Behavior prediction method and device combining Apriori algorithm and FP-Growth algorithm
By combining the Apriori algorithm and the FP-GROWTH algorithm, customer behavior is deeply excavated, and accurate repurchase and cross-purchase prediction information is generated, which solves the problem of low efficiency of the existing customer wake-up method, and improves customer wake-up efficiency and customer value.
Patent Information
- Application Number
- CN202411293314.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-14
- Publication Date
- 2025-01-28
AI Technical Summary
The existing customer wake-up method relies on the experience and simple analysis of salesmen, and lacks in-depth exploration and accurate prediction of customer behavior, resulting in inefficient customer wake-up and difficulty in achieving effective repurchase and cross-selling.
Combined with the Apriori algorithm and the FP-GROWTH algorithm, obtain the relevant information and data of the customer, convert it into a transaction form, generate frequent item sets and related rules, and adjust the support and confidence threshold of knowledge adjustment in the field of insurance in the field of insurance. Prediction information for repurchase and cross -purchase behavior.
Through in -depth mining of customer behavior, provide accurate customer portrait and action suggestions, improve customer wake -up efficiency, reduce customer loss rate, promote cross -sales, and increase customer value.
Smart Images
Figure HDA0005045228540000011 
Figure HDA0005045228540000021 
Figure HDA0005045228540000031
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data mining, and in particular to a behavior prediction method and device combining an Apriori algorithm and an FP-Growth algorithm. Background Art
[0002] When insurance companies are managing customer relationships, they often neglect to effectively awaken and guide "lonely customers". Existing methods of awakening customers mainly rely on the experience of salespeople and simple analysis of customer information, lacking in-depth mining and accurate prediction of customer behavior, resulting in low efficiency of awakening customers and difficulty in achieving effective repurchase and cross-selling.
[0003] Therefore, a behavior prediction method and device combining the Apriori algorithm and the FP-Growth algorithm are provided to solve the above problems. Summary of the invention
[0004] The main purpose of the present invention is to solve the problem in the prior art that the customer awakening method mainly relies on the experience of the salesperson and a simple analysis of customer information, lacks in-depth mining and accurate prediction of customer behavior, resulting in low customer awakening efficiency and difficulty in achieving effective repurchase and cross-selling.
[0005] A first aspect of the present invention provides a behavior prediction method combining an Apriori algorithm and an FP-Growth algorithm, wherein the behavior prediction method combining an Apriori algorithm and an FP-Growth algorithm comprises:
[0006] Obtain relevant information data of a single customer and convert the relevant information data of each customer into Transaction form;
[0007] Use Apriori algorithm and FP-Growth algorithm to mine relevant customer information data and generate frequent item sets and association rules;
[0008] Obtain business demand information and insurance domain knowledge information, generate support and confidence thresholds based on the business demand information and insurance domain knowledge information, and select association rules based on the support and confidence thresholds;
[0009] Based on association rules, the repurchase behavior prediction information and cross-purchase behavior prediction information of a single customer are generated.
[0010] Furthermore, the relevant information data includes: basic customer information, purchase record information, claim record information and communication record information.
[0011] Furthermore, the basic customer information includes customer age information, customer gender information, customer marital status information, customer work information, customer education information and customer family status information;
[0012] The purchase record information includes insurance type information, amount information, term information and number of repurchases information.
[0013] Furthermore, the use of the Apriori algorithm and the FP-Growth algorithm to mine the relevant information data of the customer to generate frequent item sets and association rules includes:
[0014] The Apriori algorithm is used to iteratively generate candidate frequent item sets, and the support threshold is used to filter out frequent item sets, and association rules are generated based on frequent item sets;
[0015] The FP-Growth algorithm is used to construct an FP tree to find frequent item sets and association rules.
[0016] Furthermore, the generation of the repurchase behavior prediction information and cross-purchase behavior prediction information of a single customer based on the association rules also includes:
[0017] A customer behavior prediction table is generated based on the cross-purchase behavior prediction information of a single customer's repurchase behavior prediction information set, and the customer behavior prediction table is sent to the sales business end.
[0018] Furthermore, it also includes:
[0019] Evaluate the repurchase behavior prediction information and cross-purchase behavior prediction information of a single customer and generate evaluation result information;
[0020] Adjust the support and confidence thresholds according to the evaluation result information; send the adjusted support and confidence thresholds to the step "obtain business demand information and insurance field knowledge information, generate support and confidence thresholds according to the business demand information and insurance field knowledge information, and filter out association rules according to the support and confidence thresholds".
[0021] A second aspect of the present invention provides a behavior prediction device combining the Apriori algorithm and the FP-Growth algorithm, comprising:
[0022] The customer information acquisition module is used to obtain the relevant information data of a single customer and convert the relevant information data of each customer into a Transaction form;
[0023] The association rule generation module is used to mine the relevant information data of customers using the Apriori algorithm and the FP-Growth algorithm to generate frequent item sets and association rules;
[0024] The association rule screening module is used to obtain business demand information and insurance domain knowledge information, generate support and confidence thresholds based on the business demand information and insurance domain knowledge information, and screen association rules based on the support and confidence thresholds;
[0025] The customer behavior prediction module is used to generate a single customer's repurchase behavior prediction information and cross-purchase behavior prediction information based on association rules.
[0026] A third aspect of the present invention provides an electronic device, the electronic device comprising a memory and at least one processor, the memory storing instructions;
[0027] The at least one processor calls the instructions in the memory so that the electronic device executes the various steps of the behavior prediction method combining the Apriori algorithm and the FP-Growth algorithm as described above.
[0028] A fourth aspect of the present invention provides a computer-readable storage medium having instructions stored thereon, which, when executed by a processor, implement the various steps of the behavior prediction method combining the Apriori algorithm and the FP-Growth algorithm as described above.
[0029] The technical solution of the present invention provides a behavior prediction algorithm that combines the Apriori algorithm and the FP-Growth algorithm. By mining association rules, it can effectively predict the probability of repurchase and cross-purchase of isolated customers, and provide sales staff with accurate customer portraits and action suggestions. Based on the prediction results, sales staff can awaken customers in a more targeted manner, improve customer awakening efficiency, and reduce customer churn rate. By mining the correlation between customer purchase behaviors, cross-selling can be effectively promoted and customer value can be increased. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 A flowchart of a first embodiment of a behavior prediction method combining the Apriori algorithm and the FP-Growth algorithm provided in an embodiment of the present invention;
[0031] Figure 2 A flowchart of a second embodiment of a behavior prediction method combining the Apriori algorithm and the FP-Growth algorithm provided in an embodiment of the present invention;
[0032] Figure 3 A flowchart of a third embodiment of a behavior prediction method combining the Apriori algorithm and the FP-Growth algorithm provided in an embodiment of the present invention;
[0033] Figure 4A schematic diagram of the structure of a behavior prediction device combining the Apriori algorithm and the FP-Growth algorithm provided in an embodiment of the present invention;
[0034] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0035] The embodiment of the present invention provides a behavior prediction method combining the Apriori algorithm and the FP-Growth algorithm, including obtaining relevant information data of a single customer, converting the relevant information data of each customer into a Transaction form; using the Apriori algorithm and the FP-Growth algorithm to mine the relevant information data of the customer, generating frequent item sets and association rules; obtaining business demand information and insurance field knowledge information, generating support and confidence thresholds according to the business demand information and the insurance field knowledge information, and selecting association rules according to the support and confidence thresholds; based on the association rules, generating repurchase behavior prediction information and cross-purchase behavior prediction information of a single customer. The present invention solves the problems in the prior art that the customer awakening method mainly relies on the experience of the salesperson and the simple analysis of the customer information, lacks in-depth mining and accurate prediction of the customer behavior, resulting in low efficiency of customer awakening, and difficulty in achieving effective repurchase and cross-selling.
[0036] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0037] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 The first embodiment of the behavior prediction method combining the Apriori algorithm and the FP-Growth algorithm in the embodiment of the present invention includes:
[0038] Obtain relevant information data of a single customer, and convert the relevant information data of each customer into Transaction form; the relevant information data includes: basic customer information, purchase record information, claim record information and communication record information; the basic customer information includes customer age information, customer gender information, customer marital status information, customer work information, customer education information and customer family status information; the purchase record information includes insurance type information, amount information, years information and number of repurchases; convert the information into Transaction form to facilitate association rule mining.
[0039] Use Apriori algorithm and FP-Growth algorithm to mine relevant customer information data and generate frequent item sets and association rules;
[0040] Specifically, it includes: using the Apriori algorithm to iteratively generate candidate frequent item sets, and using the support threshold to filter out frequent item sets, and generating association rules based on frequent item sets;
[0041] The FP-Growth algorithm is used to construct an FP tree to find frequent item sets and association rules.
[0042] Obtain business demand information and insurance domain knowledge information, generate support and confidence thresholds based on the business demand information and insurance domain knowledge information, and select association rules based on the support and confidence thresholds;
[0043] Based on association rules, we generate prediction information on a single customer’s repurchase behavior and cross-purchase behavior. Based on the mined association rules, we design an effective prediction model that can accurately predict a customer’s repurchase and cross-purchase behavior.
[0044] See also Figure 2 The second embodiment of the behavior prediction method combining the Apriori algorithm and the FP-Growth algorithm in the embodiment of the present invention includes:
[0045] Obtain relevant information data of a single customer, and convert the relevant information data of each customer into Transaction form; the relevant information data includes: basic customer information, purchase record information, claim record information and communication record information; the basic customer information includes customer age information, customer gender information, customer marital status information, customer work information, customer education information and customer family status information; the purchase record information includes insurance type information, amount information, years information and number of repurchases; convert the information into Transaction form to facilitate association rule mining.
[0046] Use Apriori algorithm and FP-Growth algorithm to mine relevant customer information data and generate frequent item sets and association rules;
[0047] Specifically, it includes: using the Apriori algorithm to iteratively generate candidate frequent item sets, and using the support threshold to filter out frequent item sets, and generating association rules based on frequent item sets;
[0048] The FP-Growth algorithm is used to construct an FP tree to find frequent item sets and association rules.
[0049] Obtain business demand information and insurance domain knowledge information, generate support and confidence thresholds based on the business demand information and insurance domain knowledge information, and select association rules based on the support and confidence thresholds;
[0050] Generate prediction information on a single customer's repurchase behavior and cross-purchase behavior based on association rules. Design an effective prediction model based on the mined association rules to accurately predict a customer's repurchase and cross-purchase behavior.
[0051] The generating of the repurchase behavior prediction information and cross-purchase behavior prediction information of a single customer based on the association rules also includes:
[0052] A customer behavior prediction table is generated based on the cross-purchase behavior prediction information of a single customer's repurchase behavior prediction information set, and the customer behavior prediction table is sent to the sales business end.
[0053] See also Figure 3 The third embodiment of the behavior prediction method combining the Apriori algorithm and the FP-Growth algorithm in the embodiment of the present invention includes:
[0054] Obtain relevant information data of a single customer, and convert the relevant information data of each customer into Transaction form; the relevant information data includes: basic customer information, purchase record information, claim record information and communication record information; the basic customer information includes customer age information, customer gender information, customer marital status information, customer work information, customer education information and customer family status information; the purchase record information includes insurance type information, amount information, years information and number of repurchases; convert the information into Transaction form to facilitate association rule mining.
[0055] Use Apriori algorithm and FP-Growth algorithm to mine relevant customer information data and generate frequent item sets and association rules;
[0056] Specifically, it includes: using the Apriori algorithm to iteratively generate candidate frequent item sets, and using the support threshold to filter out frequent item sets, and generating association rules based on frequent item sets;
[0057] The FP-Growth algorithm is used to construct an FP tree to find frequent item sets and association rules.
[0058] Obtain business demand information and insurance domain knowledge information, generate support and confidence thresholds based on the business demand information and insurance domain knowledge information, and select association rules based on the support and confidence thresholds;
[0059] Generate prediction information on a single customer's repurchase behavior and cross-purchase behavior based on association rules. Design an effective prediction model based on the mined association rules to accurately predict a customer's repurchase and cross-purchase behavior.
[0060] The generating of the repurchase behavior prediction information and cross-purchase behavior prediction information of a single customer based on the association rules also includes:
[0061] Generate a customer behavior prediction table based on the cross-purchase behavior prediction information of a single customer's repurchase behavior prediction information set, and send the customer behavior prediction table to the sales business end;
[0062] It also includes: evaluating the repurchase behavior prediction information and cross-purchase behavior prediction information of a single customer and generating evaluation result information;
[0063] Adjust the support and confidence thresholds according to the evaluation result information; send the adjusted support and confidence thresholds to the step "obtain business demand information and insurance field knowledge information, generate support and confidence thresholds according to the business demand information and insurance field knowledge information, and filter out association rules according to the support and confidence thresholds".
[0064] The above describes the behavior prediction method combining the Apriori algorithm and the FP-Growth algorithm in the embodiment of the present invention. The following describes the behavior prediction device combining the Apriori algorithm and the FP-Growth algorithm in the embodiment of the present invention. Figure 4 In the embodiment of the present invention, the behavior prediction device combining the Apriori algorithm and the FP-Growth algorithm includes:
[0065] The customer information acquisition module 401 is used to acquire the relevant information data of a single customer and convert the relevant information data of each customer into a Transaction form;
[0066] The association rule generation module 402 is used to mine the relevant information data of the customer using the Apriori algorithm and the FP-Growth algorithm to generate frequent item sets and association rules;
[0067] The association rule screening module 403 is used to obtain business demand information and insurance domain knowledge information, generate support and confidence thresholds according to the business demand information and insurance domain knowledge information, and screen association rules according to the support and confidence thresholds;
[0068] The customer behavior prediction module 404 is used to generate the repurchase behavior prediction information and cross-purchase behavior prediction information of a single customer based on association rules.
[0069] above Figure 4 The behavior prediction device combining the Apriori algorithm and the FP-Growth algorithm in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The electronic device in the embodiment of the present invention is described in detail from the perspective of hardware processing.
[0070] Figure 5 7 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device 700 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 710 (for example, one or more processors) and a memory 720, and one or more storage media 730 (for example, one or more mass storage devices) storing application programs 733 or data 732. Among them, the memory 720 and the storage medium 730 can be temporary storage or permanent storage. The program stored in the storage medium 730 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the electronic device 700. Furthermore, the processor 710 can be configured to communicate with the storage medium 730 to execute a series of instruction operations in the storage medium 730 on the electronic device 700.
[0071] The electronic device 700 may also include one or more power supplies 740, one or more wired or wireless network interfaces 750, one or more input and output interfaces 750, and / or one or more operating systems 731, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. It will be appreciated by those skilled in the art that Figure 5 The structure of the electronic device shown does not constitute a limitation on the electronic device, and may include more or less components than shown in the figure, or combine some components, or arrange the components differently.
[0072] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of a behavior prediction method combining an Apriori algorithm and an FP-Growth algorithm.
[0073] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device, or unit can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.
[0074] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.
[0075] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A behavior prediction method combining Apriori algorithm and FP-Growth algorithm, characterized in that: The behavior prediction method combining the Apriori algorithm and the FP-Growth algorithm includes: Obtain relevant information data of a single customer and convert the relevant information data of each customer into Transaction form; Use Apriori algorithm and FP-Growth algorithm to mine relevant customer information data and generate frequent item sets and association rules; Obtain business demand information and insurance domain knowledge information, generate support and confidence thresholds based on the business demand information and insurance domain knowledge information, and select association rules based on the support and confidence thresholds; Based on association rules, the repurchase behavior prediction information and cross-purchase behavior prediction information of a single customer are generated.
2. The behavior prediction method combining the Apriori algorithm and the FP-Growth algorithm according to claim 1, characterized in that: The relevant information data includes: basic customer information, purchase record information, claim record information and communication record information.
3. The behavior prediction method combining Apriori algorithm and FP-Growth algorithm according to claim 1, characterized in that: The basic customer information includes customer age information, customer gender information, customer marital status information, customer work information, customer education information and customer family status information; The purchase record information includes insurance type information, amount information, term information and number of repurchases information.
4. The behavior prediction method combining Apriori algorithm and FP-Growth algorithm according to claim 1, characterized in that: The method of using the Apriori algorithm and the FP-Growth algorithm to mine the relevant information data of customers and generate frequent item sets and association rules includes: The Apriori algorithm is used to iteratively generate candidate frequent item sets, and the support threshold is used to filter out frequent item sets, and association rules are generated based on frequent item sets; The FP-Growth algorithm is used to construct an FP tree to find frequent item sets and association rules.
5. The behavior prediction method combining Apriori algorithm and FP-Growth algorithm according to claim 1, characterized in that: The generating of the repurchase behavior prediction information and cross-purchase behavior prediction information of a single customer based on the association rules also includes: A customer behavior prediction table is generated based on the cross-purchase behavior prediction information of a single customer's repurchase behavior prediction information set, and the customer behavior prediction table is sent to the sales business end.
6. The behavior prediction method combining Apriori algorithm and FP-Growth algorithm according to claim 1, characterized in that: Also includes: Evaluate the repurchase behavior prediction information and cross-purchase behavior prediction information of a single customer and generate evaluation result information; Adjust the support and confidence thresholds according to the evaluation result information; send the adjusted support and confidence thresholds to the step "obtaining business demand information and insurance field knowledge information, generating support and confidence thresholds according to the business demand information and insurance field knowledge information, and filtering association rules according to the support and confidence thresholds".
7. A behavior prediction device combining Apriori algorithm and FP-Growth algorithm, characterized in that: include: The customer information acquisition module is used to obtain the relevant information data of a single customer and convert the relevant information data of each customer into a Transaction form; The association rule generation module is used to mine the relevant information data of customers using the Apriori algorithm and the FP-Growth algorithm to generate frequent item sets and association rules; The association rule screening module is used to obtain business demand information and insurance domain knowledge information, generate support and confidence thresholds based on the business demand information and insurance domain knowledge information, and screen association rules based on the support and confidence thresholds; The customer behavior prediction module is used to generate a single customer's repurchase behavior prediction information and cross-purchase behavior prediction information based on association rules.
8. An electronic device, comprising a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instruction in the memory so that the electronic device executes each step of the behavior prediction method combining the Apriori algorithm and the FP-Growth algorithm as described in any one of claims 1 to 6.
9. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the steps of the behavior prediction method combining the Apriori algorithm and the FP-Growth algorithm as claimed in any one of claims 1 to 6 are implemented.