Power value-added service target customer portraying method, system, equipment and medium

Abstract

The invention discloses an electric power value-added service target customer portraying method, system and device and a medium, and belongs to the technical field of customer behavior data analysis, and the method comprises the following steps: obtaining the electricity utilization information of an electricity utilization customer through an electric power information system; the method comprises the following steps: dividing power utilization information, outputting a division result as power utilization behavior characteristics, and sorting the divided power utilization behavior characteristics; feature tags are extracted from the sorted electricity consumption behavior features, and an electricity consumption customer portrait is constructed; and analyzing the satisfaction value of the electricity customer based on the portrait of the electricity customer, calculating the risk value of the electricity customer, and obtaining a target customer by combining the satisfaction value of the electricity customer and the risk value of the electricity customer. According to the invention, through comprehensive analysis of the electricity customer data, the portrait of the electricity customer can be more accurately constructed, the target and success rate of service promotion are improved, and through deep analysis of the satisfaction of the electricity customer and risk assessment, the customer relationship can be more effectively managed, and the marketing risk is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of customer behavior data analysis, and in particular to a method, system, device and medium for profiling target customers of power value-added services. Background Art

[0002] With the continuous development of the electricity market, the variety and scope of value-added electricity services are becoming increasingly diverse. Furthermore, with the advancement of data science and artificial intelligence technologies, targeted customer acquisition has become a key strategy in modern marketing. However, traditional customer analysis focuses solely on basic electricity usage information, such as electricity consumption and time of day. Currently, power companies lack a deep understanding of their customers when developing value-added services and lack accurate customer analysis tools. In particular, analysis of customers' electricity usage behaviors, needs, and preferences has become crucial. The lack of in-depth customer profiling and behavioral analysis leads to inaccurate marketing strategies and inefficient resource utilization. Furthermore, analysis of customer satisfaction and complaints is often fragmented, lacking a systematic customer risk assessment system. Consequently, existing technologies have significant shortcomings in the comprehensive utilization of customer data, in-depth analysis of customer behavior, and risk assessment. Summary of the Invention

[0003] In view of the above-mentioned problems, the present invention is proposed.

[0004] Therefore, the technical problem solved by the present invention is that the existing technology has significant deficiencies in the comprehensive utilization of customer data, in-depth analysis of electricity customer behavior and risk assessment, and the traditional customer acquisition method has deficiencies in in-depth analysis and risk assessment of electricity customers.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for profiling target customers of power value-added services, which comprises the following steps:

[0006] Obtain electricity consumption information of electricity users through the power information system;

[0007] By dividing the electricity usage information, outputting the division results as electricity usage behavior features, and sorting the divided electricity usage behavior features;

[0008] Extract feature labels from the sorted electricity usage behavior features and build a profile of electricity users;

[0009] Based on the electricity user portrait, the electricity user customer satisfaction value is analyzed and the electricity user risk value is calculated. The target customers are obtained by combining the electricity user customer satisfaction value and the electricity user risk value.

[0010] As a preferred solution of the method for profiling target customers of power value-added services described in the present invention, when obtaining the electricity consumption information of electricity users, the following steps are included:

[0011] The maximum time period for obtaining electricity consumption information; a preset collection factor, multiplying the maximum time period by the collection factor to obtain the collection period for electricity users; and collecting electricity consumption information for electricity users within the collection period.

[0012] As a preferred solution of the method for profiling target customers of value-added electricity services described in the present invention, the step of dividing electricity usage information includes analyzing electricity usage behavior, electricity usage customer needs and electricity usage preferences through electricity usage information of electricity users; integrating the analyzed electricity usage behavior, electricity usage customer needs and electricity usage preferences, and outputting them as electricity usage behavior characteristics.

[0013] As a preferred solution of the method for profiling target customers of power value-added services described in the present invention, the step of extracting feature labels from power consumption behavior features includes:

[0014] Through natural language processing technology, electricity consumption behavior characteristics are analyzed, and service sensitivity characteristics and response time requirement characteristics are extracted from electricity consumption behavior characteristics; the extracted results are integrated with equipment operation log data to generate various feature types; the feature importance evaluation mechanism is used to reduce the dimension of equipment health characteristics, select various feature types with contributions higher than the preset threshold, and output the selected feature types as feature labels.

[0015] The beneficial effects of this preferred technical solution are: through the fusion of natural language processing and device data, it can accurately capture the characteristics of customer service response needs, combine with the feature importance screening mechanism, effectively eliminate data redundancy, and improve the interpretability and business adaptability of the labeling system.

[0016] As a preferred solution of the method for profiling target customers of power value-added services described in the present invention, the step of analyzing the satisfaction value of electricity users includes:

[0017] Construct a satisfaction value evaluation index system; establish a dynamic weight allocation mechanism to adjust the weight ratio according to the feature label; calculate the comprehensive satisfaction index based on the adjusted weight ratio and output it as the electricity customer satisfaction value.

[0018] The beneficial effects of this preferred technical solution are: the dynamic weight allocation mechanism overcomes the rigidity of traditional evaluation methods, enables satisfaction value evaluation to adapt to the characteristic differences of different customer groups, and the comprehensive index generation enhances the multi-dimensional characterization of customer experience.

[0019] As a preferred solution of the method for profiling target customers of power value-added services described in the present invention, the step of calculating the risk value of electricity users includes:

[0020] A risk assessment matrix is ​​constructed based on electricity consumption information of electricity users; a time decay function is used to process risk events in the risk assessment matrix; and customers are divided into low-risk customers, medium-risk customers, and high-risk customers through a risk stratification algorithm.

[0021] The beneficial effects of this preferred technical solution are: time decay function processing enhances the time sensitivity of risk assessment, and the risk stratification algorithm realizes the gradient management of customer risks, providing quantitative support for the formulation of differentiated service strategies.

[0022] As a preferred solution of the method for profiling target customers of power value-added services described in the present invention, the step of obtaining target customers includes:

[0023] Target customers are acquired based on electricity user profiles, electricity user satisfaction values, and electricity user risk values; the target customers do not include high-risk customers.

[0024] Another object of the present invention is to provide a system for profiling target customers of power value-added services.

[0025] To solve the above technical problems, the present invention provides the following technical solutions: a system for profiling target customers of power value-added services, comprising a data acquisition module, a power consumption behavior feature classification module, a label extraction module, and an analysis and evaluation module;

[0026] The data acquisition module collects electricity consumption information of electricity users through the power information system;

[0027] The electricity usage behavior feature classification module divides the electricity usage information, outputs the classification results as electricity usage behavior features, and sorts the divided electricity usage behavior features;

[0028] The label extraction module sorts the divided electricity consumption behavior features, extracts feature labels, and constructs a profile of the electricity user;

[0029] The analysis and evaluation module analyzes the satisfaction value of electricity users based on the electricity user portrait, calculates the risk value of electricity users, and determines the target customers.

[0030] The present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the method for profiling target customers of value-added power services are implemented.

[0031] The present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the method for creating a target customer profile for a value-added power service are implemented.

[0032] The beneficial effects of this invention are: significantly improving the accuracy and efficiency of customer acquisition. By comprehensively analyzing electricity customer data, a more accurate customer profile can be constructed, improving the targeting and success rate of service promotion. Furthermore, through in-depth analysis of customer satisfaction and risk assessment, customer relationships can be more effectively managed and marketing risks reduced. Furthermore, the application of this method helps to improve customer satisfaction and loyalty, thereby driving business growth and market competitiveness. This includes customer data utilization, electricity customer behavior analysis, and risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0034] Figure 1 An overall flow chart of a method for profiling target customers for value-added power services provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0035] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0036] Example 1, with reference to Figure 1 , which is the first embodiment of the present invention, provides a method for profiling target customers of power value-added services, comprising the following steps:

[0037] S1. Obtain electricity consumption information of electricity users through the power information system.

[0038] Obtaining electricity consumption information of electricity users includes the following steps:

[0039] The maximum time period for obtaining electricity consumption information; a preset collection factor, multiplying the maximum time period by the collection factor to obtain the collection period for electricity users; and collecting electricity consumption information for electricity users within the collection period.

[0040] In an optional embodiment, a provincial power information platform database is accessed to obtain customer electricity data storage configuration parameters of a certain industrial park, and it is determined that the maximum historical data storage period supported by the system is 8 years.

[0041] When calculating the effective collection period, we set the collection factor α to 0.6 and calculated the effective collection period: T = 8 years × 0.6 = 4.8 years ≈ 5 years. Therefore, the construction time window is 5 years forward from January 1, 2025 (January 1, 2019 - December 31, 2024).

[0042] In another optional embodiment, electricity consumption information can be directly collected over the past three years. The user's electricity usage behavior is then analyzed to calculate electricity consumption and cost scores. Customer needs are analyzed based on their preferences and activity frequency. Customer preferences are analyzed based on their preferred selections and selection frequency. These three analysis results provide foundational data for subsequent steps.

[0043] S2. Divide the electricity usage information, output the division results as electricity usage behavior features, and sort the divided electricity usage behavior features.

[0044] The steps for dividing electricity usage information include:

[0045] Analyze electricity consumption behavior, customer needs and preferences through electricity consumption information of electricity users;

[0046] The analyzed electricity usage behavior, electricity customer needs and electricity usage preferences are integrated and output as electricity usage behavior characteristics.

[0047] In an optional embodiment, when analyzing the electricity consumption information of the electricity user, the electricity consumption behavior E of the electricity user is analyzed based on the electricity consumption and electricity cost of the electricity user. usage :

[0048]

[0049] Where n represents the number of power consumption measurement periods, α i represents the weighted coefficient of the i-th period, P i represents the power consumption in the i-th period, p i represents the power coefficient of electricity consumption in the i-th period, m represents the number of electricity cost measurement periods, β j represents the weighted coefficient of the jth period, Q j represents the electricity cost in the jth period, δ j Indicates the offset of electricity cost in the jth period.

[0050] Assume a business user, n = m = 24, analyze by hour, there are 24 time periods in a day, 9-11 am (assuming it is the 9th to 11th period), which is the business peak, the weighted coefficients α9, α 10 and α 11Both are 0.08. 14-16 o'clock in the afternoon (the 14th-16th period) is also a busy business period, so the output α 14 , α 15 and α 16 The power consumption in the 9th period is P9=50 degrees, assuming P9=1.2; the power consumption in the 10th period is P 10 =60 degrees, P 10 = 1.2. The contributions of these two periods to electricity consumption are 10.07 and 12.98, respectively. By calculating the lowest electricity cost, the contribution value for each period is output. By adding the electricity consumption and electricity cost contributions for all periods, we can obtain the total electricity consumption for this commercial user for the day.

[0051] In another optional embodiment, by analyzing the electricity consumption information of electricity users, it was found that among 100 industrial and commercial users in a certain park, 30 had regular load troughs between 12:00 and 2:00 PM due to centralized shutdowns for maintenance; 25 experienced peak electricity consumption between 6:00 PM and 10:00 PM when production equipment was running at full capacity; and 45 had periodic characteristics of weekend electricity consumption dropping by more than 60%. When analyzing the needs of electricity users, customer surveys revealed that 60% of users were concerned about tiered electricity pricing policies, 35% required a power outage response time of less than 10 minutes, and 5% requested customized load control solutions. When analyzing the preferences of electricity users, 20 high-tech enterprises preferred a renewable energy power supply ratio of ≥30%, 30 traditional manufacturing industries preferred fixed capacity packages, and 50 small and micro enterprises chose flexible billing models.

[0052] After the analysis is completed, the output is the electricity consumption behavior characteristics, which include load periodicity characteristics, response demand characteristics, and energy preference characteristics.

[0053] In an optional embodiment, the needs of electricity customers may also be analyzed based on their preference characteristics and activity frequencies:

[0054]

[0055] Where o represents the number of electricity customer preference characteristics, γ k represents the weighting coefficient of the kth feature, R k represents the preference characteristics of the k-th electricity customer, represents the power coefficient of the kth feature, p represents the activity frequency of electricity customers, and λ l represents the weighting coefficient of the lth activity frequency, S l Indicates the activity frequency of the lth electricity customer.

[0056] In another optional embodiment, the electricity customer preferences can be analyzed based on the electricity customer preference selection items and the electricity customer selection frequency:

[0057]

[0058] Among them, r represents the number of preferred options for electricity customers, μ q represents the weighted coefficient of the qth electricity customer’s preference option, T q represents the qth electricity customer’s preference option, v q The weighted coefficient of the qth electricity customer’s preferred frequency selection, U q Represents the selection frequency of the qth electricity customer.

[0059] S3. Extract feature labels from the sorted electricity usage behavior features and build a profile of electricity users.

[0060] The steps of extracting feature labels for electricity usage behavior features include: parsing electricity usage behavior features through natural language processing technology, extracting service sensitivity features and response time requirement features from electricity usage behavior features; integrating the extracted results with equipment operation log data to generate various feature types; using a feature importance evaluation mechanism to perform dimensionality reduction processing on equipment health features, selecting various feature types whose contribution is higher than a preset threshold, and outputting the selected feature types as feature labels.

[0061] In one optional embodiment, electricity usage characteristics are first analyzed. For example, an industrial user's electricity usage characteristics may be described as "extremely sensitive to electricity price fluctuations, requiring a response to power outage and maintenance notifications within 15 minutes, otherwise production line operations will be impacted." Natural language processing technology is used to analyze and extract the service sensitivity characteristics (sensitivity to electricity price fluctuations) and the response time requirement (responding to power outage notifications within 15 minutes).

[0062] When integrating the user device operation log, for example, "In the past 30 days, key equipment has shut down four times due to unstable voltage, with an average recovery time of 20 minutes each time," the following feature types are generated: service sensitivity feature, response time requirement feature, and device health feature.

[0063] Using a feature importance assessment mechanism with a preset threshold of 0.5, the calculations show that: sensitivity to electricity price fluctuations has a contribution of 0.7; responding to power outage notifications within 15 minutes has a contribution of 0.65; frequency of outages due to voltage instability has a contribution of 0.45; and outage recovery time has a contribution of 0.4. "Sensitivity to electricity price fluctuations" and "responding to power outage notifications within 15 minutes" with contributions above the threshold are selected and output as feature labels.

[0064] Based on feature tags, the industrial user profile is sensitive to electricity price fluctuations and requires power supply companies to respond quickly to power outage notifications. It is a time-sensitive and price-sensitive industrial customer, and priority must be given to ensuring power supply stability and providing real-time electricity price warning services.

[0065] In this embodiment, the feature importance evaluation mechanism includes but is not limited to a random forest algorithm or a neural network algorithm.

[0066] In another optional embodiment, based on the analysis results of the electricity consumption behavior, electricity consumption needs and preferences of the electricity consumption customers, a feature label F is extracted for the electricity consumption customers. feat ·:

[0067]

[0068] Among them, t represents the number of electricity customers’ characteristics, ξ s Represents the weight coefficient of the sth feature, π s Represents the power coefficient of the sth feature, E usage Represents the quantitative value of electricity consumption behavior, N demand Represents the quantitative value of electricity customer demand, P pref It represents the quantitative value of electricity customer preference.

[0069] Build electricity customer portrait U based on electricity customer feature labels profile :

[0070]

[0071] Among them, v represents the number of electricity customer portrait features, w u represents the weight coefficient of the u-th feature, φ u represents the power coefficient of the u-th feature, ψ u represents the weight coefficient of the u-th feature, θ u Represents the weight coefficient of the u-th feature.

[0072] It should be noted that the combination of logarithmic and summation formulas can more accurately characterize the needs and preferences of electricity users.

[0073] S4. Analyze the customer satisfaction value based on the customer portrait, calculate the customer risk value, and obtain target customers by combining the customer satisfaction value and the customer risk value.

[0074] In an optional embodiment, when analyzing the customer satisfaction value of electricity users, a satisfaction value evaluation index system is constructed; a dynamic weight allocation mechanism is established to adjust the weight ratio according to the feature label; and a comprehensive satisfaction index is calculated based on the adjusted weight ratio, and the output is the customer satisfaction value of electricity users. Specifically, it is expressed as:

[0075] First, we developed a satisfaction evaluation index system. Taking residential electricity users in a specific region as an example, we constructed a system consisting of four primary indicators: power supply reliability, electricity price rationality, service response speed, and timely fault repair rate. Each indicator was scored on a 100-point scale, and scores were obtained through questionnaire surveys and electricity consumption data statistics.

[0076] Assume that preliminary analysis reveals that a certain type of residential customer in the area is characterized by being "sensitive to electricity prices and prioritizing power supply reliability." The initial weights are (power supply reliability: 0.3, electricity price rationality: 0.3, service response speed: 0.2, and fault repair timeliness: 0.2). These weights are dynamically adjusted based on the characteristic labels to (power supply reliability: 0.4, electricity price rationality: 0.4, service response speed: 0.1, and fault repair timeliness: 0.1).

[0077] Then calculate the comprehensive satisfaction index and the scores of various indicators of a resident customer.

[0078] In another optional embodiment, the electricity customer satisfaction, demands and complaint information are analyzed.

[0079] Calculate the ratio of positive and negative sentiment scores of electricity users and analyze the customer satisfaction S sat :

[0080]

[0081] Where x represents the number of satisfaction feedback, χ w Represents the sentiment analysis weight coefficient, Pos w represents the positive sentiment score of the w-th feedback, Neg w represents the negative sentiment score of the w-th feedback, ζ w represents the power coefficient of sentiment analysis, z represents the number of negative feedback, η y represents the negative feedback weight coefficient, Represents the negative feedback weight coefficient, Neg y represents the negative sentiment score of the y-th feedback.

[0082] Analyze the demands of electricity users based on their feedback, complaints, product selection frequency, service satisfaction scores, and problem resolution time. demand :

[0083]

[0084] Where b represents the number of electricity customers’ demands, λ a Req represents the weighted coefficient of electricity customer demands, a Indicates the appeal score of the a-th feedback, Com a represents the complaint score of the a-th feedback, V a represents the number of product selections in the a-th feedback, W a represents the service satisfaction score in the a-th feedback, μ a represents the power coefficient of electricity customer demand, ξ a represents the weighted coefficient of electricity customer demands, X a Indicates the time it takes to resolve the a-th feedback issue.

[0085] The steps for calculating the risk value of electricity users include: constructing a risk assessment matrix based on the electricity usage information of electricity users; processing risk events of the risk assessment matrix using a time decay function; and classifying customers into low-risk customers, medium-risk customers, and high-risk customers through a risk stratification algorithm.

[0086] The electricity customer risk score is calculated based on the electricity customer's credit score, the number of electricity customer defaults, the number of years of cooperation with the electricity customer, the number of electricity customer overdue payments, the number of electricity customer complaints, and the electricity customer return rate.

[0087] The risk score of electricity customers is expressed as:

[0088]

[0089] Among them, e represents the number of electricity customers, α f Indicates that p f Represents the power coefficient of credit score, C f represents the credit score of the f-th electricity customer, β represents the weighted coefficient of the number of defaults of the electricity customer, D represents the number of defaults of the electricity customer, γ represents the weighted coefficient of the number of years of cooperation with the electricity customer, E represents the number of years of cooperation with the electricity customer, δ represents the weighted coefficient of the number of overdue payments of the electricity customer, F represents the number of overdue payments of the electricity customer, ∈ represents the weighted coefficient of the number of complaints of the electricity customer, G represents the number of complaints of the electricity customer, ζ represents the weighted coefficient of the return rate of the electricity customer, and H represents the return rate of the electricity customer.

[0090] Analyze electricity customer demands based on customer feedback, complaints, number of product problems, problem severity scores, and customer churn risk. complaint :

[0091]

[0092] Where d represents the number of complaints from electricity customers, v c Com represents the weighted coefficient of electricity customer complaints. c Req represents the complaint score of the cth feedback, c represents the appeal score of the cth feedback, Y c represents the number of recurring problems in the c-th feedback, Z c represents the severity score of the c-th feedback problem, ξ c represents the power coefficient of electricity customer complaints, η c represents the weighted coefficient of electricity customer complaints, U c represents the customer churn risk score of the c-th feedback.

[0093] It should be noted that appeal analysis and complaint information analysis further enhance the accuracy of customer feedback.

[0094] The steps to acquire target customers include,

[0095] Acquire target customers based on electricity customer profiles, electricity customer satisfaction scores, and electricity customer risk scores;

[0096] Target customers do not include high-risk customers.

[0097] Based on the electricity customer profile, electricity customer satisfaction, demands, complaint information and electricity customer risk score, potential demand target users are identified as follows:

[0098]

[0099] Among them, h represents the number of potential target users, η s represents the weighted coefficient of matching feature s, θ s Represents the power coefficient of the matching feature s.

[0100] It should be noted that the above formula comprehensively considers electricity user profiles, satisfaction, demands, complaints, and risk scores, providing a comprehensive assessment for intelligent target customer matching. This ensures that target customer selection not only meets electricity user needs and preferences, but also fully considers their risk factors, providing safety and security for users.

[0101] Example 2 is the second embodiment of the present invention, which is different from the previous embodiment in that:

[0102] If the function 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, or the part that contributes to the prior art, or the 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 and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0103] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0104] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0105] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0106] Example 3 is the third embodiment of the present invention, which provides a system for profiling target customers of power value-added services, including a data acquisition module, a power consumption behavior feature classification module, a label extraction module, and an analysis and evaluation module;

[0107] The data acquisition module collects electricity consumption information of electricity users through the power information system;

[0108] The electricity usage behavior feature segmentation module segments the electricity usage information, outputs the segmentation results as electricity usage behavior features, and sorts the segmented electricity usage behavior features;

[0109] The label extraction module sorts the divided electricity consumption behavior features, extracts feature labels, and constructs a profile of the electricity user;

[0110] The analysis and evaluation module analyzes the customer satisfaction value based on the customer portrait, calculates the customer risk value, and determines the target customers.

[0111] Example 4 is the fourth embodiment of the present invention, which provides a method for creating a profile of potential customers with demand for value-added power services. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0112] During the experiment preparation phase, we collected electricity consumption data, historical service records, satisfaction feedback, and other relevant data from electricity users to conduct the experiment:

[0113] This method is used to analyze electricity customer data and construct electricity customer portraits.

[0114] Screen potential target user groups through electricity customer satisfaction and risk assessment.

[0115] Compare the customer acquisition effects of the traditional method and the method of the present invention.

[0116] The experimental results are shown in Table 1.

[0117] Table 1: Experimental results of different methods

[0118]

[0119] It can be seen that this method is superior to existing traditional methods in terms of customer acquisition accuracy, electricity customer satisfaction, and risk assessment accuracy.

[0120] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for profiling target customers for power value-added services, characterized by: The following steps are included: Obtain electricity consumption information of electricity users through the power information system; By dividing the electricity usage information, outputting the division results as electricity usage behavior features, and sorting the divided electricity usage behavior features; Extract feature labels from the sorted electricity usage behavior features and build a profile of electricity users; Based on the electricity user portrait, the electricity user customer satisfaction value is analyzed and the electricity user risk value is calculated. The target customers are obtained by combining the electricity user customer satisfaction value and the electricity user risk value.

2. The method for profiling target customers for power value-added services according to claim 1, characterized in that: When obtaining electricity consumption information of electricity users, the following steps are included: The maximum time period for obtaining electricity consumption information; Preset the collection factor, multiply the maximum time period by the collection factor to obtain the collection period of the electricity customer; Collect electricity consumption information of electricity customers within the collection period.

3. The method for profiling target customers for power value-added services according to claim 2, characterized in that: The steps for dividing electricity usage information include: Analyze electricity consumption behavior, customer needs and preferences through electricity consumption information of electricity users; The analyzed electricity usage behavior, electricity customer needs and electricity usage preferences are integrated and output as electricity usage behavior characteristics.

4. The method for profiling target customers for power value-added services according to claim 3, characterized in that: The steps of extracting feature labels from electricity consumption behavior features include: Analyze electricity consumption behavior characteristics through natural language processing technology, and extract service sensitivity characteristics and response time requirement characteristics from electricity consumption behavior characteristics; The extracted results are integrated with the equipment operation log data to generate various feature types; The feature importance evaluation mechanism is used to reduce the dimension of the equipment health features, select the feature types whose contribution is higher than the preset threshold, and output the selected feature types as feature labels.

5. The method for profiling target customers for power value-added services according to claim 4, characterized in that: The steps to analyze the customer satisfaction value of electricity users include: Construct a satisfaction evaluation index system; Establish a dynamic weight allocation mechanism to adjust the weight ratio according to feature labels; The comprehensive satisfaction index is calculated through the adjusted weight ratio, and the output is the electricity customer satisfaction value.

6. The method for profiling target customers for power value-added services according to claim 5, characterized in that: The step of calculating the risk value of electricity users includes: Construct a risk assessment matrix based on electricity consumption information of electricity users; Use time decay function to process risk events in risk assessment matrix; Customers are divided into low-risk customers, medium-risk customers, and high-risk customers through risk stratification algorithms.

7. The method for profiling target customers for power value-added services according to claim 6, characterized in that: The steps to acquire target customers include, Acquire target customers based on electricity customer profiles, electricity customer satisfaction scores, and electricity customer risk scores; The target customers mentioned do not include high-risk customers.

8. A system for profiling target customers of power value-added services, applying a method for profiling target customers of power value-added services according to any one of claims 1 to 7, characterized in that: It includes data acquisition module, electricity consumption behavior feature classification module, label extraction module, and analysis and evaluation module; The data acquisition module collects electricity consumption information of electricity users through the power information system; The electricity usage behavior feature classification module divides the electricity usage information, outputs the classification results as electricity usage behavior features, and sorts the divided electricity usage behavior features; The label extraction module sorts the divided electricity consumption behavior features, extracts feature labels, and constructs a profile of the electricity user; The analysis and evaluation module analyzes the satisfaction value of electricity users based on the electricity user portrait, calculates the risk value of electricity users, and determines the target customers.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a method for profiling target customers of power value-added services according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for profiling target customers of power value-added services according to any one of claims 1 to 7 are implemented.

Images