A method for constructing a portrait of a charging and swapping user based on multi-dimensional behavior data

Abstract

This invention belongs to the field of user profile construction technology, and specifically relates to a method for constructing charging and swapping user profiles based on multi-dimensional behavioral data. This method collects multi-dimensional behavioral data from the entire charging and swapping user process, completes data cleaning, fusion, and feature engineering processing, extracts user behavioral features from dimensions such as charging habits and swapping frequency, combines user demand and preference survey data to mine core needs and preference tendencies, and uses clustering algorithms to achieve precise user segmentation. This results in the construction of multi-dimensional and refined charging and swapping user profiles, and the establishment of a user value assessment model to complete value grading and potential analysis of different groups. This invention solves the problems of traditional charging and swapping user analysis having single dimensions, vague profiles, and poor targeting of marketing services. The constructed user profiles can directly provide data support for precise marketing, personalized service optimization, and resource allocation adjustment in the charging and swapping business, effectively improving the adaptability and commercial value of charging and swapping services.

Description

Technical Field

[0001] This invention belongs to, but is not limited to, the field of user profile construction technology, and particularly relates to a method for constructing charging and swapping user profiles based on multi-dimensional behavioral data. Background Technology

[0002] With the rapid development of the new energy vehicle industry, charging and battery swapping, as a core supporting service for new energy vehicles, has seen its market size continue to expand, and competition within the industry has become increasingly fierce. The user base for charging and battery swapping services exhibits diversified characteristics, with significant differences among users in terms of charging and battery swapping habits, needs, preferences, and purchasing power. However, traditional charging and battery swapping service operation models often adopt a "one-size-fits-all" marketing and service strategy, lacking refined analysis and precise operation of users.

[0003] Currently, most existing user analysis methods for charging and swapping devices only analyze user transaction data from a single dimension, resulting in the following problems: First, the data collection dimensions are limited, failing to integrate multi-dimensional data such as charging and swapping device usage, time, geography, and service evaluations, leading to an incomplete portrayal of user behavior; second, the depth of user feature mining is insufficient, focusing only on basic indicators such as consumption amount and usage frequency, without delving into core features such as user needs, preferences, and behavioral motivations; third, the user segmentation methods are crude, lacking scientific clustering algorithms, resulting in segmentation results without significant feature differences, failing to form refined user profiles; fourth, a user value evaluation system deeply integrated with business operations has not been established, and the formulation of marketing and service strategies lacks data support, leading to low resource allocation efficiency, poor user experience, and poor business conversion results.

[0004] Therefore, how to construct a scientific, detailed, and dynamic user profile for charging and swapping based on multi-dimensional user behavior data and user needs and preferences, while simultaneously achieving accurate assessment of user value, and providing a basis for precise marketing, personalized services, and optimized resource allocation in the charging and swapping business, has become a pressing technical problem that the current charging and swapping industry needs to solve. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a method for constructing user profiles for charging and swapping based on multi-dimensional behavioral data.

[0006] This invention is implemented as follows: a method for constructing charging and swapping user profiles based on multi-dimensional behavioral data, the method comprising:

[0007] S1. Multi-dimensional data collection and preprocessing: Collect behavioral data and demand preference survey data of charging and swapping users, clean, deduplicate, complete and standardize the raw data to obtain a structured dataset to be analyzed.

[0008] S2. User behavior feature extraction: Based on preprocessed behavior data, a behavior feature system is constructed from four core dimensions: charging habits, battery swapping frequency, usage time period, and consumption scenario, and feature quantification and encoding are completed.

[0009] S3. User demand and preference feature mining: Statistical analysis and correlation mining are performed on demand and preference survey data to extract three core demand and preference features: service quality demand, price sensitivity, and service experience preference, forming a demand and preference feature vector.

[0010] S4. Precise user segmentation: The user behavior feature vector is fused with the demand preference feature vector, and a clustering algorithm is used to perform cluster analysis on the fused feature dataset to obtain multiple charging and swapping user groups with significant feature differences.

[0011] S5. Multi-dimensional user profile construction: For each user group, feature summarization and tagging are carried out from three levels: basic attributes, behavioral characteristics, and demand preferences to build a structured and visualized charging and swapping user group profile, while generating a personalized profile for each user.

[0012] S6. User value assessment and strategy formulation: Establish a user value assessment model for charging and swapping, classify the value of each user group, analyze the commercial contribution and potential development space of different groups, and formulate targeted marketing and service strategies based on user profile characteristics and value levels.

[0013] Furthermore, in step S1, the behavioral data of the charging and swapping users is full-process business data, including user charging and swapping transaction data, device usage data, geographical trajectory data, and time dimension data. The data collection methods include data collection from the charging and swapping platform backend, data collection from the charging and swapping device terminal, and data synchronization from the user terminal APP. The demand preference survey data includes online questionnaire survey data, offline interview data, and service evaluation feedback data. During the data collection process, user identity information is anonymized to ensure data compliance.

[0014] Furthermore, in step S1, the specific process of data preprocessing is as follows: first, extreme abnormal data in the dataset is removed by an outlier detection algorithm, missing values ​​are filled in by interpolation, and duplicate data is deleted;

[0015] Then, the unstructured data is transformed into a structured data set, the numerical features are normalized / standardized, and the categorical features are one-hot encoded / label encoded, ultimately resulting in a dataset with uniform dimensions and standardized format for analysis.

[0016] Furthermore, in step S2, the specific dimensions and quantitative indicators of the behavioral characteristic system are as follows:

[0017] (1) Charging habit dimension: including average charging time, charging power preference, charging amount range, percentage of charging method selection, and percentage of fixed charging points;

[0018] (2) Battery swapping frequency dimension: including monthly average number of battery swaps, battery swapping interval days, battery swapping capacity preference, and battery swapping package purchase indicators;

[0019] (3) Usage time dimension: including peak hour usage ratio, off-peak hour usage ratio, low-peak hour usage ratio, weekend / weekday usage difference rate, and nighttime charging and swapping ratio;

[0020] (4) Consumption scenario dimension: including indicators such as the proportion of household scenarios, commercial scenarios, public scenarios, and the proportion of long-distance travel-related usage;

[0021] Each indicator is assigned a weight based on actual business needs, and a weighted summation method is used to construct the behavioral feature vector.

[0022] Furthermore, in step S3, the service quality demands include preference levels for charging and swapping equipment availability, service response speed, and fault handling efficiency; the price sensitivity includes a quantitative value of sensitivity to charging and swapping pricing, package discounts, and membership discounts, and a Likert scale is used to complete the sensitivity classification; the service experience preferences include preference tendencies for charging and swapping site environment, terminal operation convenience, online platform services, and value-added services, and the correlation between each experience factor and user behavior is obtained through association rule mining; finally, the three types of features are quantified and concatenated into a demand preference feature vector, which is matched with the dimension of the behavioral feature vector.

[0023] Furthermore, in step S4, the feature fusion adopts the feature concatenation method, which concatenates the user behavior feature vector and the demand preference feature vector in dimensional order into a unified comprehensive user feature vector; the clustering algorithm is an improved K-means clustering algorithm or DBSCAN density clustering algorithm. Before clustering, the optimal number of clusters is determined by the silhouette coefficient method and the elbow rule. During the clustering process, the feature vector is dimensionality reduced to improve the clustering efficiency and accuracy.

[0024] Furthermore, in step S5, the tagging process adopts a three-level tagging system. The first-level tags include basic attribute tags, behavioral feature tags, and demand preference tags; the second-level tags are the subdivisions of the first-level tags; and the third-level tags are specific feature quantification values ​​or descriptive tags. The user group profile is visualized in the form of tag cloud, feature radar chart, and data report. The personalized profile of a single user is generated based on the profile of the user's group and its own unique characteristics, so as to achieve full coverage of the profile of "group commonality + individual individuality".

[0025] Furthermore, in step S6, the charging and swapping user value assessment model is constructed using the RFM model combined with business-customized indicators. Based on the traditional R (most recent consumption), F (consumption frequency), and M (consumption amount) indicators, it adds charging and swapping business-specific indicators: user lifecycle, platform stickiness, potential consumption capacity, and word-of-mouth value. The weights of each indicator are determined using the analytic hierarchy process, and the comprehensive value score of each user group is calculated using a weighted scoring method, dividing the user group into three levels: high-value core users, medium-value potential users, and low-value ordinary users.

[0026] Furthermore, in step S6, the principles for formulating targeted marketing and service strategies are as follows: for high-value core users, provide exclusive customized services, member-only discounts, and priority resource guarantees; for mid-value potential users, enhance user stickiness and consumption frequency through personalized package recommendations, service experience optimization, and promotional activities; for low-value ordinary users, provide basic service guarantees, low-cost marketing outreach, and explore potential upgrade needs; and at the same time, optimize the resource allocation and layout of charging and swapping equipment based on the spatiotemporal characteristics and behavioral preferences of user profiles.

[0027] Another object of the present invention is to provide a computer device, the computer device including a memory and a processor, the memory storing a computer program, the computer program being executed by the processor causing the processor to perform the steps of the charging and swapping user profile construction method based on multi-dimensional behavioral data.

[0028] Based on the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solution to be protected by this invention are as follows:

[0029] First, comprehensive data dimensions for more accurate characterization: This invention collects multi-dimensional behavioral data of charging and swapping users, including transaction, device, time, and geographical data, and combines it with demand preference survey data. This breaks through the limitations of traditional single-dimensional data and achieves a comprehensive characterization of user behavior and needs, laying a data foundation for building refined profiles.

[0030] The feature system is scientific and allows for deeper mining: It has constructed a behavioral feature system based on charging habits, battery swapping frequency, usage time, and consumption scenarios, as well as a demand preference feature system based on service quality, price sensitivity, and service experience. Through quantification, coding, and fusion, it deeply mines the core features of users, solving the problem of insufficient depth in traditional feature mining.

[0031] The clustering method is scientific and the group characteristics are significant: The user is accurately segmented by using feature fusion combined with an improved clustering algorithm. The optimal number of clusters is determined by the silhouette coefficient method and the elbow rule, which effectively solves the problem of coarse traditional clustering. The clustering results have significant feature differences, providing a scientific basis for profile construction.

[0032] The user profile system is comprehensive and presented more intuitively: a charging and swapping user profile system consisting of a "three-level tag system + group profile + individual user profile" has been constructed. It is presented using visualization methods such as tag cloud and feature radar chart, achieving full coverage of "group commonality + individual personality". The profile structure is clear, the content is comprehensive, and the presentation is intuitive.

[0033] Customized value assessment and more targeted strategies: Based on the traditional RFM model, a customized value assessment model is built in combination with the characteristics of the charging and swapping business to complete the value classification of user groups. Based on the profile characteristics and value levels, a "personalized" marketing and service strategy is formulated. At the same time, resource allocation is optimized, which effectively improves the operational efficiency and commercial value of the charging and swapping business.

[0034] Dynamic profile updates to adapt to business development: The system includes dynamic profile updates and model optimization. It iterates and updates profiles based on real-time data and optimizes model parameters based on business feedback. This ensures that the profiles and models always adapt to the development needs of the charging and swapping business, demonstrating good practicality and scalability. Attached Figure Description

[0035] Figure 1 This is a flowchart of a method for constructing a charging and swapping user profile based on multi-dimensional behavioral data, provided by an embodiment of the present invention.

[0036] Figure 2 This is a framework diagram of the user behavior feature system for charging and swapping provided in an embodiment of the present invention;

[0037] Figure 3 This is a framework diagram of the three-level tagging system for charging and swapping users provided in an embodiment of the present invention;

[0038] Figure 4 This is a framework diagram of the indicator system for the user value assessment model provided in this embodiment of the invention;

[0039] Figure 5 This is a visualization of user group clustering results provided in an embodiment of the present invention;

[0040] Figure 6 This is a feature radar map of a high-value user group provided in an embodiment of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0042] like Figure 1As shown in the figure, this embodiment of the invention provides a method for constructing a charging and swapping user profile based on multi-dimensional behavioral data. The method includes:

[0043] S1. Multi-dimensional data collection and preprocessing: Collect behavioral data and demand preference survey data of charging and swapping users, clean, deduplicate, complete and standardize the raw data to obtain a structured dataset to be analyzed.

[0044] S2. User behavior feature extraction: Based on preprocessed behavior data, a behavior feature system is constructed from four core dimensions: charging habits, battery swapping frequency, usage time period, and consumption scenario, and feature quantification and encoding are completed.

[0045] S3. User demand and preference feature mining: Statistical analysis and correlation mining are performed on demand and preference survey data to extract three core demand and preference features: service quality demand, price sensitivity, and service experience preference, forming a demand and preference feature vector.

[0046] S4. Precise user segmentation: The user behavior feature vector is fused with the demand preference feature vector, and a clustering algorithm is used to perform cluster analysis on the fused feature dataset to obtain multiple charging and swapping user groups with significant feature differences.

[0047] S5. Multi-dimensional user profile construction: For each user group, feature summarization and tagging are carried out from three levels: basic attributes, behavioral characteristics, and demand preferences to build a structured and visualized charging and swapping user group profile, while generating a personalized profile for each user.

[0048] S6. User value assessment and strategy formulation: Establish a user value assessment model for charging and swapping, classify the value of each user group, analyze the commercial contribution and potential development space of different groups, and formulate targeted marketing and service strategies based on user profile characteristics and value levels.

[0049] In step S1, the behavioral data of the charging and swapping users is full-process business data, including user charging and swapping transaction data, device usage data, geographical trajectory data, and time dimension data. The data collection methods include data collection from the charging and swapping platform backend, data collection from the charging and swapping device terminal, and data synchronization from the user terminal APP. The demand preference survey data includes online questionnaire survey data, offline interview data, and service evaluation feedback data. During the data collection process, user identity information is anonymized to ensure data compliance.

[0050] In step S1, the specific process of data preprocessing is as follows: first, extreme abnormal data in the dataset is removed by an outlier detection algorithm, missing values ​​are filled in by interpolation, and duplicate data is deleted;

[0051] Then, the unstructured data is transformed into a structured data set, the numerical features are normalized / standardized, and the categorical features are one-hot encoded / label encoded, ultimately resulting in a dataset with uniform dimensions and standardized format for analysis.

[0052] like Figure 2 As shown, in step S2, the specific dimensions and quantitative indicators of the behavioral feature system are as follows:

[0053] (1) Charging habit dimension: including average charging time, charging power preference, charging amount range, percentage of charging method selection, and percentage of fixed charging points;

[0054] (2) Battery swapping frequency dimension: including monthly average number of battery swaps, battery swapping interval days, battery swapping capacity preference, and battery swapping package purchase indicators;

[0055] (3) Usage time dimension: including peak hour usage ratio, off-peak hour usage ratio, low-peak hour usage ratio, weekend / weekday usage difference rate, and nighttime charging and swapping ratio;

[0056] (4) Consumption scenario dimension: including indicators such as the proportion of household scenarios, commercial scenarios, public scenarios, and the proportion of long-distance travel-related usage;

[0057] Each indicator is assigned a weight based on actual business needs, and a weighted summation method is used to construct the behavioral feature vector.

[0058] In step S3, the service quality demands include preference levels for charging and swapping equipment availability, service response speed, and fault handling efficiency; the price sensitivity includes a quantitative value of sensitivity to charging and swapping pricing, package discounts, and membership discounts, and a Likert scale is used to complete the sensitivity classification; the service experience preferences include preferences for charging and swapping site environment, terminal operation convenience, online platform services, and value-added services, and the correlation between each experience factor and user behavior is obtained through association rule mining; finally, the three types of features are quantified and concatenated into a demand preference feature vector, which is matched with the behavioral feature vector dimension.

[0059] In step S4, the feature fusion adopts the feature concatenation method, which concatenates the user behavior feature vector and the demand preference feature vector in dimensional order into a unified comprehensive user feature vector; the clustering algorithm is an improved K-means clustering algorithm or DBSCAN density clustering algorithm. Before clustering, the optimal number of clusters is determined by the silhouette coefficient method and elbow rule. During the clustering process, the feature vector is dimensionality reduced to improve the clustering efficiency and accuracy.

[0060] like Figure 3As shown, in step S5, the tagging process adopts a three-level tagging system. The first-level tags include basic attribute tags, behavioral feature tags, and demand preference tags; the second-level tags are the subdivisions of the first-level tags; and the third-level tags are specific feature quantification values ​​or descriptive tags. The user group profile is visualized in the form of tag cloud, feature radar chart, and data report. The personalized profile of a single user is generated by supplementing the profile of the user's group and its own unique characteristics, so as to achieve full coverage of the profile of "group commonality + individual personality".

[0061] like Figure 4 As shown, in step S6, the charging and swapping user value assessment model is constructed using the RFM model combined with business-customized indicators. Based on the traditional R (most recent consumption), F (consumption frequency), and M (consumption amount) indicators, it adds charging and swapping business-specific indicators: user lifecycle, platform stickiness, potential consumption capacity, and word-of-mouth value. The weight of each indicator is determined using the analytic hierarchy process, and the comprehensive value score of each user group is calculated using the weighted scoring method, dividing the user group into three levels: high-value core users, medium-value potential users, and low-value ordinary users.

[0062] In step S6, the principles for formulating targeted marketing and service strategies are as follows: for high-value core users, provide exclusive customized services, member-only discounts, and priority resource guarantees; for mid-value potential users, enhance user stickiness and consumption frequency through personalized package recommendations, service experience optimization, and promotional activities; for low-value ordinary users, provide basic service guarantees, low-cost marketing outreach, and explore potential upgrade needs; and at the same time, optimize the resource allocation and layout of charging and swapping equipment based on the spatiotemporal characteristics and behavioral preferences of user profiles.

[0063] Figure 5 This is a visualization of user group clustering results provided in an embodiment of the present invention;

[0064] Figure 6 This is a feature radar map of a high-value user group provided in an embodiment of the present invention.

[0065] The specific application areas or related products of this invention.

[0066] This invention provides a computer device, which includes a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the charging and swapping user profile construction method based on multi-dimensional behavioral data.

[0067] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.

[0068] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for constructing user profiles for charging and battery swapping based on multi-dimensional behavioral data, characterized in that, The method includes: S1. Multi-dimensional data collection and preprocessing: Collect behavioral data and demand preference survey data of charging and swapping users, clean, deduplicate, complete and standardize the raw data to obtain a structured dataset to be analyzed. S2. User behavior feature extraction: Based on preprocessed behavior data, a behavior feature system is constructed from four core dimensions: charging habits, battery swapping frequency, usage time period, and consumption scenario, and feature quantification and encoding are completed. S3. User demand and preference feature mining: Statistical analysis and correlation mining are performed on demand and preference survey data to extract three core demand and preference features: service quality demand, price sensitivity, and service experience preference, forming a demand and preference feature vector. S4. Precise user segmentation: The user behavior feature vector is fused with the demand preference feature vector, and a clustering algorithm is used to perform cluster analysis on the fused feature dataset to obtain multiple charging and swapping user groups with significant feature differences. S5. Multi-dimensional user profile construction: For each user group, feature summarization and tagging are carried out from three levels: basic attributes, behavioral characteristics, and demand preferences to build a structured and visualized charging and swapping user group profile, while generating a personalized profile for each user. S6. User value assessment and strategy formulation: Establish a user value assessment model for charging and swapping, classify the value of each user group, analyze the commercial contribution and potential development space of different groups, and formulate targeted marketing and service strategies based on user profile characteristics and value levels.

2. The method for constructing charging and swapping user profiles based on multi-dimensional behavioral data according to claim 1, characterized in that, In step S1, the behavioral data of the charging and swapping users is full-process business data, including user charging and swapping transaction data, device usage data, geographic trajectory data, and time dimension data. The data collection methods include data collection from the charging and swapping platform backend embedded points, data collection from the charging and swapping equipment terminal, and data synchronization from the user terminal APP. The demand preference survey data includes online questionnaire survey data, offline interview data, and service evaluation feedback data. User identity information is anonymized during the data collection process to ensure data compliance.

3. The method for constructing charging and swapping user profiles based on multi-dimensional behavioral data according to claim 1, characterized in that, In step S1, the specific process of data preprocessing is as follows: first, extreme abnormal data in the dataset is removed by an outlier detection algorithm, missing values ​​are filled in by interpolation, and duplicate data is deleted; Then, the unstructured data is transformed into a structured data set, the numerical features are normalized / standardized, and the categorical features are one-hot encoded / label encoded, ultimately resulting in a dataset with uniform dimensions and standardized format for analysis.

4. The method for constructing charging and swapping user profiles based on multi-dimensional behavioral data according to claim 1, characterized in that, In step S2, the specific dimensions and quantitative indicators of the behavioral feature system are as follows: (1) Charging habit dimension: including average charging time, charging power preference, charging amount range, percentage of charging method selection, and percentage of fixed charging points; (2) Battery swapping frequency dimension: including monthly average number of battery swaps, battery swapping interval days, battery swapping capacity preference, and battery swapping package purchase indicators; (3) Usage time dimension: including peak hour usage ratio, off-peak hour usage ratio, low-peak hour usage ratio, weekend / weekday usage difference rate, and nighttime charging and swapping ratio; (4) Consumption scenario dimension: including indicators such as the proportion of household scenarios, commercial scenarios, public scenarios, and the proportion of long-distance travel-related usage; Each indicator is assigned a weight based on actual business needs, and a weighted summation method is used to construct the behavioral feature vector.

5. The method for constructing charging and swapping user profiles based on multi-dimensional behavioral data according to claim 1, characterized in that, In step S3, the service quality demands include preference levels for charging and swapping equipment availability, service response speed, and fault handling efficiency; the price sensitivity includes a quantitative value of sensitivity to charging and swapping pricing, package discounts, and membership discounts, and a Likert scale is used to complete the sensitivity classification; the service experience preferences include preferences for charging and swapping site environment, terminal operation convenience, online platform services, and value-added services, and the correlation between each experience factor and user behavior is obtained through association rule mining; finally, the three types of features are quantified and concatenated into a demand preference feature vector, which is matched with the behavioral feature vector dimension.

6. The method for constructing charging and swapping user profiles based on multi-dimensional behavioral data according to claim 1, characterized in that, In step S4, the feature fusion adopts the feature concatenation method, which concatenates the user behavior feature vector and the demand preference feature vector in dimensional order into a unified comprehensive user feature vector; the clustering algorithm is an improved K-means clustering algorithm or DBSCAN density clustering algorithm. Before clustering, the optimal number of clusters is determined by the silhouette coefficient method and elbow rule. During the clustering process, the feature vector is dimensionality reduced to improve the clustering efficiency and accuracy.

7. The method for constructing charging and swapping user profiles based on multi-dimensional behavioral data according to claim 1, characterized in that, In step S5, the tagging process adopts a three-level tagging system. The first-level tags include basic attribute tags, behavioral feature tags, and demand preference tags. The second-level tags are the subdivisions of the first-level tags, and the third-level tags are specific feature quantification values ​​or descriptive tags. The user group profile is visualized in the form of tag cloud, feature radar chart, and data report. The personalized profile of a single user is generated based on the profile of the user's group and its own unique characteristics, so as to achieve full coverage of the profile of "group commonality + individual individuality".

8. The method for constructing charging and swapping user profiles based on multi-dimensional behavioral data according to claim 1, characterized in that, In step S6, the charging and swapping user value assessment model is constructed using the RFM model combined with business-customized indicators. Based on the traditional R (recent consumption), F (consumption frequency), and M (consumption amount) indicators, it adds charging and swapping business-specific indicators: user lifecycle, platform stickiness, potential consumption capacity, and word-of-mouth value. The weight of each indicator is determined using the analytic hierarchy process, and the comprehensive value score of each user group is calculated using a weighted scoring method, dividing the user group into three levels: high-value core users, medium-value potential users, and low-value ordinary users.

9. The method for constructing charging and swapping user profiles based on multi-dimensional behavioral data according to claim 8, characterized in that, In step S6, the principle for formulating the targeted marketing and service strategy is: to provide exclusive customized services, exclusive member discounts, and priority resource guarantees for high-value core users; For mid-value potential users, we enhance user stickiness and purchase frequency through personalized package recommendations, service experience optimization, and promotional activities; for low-value ordinary users, we provide basic service guarantees, low-cost marketing outreach, and explore potential upgrade needs; at the same time, we optimize the resource allocation and layout of charging and swapping equipment based on the spatiotemporal characteristics and behavioral preferences of user profiles.

10. A computer device, characterized in that, The computer device includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor performs the steps of the charging and swapping user profile construction method based on multi-dimensional behavioral data as described in any one of claims 1-9.

Images