Method and system for identifying real cigarette abnormal flow retailers

Abstract

The invention relates to the technical field of machine learning, and particularly provides a method and a system for identifying abnormal mobile retailers of real cigarettes. Firstly, data resources in all aspects are managed in a centralized manner and stored in a unified manner, a data mart is established, and the abnormal mobile retailers of the real cigarettes are identified according to the characteristics of outflow of the tobacco cigarettes; quantifiable outflow features are combed in the whole process of cigarette orders, distribution, sales, inventory, historical outflow and cigarette-related case terminal sales, and data support is provided for illegal circulation early warning of real cigarettes. Compared with the prior art, the method has the advantages that the real smoke outflow prediction decision tree model can be established, and the accuracy of the prediction model is improved through training of machine learning and continuous optimization and improvement of the decision tree. Technical support is provided for business personnel, and illegal circulation of real cigarettes is effectively prevented.

Description

Technical Field

[0001] The present invention relates to the field of machine learning technology, and specifically provides a method and system for identifying abnormal mobile retailers of genuine cigarettes. Background Art

[0002] Currently, the focus is on analyzing and rectifying leaked cigarettes and retailers after genuine cigarettes leak out, and there is a lack of forward-looking prevention of genuine cigarettes leaking out. The daily supervision of retailers is also mainly from the terminal, marketing, monopoly, internal management and other business areas, sorting out the early warning indicators of illegal circulation of genuine cigarettes, formulating early warning rules, and issuing regular warnings. For retailers that trigger the early warning rules, account managers are required to visit offline to determine whether the early warning results are misjudged. The work is cumbersome and inefficient.

[0003] How to use cutting-edge technologies such as machine learning and artificial intelligence to analyze and rectify leaked cigarettes and retailers after real cigarette outflow occurs, build a real cigarette outflow prediction decision tree model, and accurately grasp the trend of cigarette outflow is an urgent problem that technical personnel in this field need to solve. Summary of the invention

[0004] The present invention aims at the above-mentioned deficiencies of the prior art and provides a method for identifying abnormal mobile retailers of genuine cigarettes with strong practicability.

[0005] A further technical task of the present invention is to provide a system for identifying abnormal mobile retailers of genuine cigarettes that is reasonably designed, safe and applicable.

[0006] The technical solution adopted by the present invention to solve its technical problem is:

[0007] A method for identifying retailers with abnormal flow of genuine cigarettes. First, various data resources are centrally managed and uniformly stored to establish a data mart. According to the characteristics of tobacco cigarette outflow, quantifiable outflow characteristics are sorted out from the entire process of cigarette orders, distribution, sales, inventory, historical outflows and terminal sales of tobacco-related cases to provide data support for early warning of illegal circulation of genuine cigarettes.

[0008] Furthermore, a decision tree model for predicting the outflow of genuine cigarettes is established based on the data mart, and big data and machine learning technologies are used to conduct in-depth analysis and mining of the data. Machine autonomous learning is carried out based on business data and outflow characteristics, so as to timely discover potential risks of illegal circulation of genuine cigarettes, push them to regulators for prevention and control, and adjust the outflow risk coefficient according to the processing results.

[0009] Furthermore, the real smoke outflow prediction decision tree model is:

[0010] (1) Use data resources to extract common outflow features, remove or correct errors in the data, form an outflow feature library, and provide data support for early warning of illegal circulation of genuine cigarettes;

[0011] (2) Use decision tree binary search algorithm;

[0012] (3) Use machine learning methods.

[0013] Furthermore, in step (2), a decision tree binary search algorithm is used to generate a retailer portrait in combination with the market status of cigarette specifications, and the retailer's probability of genuine cigarette outflow tendency is calculated.

[0014] Furthermore, in step (3), machine learning is used to optimize the decision tree model through the investigation and punishment results continuously fed back by market supervision, dynamically adjust the outflow risk coefficient, and screen out invalid feature indicators.

[0015] A system for identifying retailers with abnormal flows of genuine cigarettes. First, it centrally manages and uniformly stores various data resources to establish a data mart. Based on the characteristics of tobacco cigarette outflow, it sorts out quantifiable outflow characteristics from the entire process of cigarette orders, distribution, sales, inventory, historical outflows, and terminal sales in tobacco-related cases, providing data support for early warning of illegal circulation of genuine cigarettes.

[0016] Furthermore, a decision tree model for predicting the outflow of genuine cigarettes is established based on the data mart, and big data and machine learning technologies are used to conduct in-depth analysis and mining of the data. Machine autonomous learning is carried out based on business data and outflow characteristics, so as to timely discover potential risks of illegal circulation of genuine cigarettes, push them to regulators for prevention and control, and adjust the outflow risk coefficient according to the processing results.

[0017] Furthermore, the real smoke outflow prediction decision tree model is:

[0018] (1) Use data resources to extract common outflow features, remove or correct errors in the data, form an outflow feature library, and provide data support for early warning of illegal circulation of genuine cigarettes;

[0019] (2) Use decision tree binary search algorithm;

[0020] (3) Use machine learning methods.

[0021] Furthermore, in step (2), a decision tree binary search algorithm is used to generate a retailer portrait in combination with the market status of cigarette specifications, and the retailer's probability of genuine cigarette outflow tendency is calculated.

[0022] In step (3), machine learning is used to optimize the decision tree model through the continuous feedback of investigation and punishment results from market supervision, dynamically adjust the outflow risk coefficient, and screen out invalid feature indicators.

[0023] Compared with the prior art, the method and system for identifying abnormal mobile retailers of genuine cigarettes of the present invention have the following outstanding beneficial effects:

[0024] The present invention establishes a real cigarette outflow prediction decision tree model, and improves the accuracy of the prediction model through machine learning training and continuous optimization and improvement of the decision tree, providing technical support for business personnel and effectively preventing the illegal circulation of real cigarettes. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0026] Attached Figure 1 It is a flow chart of a method for identifying abnormal mobile retailers of genuine cigarettes;

[0027] Attached Figure 2 A schematic diagram of a decision tree model in a method for identifying abnormal mobile retailers of genuine cigarettes. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the solution of the present invention, the present invention is further described in detail below in conjunction with specific implementation methods. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] A best embodiment is given below:

[0030] like Figure 1-2 As shown, in this embodiment, a method for identifying abnormal mobile retailers of genuine cigarettes is firstly to centrally manage and uniformly store various data resources, establish a data mart, and based on the characteristics of tobacco cigarette outflow, quantifiable outflow characteristics are sorted out from the entire process of cigarette orders, distribution, sales, inventory, historical outflows and terminal sales of tobacco-related cases to provide data support for the early warning of illegal circulation of genuine cigarettes.

[0031] Based on the data mart, a decision tree model for predicting the outflow of genuine cigarettes is established. Big data and machine learning technology are used to conduct in-depth analysis and mining of data. Based on business data and outflow characteristics, machine autonomous learning is carried out to timely discover potential risks of illegal circulation of genuine cigarettes, push them to supervisors for prevention and control, and adjust the outflow risk coefficient according to the processing results. Effectively prevent the illegal circulation of genuine cigarettes and improve the level of standardized operations.

[0032] Among them, the real smoke outflow prediction decision tree model is:

[0033] (1) Use data resources to extract common outflow features, remove or correct errors in the data, such as missing values ​​or outliers, and form an outflow feature library to provide effective data support for the early warning of illegal circulation of genuine cigarettes.

[0034] (2) Using the decision tree binary search algorithm and the market status of cigarette specifications, a retailer portrait is generated and the probability of the retailer's tendency to leak genuine cigarettes is calculated.

[0035] (3) Use machine learning methods to optimize the decision tree model through continuous feedback from market supervision on investigation and punishment results, dynamically adjust the outflow risk coefficient, and screen out invalid feature indicators.

[0036] Based on the above method, a system for identifying abnormal mobile retailers of genuine cigarettes in this embodiment first centrally manages and uniformly stores various data resources to establish a data mart. According to the characteristics of tobacco cigarette outflow, quantifiable outflow characteristics are sorted out from the entire process of cigarette orders, distribution, sales, inventory, historical outflows and terminal sales of tobacco-related cases to provide data support for the early warning of illegal circulation of genuine cigarettes.

[0037] Among them, a decision tree model for predicting the outflow of genuine cigarettes is established based on the data mart, and big data and machine learning technologies are used to conduct in-depth analysis and mining of the data. Machine autonomous learning is carried out based on business data and outflow characteristics to timely discover potential risks of illegal circulation of genuine cigarettes, push them to regulators for prevention and control, and adjust the outflow risk coefficient based on the processing results.

[0038] The real smoke outflow prediction decision tree model is:

[0039] (1) Use data resources to extract common outflow features, remove or correct errors in the data, form an outflow feature library, and provide data support for early warning of illegal circulation of genuine cigarettes;

[0040] (2) Use decision tree binary search algorithm;

[0041] (3) Use machine learning methods.

[0042] In step (2), a decision tree binary search algorithm is used to generate a retailer portrait in combination with the market status of cigarette specifications, and the retailer's probability of genuine cigarette outflow tendency is calculated.

[0043] In step (3), machine learning is used to optimize the decision tree model through the continuous feedback of investigation and punishment results from market supervision, dynamically adjust the outflow risk coefficient, and screen out invalid feature indicators.

[0044] The present invention further improves the accuracy of real cigarette outflow warning by conducting deep learning on abnormal retailers after field verification and optimizing the prediction decision tree model, effectively preventing the illegal circulation of real cigarettes, purifying the market environment, and creating a fair market competition situation. At the same time, it improves the level of standardized operation, enhances the core competitiveness of enterprises, and contributes to the sustainable development of the industry.

[0045] Establish a decision tree model for predicting the outflow of genuine cigarettes, and improve the accuracy of the prediction model through machine learning training and continuous optimization and improvement of the decision tree. Provide technical support for business personnel and effectively prevent the illegal circulation of genuine cigarettes.

[0046] The above-mentioned specific implementations are only specific cases of the present invention. The patent protection scope of the present invention includes but is not limited to the above-mentioned specific implementations. Any technical solutions that conform to the above-mentioned specific implementations of the present invention and any appropriate changes or substitutions made by ordinary technicians in the relevant technical field shall fall within the patent protection scope of the present invention.

[0047] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for identifying abnormal mobile retailers of genuine cigarettes, characterized in that: First, we will centrally manage and uniformly store data resources from all aspects, establish a data mart, and based on the characteristics of tobacco cigarette outflow, we will sort out quantifiable outflow characteristics from the entire process of cigarette orders, distribution, sales, inventory, historical outflows, and terminal sales in tobacco-related cases to provide data support for early warning of illegal circulation of genuine cigarettes.

2. A method for identifying abnormal mobile retailers of genuine cigarettes according to claim 1, characterized in that: Based on the data mart, a decision tree model for predicting the outflow of genuine cigarettes is established. Big data and machine learning technologies are used to conduct in-depth analysis and mining of the data. Machine autonomous learning is carried out based on business data and outflow characteristics to timely discover potential risks of illegal circulation of genuine cigarettes, push them to regulators for prevention and control, and adjust the outflow risk coefficient based on the processing results.

3. A method for identifying abnormal mobile retailers of genuine cigarettes according to claim 2, characterized in that: The real smoke outflow prediction decision tree model is: (1) Use data resources to extract common outflow features, remove or correct errors in the data, form an outflow feature library, and provide data support for early warning of illegal circulation of genuine cigarettes; (2) Use decision tree binary search algorithm; (3) Use machine learning methods.

4. A method for identifying abnormal mobile retailers of genuine cigarettes according to claim 3, characterized in that: In step (2), a decision tree binary search algorithm is used to generate a retailer portrait in combination with the market status of cigarette specifications, and the retailer's probability of genuine cigarette outflow tendency is calculated.

5. The method for identifying abnormal mobile retailers of genuine cigarettes according to claim 3, characterized in that: In step (3), machine learning is used to optimize the decision tree model through the continuous feedback of investigation and punishment results from market supervision, dynamically adjust the outflow risk coefficient, and screen out invalid feature indicators.

6. A system for identifying abnormal mobile retailers of genuine cigarettes, characterized in that: First, we will centrally manage and uniformly store data resources from all aspects, establish a data mart, and based on the characteristics of tobacco cigarette outflow, we will sort out quantifiable outflow characteristics from the entire process of cigarette orders, distribution, sales, inventory, historical outflows, and terminal sales in tobacco-related cases to provide data support for early warning of illegal circulation of genuine cigarettes.

7. A system for identifying abnormal mobile retailers of genuine cigarettes according to claim 6, characterized in that: Based on the data mart, a decision tree model for predicting the outflow of genuine cigarettes is established. Big data and machine learning technologies are used to conduct in-depth analysis and mining of the data. Machine autonomous learning is carried out based on business data and outflow characteristics to timely discover potential risks of illegal circulation of genuine cigarettes, push them to regulators for prevention and control, and adjust the outflow risk coefficient based on the processing results.

8. A system for identifying abnormal mobile retailers of genuine cigarettes according to claim 7, characterized in that: The real smoke outflow prediction decision tree model is: (1) Use data resources to extract common outflow features, remove or correct errors in the data, form an outflow feature library, and provide data support for early warning of illegal circulation of genuine cigarettes; (2) Use decision tree binary search algorithm; (3) Use machine learning methods.

9. A system for identifying abnormal mobile retailers of genuine cigarettes according to claim 8, characterized in that: In step (2), a decision tree binary search algorithm is used to generate a retailer portrait in combination with the market status of cigarette specifications, and the retailer's probability of genuine cigarette outflow tendency is calculated.

10. A system for identifying abnormal mobile retailers of genuine cigarettes according to claim 9, characterized in that: In step (3), machine learning is used to optimize the decision tree model through the continuous feedback of investigation and punishment results from market supervision, dynamically adjust the outflow risk coefficient, and screen out invalid feature indicators.

Images