisible samples of a low-dimensional input space are converted into a high-dimensional feature space by using a nonlinear mapping algorithm so as to be linearly divisible, and therefore the linear analysis of the nonlinear features of the samples by the high-dimensional feature space through a linear algorithm becomes possible. The SVM constructs an optimal hyperplane in a feature space based on a structure risk minimization theory, so that a learner obtains global optimization, and the expectation of the whole sample space meets a certain upper bound with a certain probability.
A decision tree algorithm is a method of approximating discrete function values. It is a typical classification method that first processes the data, generates readable rules and decision trees using a generalisation algorithm, and then uses the decisions to analyze the new data. In essence, a decision tree is a process of classifying data through a series of rules. Typical algorithms for decision trees are ID3, C4.5, CART, etc.; how to construct a decision tree with high precision and small scale is the core content of the decision tree algorithm. The decision tree construction is performed in two steps. First, generation of a decision tree: a process of generating a decision tree from a training sample set. In general, a training sample data set is a data set which has a history according to actual needs and a certain degree of integration and is used for data analysis processing. Step two, pruning the decision tree: the pruning of the decision tree is the process of checking, correcting and repairing the decision tree generated at the previous stage, and is mainly to use the data in a new sample throttle set (called a test data set) to check the preliminary rules generated in the process of generating the decision tree and prune branches influencing the accuracy of pre-balance.
The construction process of the random forest comprises the following steps: randomly putting back and sampling m samples from an original training set by using a Bootstrap method, and performing n _ tree times of sampling to generate n _ tree training sets; for n _ tree training sets, respectively training n _ tree decision tree models; for a single decision tree model, assuming that the number of training sample features is n, selecting the best feature to split according to the information gain/information gain ratio/the kini index during each splitting; each tree is split until all training examples for that node belong to the same class. Pruning is not needed in the splitting process of the decision tree; and forming a random forest by the generated decision trees. For the classification problem, voting according to a plurality of tree classifiers to determine a final classification result; for the regression problem, the final prediction result is determined by the mean value of the predicted values of the multiple trees.
In enterprise customer management, knowing about customers, customers can be better served. The user is depicted through the attribute value of the tag, certain guiding significance is provided for marketing aspects such as scheme making, popularization channel, marketing content, personalized recommendation of website pages and the like, meanwhile, the information of the user is displayed in time in the process of communication between customer service and the user, the consultation service can be provided in a targeted mode, enterprise sales and service are enabled, the experience of the user on the website is improved, and continuous stable benefits are brought to the enterprise. Therefore, the marketing system based on the user portrait system is provided, a target user is sketched through massive client data, each piece of concrete information of the user is abstracted into labels, the labels are utilized to concretize the user, fine marketing is assisted, and a targeted service is provided for the user.
Disclosure of Invention
The invention aims to provide a marketing system based on a user portrait system under the background of the prior art, the user portrait system suitable for core business characteristics of an enterprise is built based on data such as user access behavior data, purchase records and offline records, meanwhile, the imperfect labels of the user are conjectured as comprehensively as possible by means of formulation of business logic strategies or data models and algorithms, the user portrait system is applied to a sales system and an operation system, valuable reference information is provided in time, the user requirements of websites are responded quickly and accurately, and the fine marketing effect of the enterprise is improved.
The invention adopts the technical scheme that a marketing method based on a user portrait system comprises the following steps: the system comprises a data source module, a user label system making module, a label development module and an application module.
The data source module collects multi-source data, stores the data source layer which is the most original in the data warehouse, and the main users can complete registered basic information, purchase records, access behaviors, offline consultation, third-party data and the like.
And the user label system setting module is used for setting a user label system suitable for the current core service characteristics by the big data product personnel and the service party together.
The label development module comprises three submodules of user ONEID establishment, user full data integration and statistics and data modeling.
The application module comprises three submodules of tag management, new user group creation and marketing target user group screening, and is mainly applied to the aspect of fine marketing.
The method comprises the following steps:
step 1: the website customer data collection, including basic information of the user in complete registration, purchase records, access behaviors, offline consultation, third-party data and the like, is stored in the most original data source layer in the data warehouse.
Step 2: and the big data product personnel and the business party jointly formulate a user label system suitable for the current business characteristics, and determine the label hierarchy and category.
And step 3: the only ONEID for user identification is established, only one identity can be identified in a portrait system by one user, and data generated by the same person at different positions can be identified as far as possible by means of business rules. Therefore, after ONEID, user data can be integrated, data of the same user is cleaned and processed in a centralized mode, and the user data are combined into a wide list according to subjects and stored in a detail layer of a data warehouse.
And 4, step 4: and (3) according to the label system logic rule formulated in the step (2), calculating various labels by means of a statistical method and a model, and outputting the labels to a market layer of the data warehouse.
And 5: developing a label management page, displaying a plurality of sub-labels below the current level label and user coverage corresponding to various sub-labels, and monitoring whether the label formulation is reasonable. Meanwhile, a new user group can be formed by linkage of a plurality of labels, and other characteristics of the part of users can be observed.
Step 6: developing a user grouping detail page, wherein operation can screen specified groups for fine marketing, directly output target users, transmit the target users to operation and maintenance personnel through an interface, and uniformly popularize the target users in a short message or mail mode and the like; or the system is connected to a customer service background system, supports customer service autonomous search or is directly related to a specified user portrait, displays user information and assists customers to take targeted service customers.
According to the logic rule of the label system formulated in the step 4, the calculation of various labels is realized by means of a statistical method and a model, and the calculation is output to a market layer in a data warehouse; the labels of the prediction classes are subjected to model training based on statistical data, a sampling algorithm is used for solving the problem of data imbalance, normalization processing is carried out on characteristics, then variable selection is carried out, independent variables with strong correlation with dependent variables can be selected through a lasso algorithm, multiple collinearity of the independent variables is eliminated, therefore, a training set only keeps important variables with high correlation, then different algorithms can be selected for carrying out prediction model training, decision trees, SVM, random forests and GDBT, and an optimal algorithm is selected according to model effects.
Has the advantages that: on the basis of multi-source data and target user drawing, each concrete information of a user is abstracted into labels, the labels are utilized to concretize the user, a user portrait system suitable for core business characteristics of an enterprise is built, meanwhile, the imperfect labels of the user are conjectured as comprehensively as possible by means of the formulation of business logic strategies or some data models and algorithms, the user portrait system is applied to a sales system and an operation system, valuable reference information is provided in time, the user requirements of a website are responded quickly and accurately, and the fine marketing effect of the enterprise is improved.
Drawings
FIG. 1 is a flow chart of a marketing system based on a user representation system according to an embodiment of the present invention.
FIG. 2 is a schematic diagram of a marketing system based on a user representation system according to an embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and specific examples.
Referring to fig. 1, an embodiment of the present invention is a method flow of a marketing system based on a user portrait system, and the method flow specifically includes the steps of:
step 11: the website customer data collection, including basic information of the user in complete registration, purchase records, access behaviors, offline consultation, third-party data and the like, is stored in the most original data source layer in the data warehouse.
Step 12: big data product personnel and business parties jointly make a user label system suitable for current business characteristics, personnel in a big data direction comb the user portrait system bias strategy and technical implementation, the business parties bias management and application directions, opinions of the two parties are finally fused, and labels are established from the level and category, so that the portrait system is comprehensive but not scattered, the dimensionality is not isolated, and the labels are related; the rule that the label attribute can cover all users but not intersect is established, for example, the user marriage and child bearing state, the attribute can be that there is no intersection between the three attribute values of married but not fertile, married but fertile, and if there are two states of married and fertile, there is intersection, and the label design is not reasonable. The label classifications corresponding to different services are different, and need to be formulated around core services, for example, the insurance service primary classification can be divided into basic attributes, social relations, insurance behaviors, risk information, guarantee preferences, guarantee conditions, user values, service information, and the like.
Step 13: the unique ONEID for user identity identification is established, some collected data mark the user identity, such as a website registration account number, a mobile phone number, a mailbox, a visitor cookie and the like, but one user can only have one identity identification in a portrait system, and the data generated by the same person in different positions can be identified as far as possible by means of business rules, so that the portrait accuracy can be ensured. Therefore, after ONEID, user data can be integrated, data of the same user are cleaned and processed in a centralized mode, and the data are combined into a wide table according to subjects and stored in a detail layer of a data warehouse.
Step 14: and (4) according to the label system logic rule formulated in the step (12), calculating various labels by means of a statistical method and a model, and outputting the labels to a market layer of the data warehouse. The basic information can be obtained through the registration information of the user, but if the user has a child or a car, the user consumption capacity level cannot be directly obtained from the data of the user, and the basic information can be calculated or inferred only by establishing a data model. For example, the location of the user, the area filled in when the user registers is preferentially taken, if the area is not filled in, the area matched with the mobile phone number is taken, if the mobile phone number is still matched, the area matched with the identity card is taken, and all users are covered as much as possible; for example, whether children exist or not is judged by screening out user behavior characteristics, the number of times of visiting related products of the children, the number of times of visiting in a fixed time period, the age, the region, the number of times of searching keywords and the like, model training is carried out, debugging is carried out continuously, and when the accuracy rate reaches more than 95%, whether children exist or not can be predicted. For a series of labels with liveness, interest scores of a certain product and the like, a coefficient of variation method is used for weighting and scoring, so that the labels can be simply output with good effect.
If the child exists, the car exists or not, and the like, the model with high accuracy is preferentially selected, and the selection is performed according to the recall rate under the condition that the accuracy is not high enough; the method comprises the following steps that (1) recall rate should be preferred for labels with dangerous buying intentions, accuracy is used as an evaluation basis of a model, when the recall rate reaches 80%, and the accuracy reaches more than 95%, the model effect reaches the optimum, through tests, the GDBT model effect is superior to that of other models, specific application is selected according to the model effect, and trained models are reused for predicting output labels; for a series of labels with liveness, a certain product interest score and the like, a coefficient of variation method is used for weighting and scoring, the liveness selects the latest login time, login period and login days of nearly 13 months as indexes according to the characteristics of service data, a specific implementation method is adopted, the calculation of three indexes is carried out in the first step, and the calculation of three indexes is carried out in the second step according to a formula
Vi＝σi/μi
Wherein sigmaiIs standard deviation, μiIs the mean value, the coefficient of variation of each index is calculated, and the third step is to calculate the weight of each index
The fourth step calculates an index score for each user based on Y ═ w1X1+ w2X2+ w3X3, and uses the ralida criterion method, i.e., for experimental data values where Y is greater than μ +3 σ or less than μ -3 σ as abnormal values, where the highest value is greater than μ +3 σ and the lowest value is less than μ -3 σ, and the fifth step normalizes the index scores and outputs a score value within 0-100 for each user. X1, X2, X3 are three indexes of the first step (login time, login period, and login days in the last 13 months as indexes);
step 15: compiling a label management page, displaying a plurality of sub-labels below the current primary label and user coverage corresponding to various sub-labels, and monitoring whether the label formulation is reasonable; meanwhile, a new user group is formed through linkage of a plurality of labels, the labels are selected according to the user labels generated at present, the group validity period set during new grouping is defined, the user group names are customized according to the screened labels, for example, the Nanjing area and male with the platform preferring to use APP, the total number of covered users is displayed, other label characteristics of the users are observed, and people in auxiliary operation, customer service and the like keep the user group which needs to be continuously observed. The authorized person can modify the user group, and when the effective period is over, the user group is automatically deleted, and the operator needs to redefine the user group. When the promotion activity is carried out, the related user groups are directly clicked, the user details are extracted, and then the marketing short messages or mails can be pushed to the users.
For example, if the user activity label is active for more than 80% of users, it is obvious that the label is not logical reasonable and needs to be adjusted. Meanwhile, a new user group can be formed through linkage of a plurality of labels, other characteristics of the user group are observed, and people such as operation and customer service are assisted to reserve the user group which needs to be continuously observed.
Step 16: developing a user grouping detail page, wherein operation can screen designated groups for fine marketing, directly output target users, transmit the target users to operation and maintenance personnel through an interface, and uniformly popularize the target users in a short message or mail mode, for example, three days before holidays, and screen out customers with preference to travel to push travel insurance; or the system is connected to a customer service background system, the customer service is supported to independently search or directly associate to a designated user portrait, user information is displayed, the customer service is assisted to take a targeted service client, for example, when the online customer service answers a client consultation question, a user grouping detail page is quickly matched through a client ID, a series of tags of the user are displayed, the customer service can be better communicated with the client, the current appeal of the client is accurately known, a proper product or service is recommended for the client, and the user experience is improved.
Referring to fig. 2, a system structure according to an embodiment of the present invention includes:
the system comprises a data source module 21, a user label system establishing module 22, a label development module 23 and an application module 24.
The data source module 21 obtains basic data left by the user for accessing a website or for offline consultation, and basic information and purchasing behavior data of the user according to the user rule, and may be from a service system, a text file or other data sources.
And a user label system making module 22 is used for making a label system suitable for the service direction around the core service by the big data product personnel and the service party, so that the website user is comprehensively depicted.
The tag development module 23 includes three sub-modules: and establishing a user ONEID submodule 231, a user full data integration submodule 232 and a statistical and data modeling submodule 233, and realizing one-stop development of a user tag system by means of business logic and a statistical model.
And establishing a user ONEID submodule 231, and integrating the marked user identities of the collected data into a uniform identity identification ID.
The user full data integration submodule 232 integrates, cleans and stores user data for data generated by the same ID at different positions.
The statistics and data modeling submodule 233 is used for modeling by combining portrait label rules with algorithms, the two methods are not available, in practical application, the algorithm is difficult to solve, and a good effect can be achieved by using simple rules.
The application module 24 comprises a tag management submodule, a newly-built user group submodule and a marketing target user group screening submodule. The label management submodule is used for generating certain fluctuation of a user portrait in practical application, and in order to solve the problem, a corresponding management and monitoring system is built for monitoring the portrait quality and discovering abnormal or unreasonable labels in time; the new user group submodule is added, and personnel such as operation, customer service and the like can form a new user group through linkage of a plurality of labels in a personalized manner, observe other characteristics of the part of users, reserve the definition of the user group and can continuously observe; and the screening marketing target user group submodule realizes that a user portrait system is accessed to backstage of a sales system, a customer service system and an operation system, and is in butt joint with operation and maintenance through an interface, so that a sales promotion plan is made in an auxiliary manner, customer service targeted service is provided, and the marketing conversion rate is improved.
Claims (5)
1. A marketing method based on a user portrait system is characterized in that based on a marketing system comprising a data source module, a user label system making module, a label development module and an application module:
the data source module is used for collecting multi-source data and storing the multi-source data in the most original data source layer in the data warehouse, wherein the data of the data source layer comprises basic information, purchase records, access behaviors, offline consultation and third-party data of a user in complete registration;
the user label system making module is used for making a user label system suitable for the characteristics of the current core service;
the label development module comprises three sub-modules of establishing a user ONEID, integrating user full data and performing statistics and data modeling;
the application module comprises three sub-modules of label management, user group creation and marketing target user group screening and is applied to refined marketing;
the method comprises the following steps:
step 1: collecting website client data, including basic information, purchase records, access behaviors, offline consultation, third-party data and the like of a user in complete registration, and storing the data in the most original data source layer in a data warehouse;
step 2: formulating a user label system suitable for the current service characteristics, and determining label levels and categories;
and step 3: establishing an ONEID with the only user identity identification, wherein only one user identity identification can be found in an image system, integrating user data after the ONEID is established, intensively cleaning and processing the data of the same user, and combining the user data into a broad list according to a theme and storing the broad list in a detail layer of a data warehouse;
and 4, step 4: according to the label system logic rule formulated in the step 2, calculation of various labels is realized by means of statistics and models, and the calculation is output to a market layer of a data warehouse;
and 5: developing a label management page, displaying a plurality of sub-labels below the current primary label and user coverage corresponding to various sub-labels, and monitoring whether the label formulation is reasonable; meanwhile, a new user group is formed through linkage of a plurality of labels, and other characteristics of the part of users are observed;
step 6: and developing a user grouping detail page, operating, screening and designating grouping for refined marketing, directly outputting target users, and transmitting the target users to operation and maintenance personnel through an interface.
2. The user representation based marketing method of claim 1,
and (4) according to the logic rule of the label system formulated in the step (4), realizing the calculation of various labels by means of a statistical method and a model, and outputting to a market layer in a data warehouse: the labels of prediction classes are subjected to model training based on statistical data, a sampling algorithm is used for solving the problem of data imbalance, normalization processing is carried out on characteristics, then variable selection is carried out, independent variables with strong correlation with dependent variables are selected through a lasso algorithm, multiple collinearity of the independent variables is eliminated, therefore, only important variables with high correlation are reserved in a training set, then different algorithms are selected for training of prediction models, the prediction models are trained through decision trees, SVM, random forests and GDBT, and the algorithms are selected according to model effects.
Whether children exist or not, whether cars exist or not and the like, selecting a model with high accuracy, selecting the model with low accuracy and selecting the model according to the recall rate.
3. The user representation based marketing method of claim 1,
the label of the purchase intention danger should preferably have recall rate, then the accuracy rate is used as the evaluation basis of the model, when the recall rate reaches 80 percent and the accuracy rate reaches more than 95 percent, the model effect reaches the best, and the GDBT model is adopted and the trained model is used for predicting the output label.
4. The marketing method based on user portrait system as defined in claim 1, wherein the tags with liveness and interest scores of certain products are weighted to obtain scores by using a coefficient of variation method, the liveness is characterized by using the latest login time, login period and login days of nearly 13 months as indexes according to the service data, the method is implemented by calculating three indexes in the first step, and calculating three indexes in the second step according to a formula
Vi＝σi/μi
Wherein sigmaiIs standard deviation, μiIs the mean value, the coefficient of variation of each index is calculated, and the third step is to calculate the weight of each index
The fourth step is that the index score of each user is calculated according to the fact that Y is w1X1+ w2X2+ w3X3, the Lauda criterion method is used, namely the experimental data value of Y which is larger than mu +3 sigma or smaller than mu-3 sigma is used as an abnormal value, the highest value of the value which is larger than mu +3 sigma is taken, and the lowest value of the value which is smaller than mu-3 sigma is taken, the index score is normalized, and each user outputs a score within 0-100; x1, X2, X3 are three indexes of the first step.
5. The user representation system based marketing method of claim 1, wherein in step 5: compiling a label management page, displaying a plurality of sub-labels below the current primary label and user coverage corresponding to various sub-labels, and monitoring whether the label formulation is reasonable; meanwhile, a new user group is formed through linkage of a plurality of labels, the labels are selected according to the user labels generated at present, the group validity period set during new grouping is defined, the user group names are customized according to the screened labels, for example, the Nanjing area and male with the platform preferring to use APP, the total number of covered users is displayed, other label characteristics of the users are observed, and people in auxiliary operation, customer service and the like keep the user group which needs to be continuously observed. The authorized person can modify the user group, and when the effective period is over, the user group is automatically deleted, and the operator needs to redefine the user group. When the promotion activity is carried out, the related user groups are directly clicked, the user details are extracted, and then the marketing short messages or mails can be pushed to the users.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010000618.3A CN111160992A (en) | 2020-01-02 | 2020-01-02 | Marketing system based on user portrait system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010000618.3A CN111160992A (en) | 2020-01-02 | 2020-01-02 | Marketing system based on user portrait system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111160992A true CN111160992A (en) | 2020-05-15 |
Family
ID=70560961
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010000618.3A Pending CN111160992A (en) | 2020-01-02 | 2020-01-02 | Marketing system based on user portrait system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111160992A (en) |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111738767A (en) * | 2020-06-23 | 2020-10-02 | 江苏云柜网络技术有限公司 | Express delivery operation service providing method and device, electronic equipment and storage medium |
| CN111861697A (en) * | 2020-07-02 | 2020-10-30 | 北京睿知图远科技有限公司 | Loan multi-head data-based user portrait generation method and system |
| CN111897963A (en) * | 2020-08-06 | 2020-11-06 | 沈鑫 | Commodity classification method based on text information and machine learning |
| CN111931107A (en) * | 2020-07-31 | 2020-11-13 | 上海博泰悦臻电子设备制造有限公司 | Digital citizen system construction method, system and storage medium |
| CN112000849A (en) * | 2020-08-21 | 2020-11-27 | 河南中原消费金融股份有限公司 | Unified label library management method, device, equipment and storage medium |
| CN112001757A (en) * | 2020-08-26 | 2020-11-27 | 中山世达模型制造有限公司 | Sales order prediction method |
| CN112215655A (en) * | 2020-10-13 | 2021-01-12 | 广东电网有限责任公司 | Client portrait label management method and system |
| CN112269805A (en) * | 2020-11-18 | 2021-01-26 | 杭州米雅信息科技有限公司 | Data processing method, device, equipment and medium |
| CN112365073A (en) * | 2020-11-18 | 2021-02-12 | 贵州电网有限责任公司 | Regulation and control operation signal monitoring method based on big data |
| CN112418929A (en) * | 2020-11-23 | 2021-02-26 | 中国工商银行股份有限公司 | Data sharing method and device |
| CN112529624A (en) * | 2020-12-15 | 2021-03-19 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for generating business prediction model |
| CN112734485A (en) * | 2021-01-13 | 2021-04-30 | 上海群之脉信息科技有限公司 | User intelligent operation system |
| CN112765146A (en) * | 2021-01-26 | 2021-05-07 | 四川新网银行股份有限公司 | Method for monitoring data quality of user portrait label |
| CN112862542A (en) * | 2021-03-10 | 2021-05-28 | 郑州时空隧道信息技术有限公司 | Marketing method and system |
| CN112860851A (en) * | 2021-01-22 | 2021-05-28 | 平安科技（深圳）有限公司 | Course recommendation method, device, equipment and medium based on root cause analysis |
| CN113076382A (en) * | 2021-06-07 | 2021-07-06 | 北京明略软件系统有限公司 | User label generation method and device, electronic equipment and readable storage medium |
| CN113139750A (en) * | 2021-05-14 | 2021-07-20 | 中国平安人寿保险股份有限公司 | Course recommendation method, device, server and storage medium |
| CN113347213A (en) * | 2021-08-05 | 2021-09-03 | 环球数科集团有限公司 | Trusted channel authentication system based on protection of sensitive data of evanescent member |
| CN113421133A (en) * | 2021-07-21 | 2021-09-21 | 赛诺数据科技(南京)有限公司 | Network marketing system based on customer matching |
| CN113722305A (en) * | 2021-08-26 | 2021-11-30 | 广东电网有限责任公司 | Analysis application system and method |
| CN113965416A (en) * | 2021-12-21 | 2022-01-21 | 江苏移动信息系统集成有限公司 | Website security protection capability scheduling method and system based on workflow |
| CN114912946A (en) * | 2022-04-24 | 2022-08-16 | 零犀(北京)科技有限公司 | Method, device, storage medium and electronic equipment for determining user hierarchy |
| CN115357802A (en) * | 2022-10-19 | 2022-11-18 | 山东省商业集团有限公司 | Multi-state enterprise potential customer identification method |
| CN115757396A (en) * | 2022-10-25 | 2023-03-07 | 杭州比智科技有限公司 | Oneid project implementation method and oneid project implementation system |
| CN115829615A (en) * | 2023-01-05 | 2023-03-21 | 瓴创(北京)科技有限公司 | User grouping method, system and storage medium based on multiple databases |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109583651A (en) * | 2018-12-03 | 2019-04-05 | 焦点科技股份有限公司 | A kind of method and apparatus for insuring electric business platform user attrition prediction |
-
2020
- 2020-01-02 CN CN202010000618.3A patent/CN111160992A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109583651A (en) * | 2018-12-03 | 2019-04-05 | 焦点科技股份有限公司 | A kind of method and apparatus for insuring electric business platform user attrition prediction |
Non-Patent Citations (1)
| Title |
|---|
| 陈捷: "《豆丁网》", 14 December 2019 * |
Cited By (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111738767A (en) * | 2020-06-23 | 2020-10-02 | 江苏云柜网络技术有限公司 | Express delivery operation service providing method and device, electronic equipment and storage medium |
| CN111861697A (en) * | 2020-07-02 | 2020-10-30 | 北京睿知图远科技有限公司 | Loan multi-head data-based user portrait generation method and system |
| CN111931107A (en) * | 2020-07-31 | 2020-11-13 | 上海博泰悦臻电子设备制造有限公司 | Digital citizen system construction method, system and storage medium |
| CN111931107B (en) * | 2020-07-31 | 2024-03-22 | 博泰车联网科技(上海)股份有限公司 | Digital citizen system construction method, system and storage medium |
| CN111897963A (en) * | 2020-08-06 | 2020-11-06 | 沈鑫 | Commodity classification method based on text information and machine learning |
| CN112000849A (en) * | 2020-08-21 | 2020-11-27 | 河南中原消费金融股份有限公司 | Unified label library management method, device, equipment and storage medium |
| WO2022041403A1 (en) * | 2020-08-26 | 2022-03-03 | 中山世达模型制造有限公司 | Sales order prediction method |
| CN112001757A (en) * | 2020-08-26 | 2020-11-27 | 中山世达模型制造有限公司 | Sales order prediction method |
| CN112215655A (en) * | 2020-10-13 | 2021-01-12 | 广东电网有限责任公司 | Client portrait label management method and system |
| CN112215655B (en) * | 2020-10-13 | 2023-04-28 | 广东电网有限责任公司 | Label management method and system for customer portrait |
| CN112269805A (en) * | 2020-11-18 | 2021-01-26 | 杭州米雅信息科技有限公司 | Data processing method, device, equipment and medium |
| CN112365073A (en) * | 2020-11-18 | 2021-02-12 | 贵州电网有限责任公司 | Regulation and control operation signal monitoring method based on big data |
| CN112269805B (en) * | 2020-11-18 | 2023-06-02 | 杭州米雅信息科技有限公司 | Data processing method, device, equipment and medium |
| CN112418929B (en) * | 2020-11-23 | 2023-09-19 | 中国工商银行股份有限公司 | Data sharing method and device |
| CN112418929A (en) * | 2020-11-23 | 2021-02-26 | 中国工商银行股份有限公司 | Data sharing method and device |
| CN112529624B (en) * | 2020-12-15 | 2024-01-09 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for generating business prediction model |
| CN112529624A (en) * | 2020-12-15 | 2021-03-19 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for generating business prediction model |
| CN112734485A (en) * | 2021-01-13 | 2021-04-30 | 上海群之脉信息科技有限公司 | User intelligent operation system |
| CN112860851A (en) * | 2021-01-22 | 2021-05-28 | 平安科技（深圳）有限公司 | Course recommendation method, device, equipment and medium based on root cause analysis |
| CN112860851B (en) * | 2021-01-22 | 2022-05-06 | 平安科技（深圳）有限公司 | Course recommendation method, device, equipment and medium based on root cause analysis |
| CN112765146B (en) * | 2021-01-26 | 2022-10-21 | 四川新网银行股份有限公司 | Method for monitoring data quality of user portrait label |
| CN112765146A (en) * | 2021-01-26 | 2021-05-07 | 四川新网银行股份有限公司 | Method for monitoring data quality of user portrait label |
| CN112862542A (en) * | 2021-03-10 | 2021-05-28 | 郑州时空隧道信息技术有限公司 | Marketing method and system |
| CN113139750A (en) * | 2021-05-14 | 2021-07-20 | 中国平安人寿保险股份有限公司 | Course recommendation method, device, server and storage medium |
| CN113139750B (en) * | 2021-05-14 | 2024-04-09 | 中国平安人寿保险股份有限公司 | Course recommendation method and device, server and storage medium |
| CN113076382A (en) * | 2021-06-07 | 2021-07-06 | 北京明略软件系统有限公司 | User label generation method and device, electronic equipment and readable storage medium |
| CN113421133A (en) * | 2021-07-21 | 2021-09-21 | 赛诺数据科技(南京)有限公司 | Network marketing system based on customer matching |
| CN113347213A (en) * | 2021-08-05 | 2021-09-03 | 环球数科集团有限公司 | Trusted channel authentication system based on protection of sensitive data of evanescent member |
| CN113722305A (en) * | 2021-08-26 | 2021-11-30 | 广东电网有限责任公司 | Analysis application system and method |
| CN113965416A (en) * | 2021-12-21 | 2022-01-21 | 江苏移动信息系统集成有限公司 | Website security protection capability scheduling method and system based on workflow |
| CN114912946B (en) * | 2022-04-24 | 2024-01-30 | 零犀(北京)科技有限公司 | Method and device for determining user layering, storage medium and electronic equipment |
| CN114912946A (en) * | 2022-04-24 | 2022-08-16 | 零犀(北京)科技有限公司 | Method, device, storage medium and electronic equipment for determining user hierarchy |
| CN115357802A (en) * | 2022-10-19 | 2022-11-18 | 山东省商业集团有限公司 | Multi-state enterprise potential customer identification method |
| CN115757396A (en) * | 2022-10-25 | 2023-03-07 | 杭州比智科技有限公司 | Oneid project implementation method and oneid project implementation system |
| CN115829615A (en) * | 2023-01-05 | 2023-03-21 | 瓴创(北京)科技有限公司 | User grouping method, system and storage medium based on multiple databases |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111160992A (en) | Marketing system based on user portrait system | |
| JP7043786B2 (en) | Sales activity support system, sales activity support method and sales activity support program | |
| Butler et al. | Family forest owner characteristics shaped by life cycle, cohort, and period effects | |
| US20090125382A1 (en) | Quantifying a Data Source's Reputation | |
| Kozitsin et al. | Modeling political preferences of russian users exemplified by the social network Vkontakte | |
| US11663839B1 (en) | Polarity semantics engine analytics platform | |
| Mau et al. | Forecasting the next likely purchase events of insurance customers: A case study on the value of data-rich multichannel environments | |
| CN112612942B (en) | Social big data-based fund recommendation system and method | |
| CN112837087A (en) | User portrait construction method oriented to consultation service system | |
| CN113742492A (en) | Insurance scheme generation method and device, electronic equipment and storage medium | |
| Darwiesh et al. | Business intelligence for risk management: A review | |
| Akila et al. | Opinion mining on food services using topic modeling and machine learning algorithms | |
| CN112330404A (en) | Data processing method and device, server and storage medium | |
| Tyagi et al. | Determinants of continuous usage of library resources on handheld devices: findings from PLS-SEM and fuzzy sets (fsQCA) | |
| Hu et al. | Identifying high quality ideas in the online context: evidence from a meta-analysis | |
| Fainshtein et al. | Using intelligent text analysis of online reviews to determine the main factors of restaurant value propositions | |
| Setiono et al. | Automatic knowledge extraction from survey data: Learning M-of-N constructs using a hybrid approach | |
| Wang et al. | Construction of a novel production develop decision model based on text mined | |
| Valsamidis et al. | Knowledge management for business intelligence measurement in an e-business system | |
| KR20220108970A (en) | Method of risk analysis and technology transfer opportunity recommendation by technology-product-material based on machine learning or AI | |
| US8589209B2 (en) | System and method for assessing viability and marketability of assets | |
| JP7306513B2 (en) | Sales activity support system, sales activity support method and sales activity support program | |
| Esmaelnezhad et al. | A fuzzy hybrid approach to analyse digital marketing strategies towards tourism industry | |
| Gehlhaus et al. | The US AI workforce | |
| Guseva et al. | Highly Pertinent Algorithm for the Market of Business Intelligence, Context and Native Advertising |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200515 |
|
| RJ01 | Rejection of invention patent application after publication |