A power supply service evaluation method and system based on dynamic weight adjustment

Abstract

The present invention belongs to the field of power supply service evaluation, and provides a power supply service evaluation method and system based on dynamic weight adjustment, comprising collecting users' scores on power supply services, and storing the scores in association with user information; dividing the scores into preset time periods by time, and calculating the average scores of all users in each time period; determining an environmental impact factor according to the power supply status in the preset time period; determining the dynamic weight of the user's current time period according to the average score, the environmental impact factor and the user's basic weight; calculating a weighted score for each user according to the user's dynamic weight and its score; calculating a total score of the power supply service according to the weighted score of each user; and updating the basic weight for each user according to the dynamic weight after combining each time period.

Description

Technical Field

[0001] The present invention belongs to the field of power supply service evaluation, and in particular relates to a power supply service evaluation method and system based on dynamic weight adjustment. Background Art

[0002] With the development of social economy and the acceleration of urbanization, power supply services, as an important part of public infrastructure, have a service quality that is directly related to user satisfaction and the social image of power companies. In existing technologies, power supply service units generally evaluate service quality through methods such as collecting user feedback information, user satisfaction questionnaires, and regular evaluation reports. However, existing power supply service evaluation methods generally have the following technical problems:

[0003] First, user ratings are not weighted, and all user ratings are generally considered equivalent, having the same impact on the final evaluation results. This fails to effectively reflect the authenticity and credibility of user ratings. For example, some users may give extremely high or low scores due to personal emotions, non-service factors, or malicious cheating, resulting in distorted evaluation results.

[0004] Second, existing methods lack a dynamic management mechanism for rating behavior. Users' ratings vary significantly over time, especially when power supply services are impacted by environmental factors (such as power outages, natural disasters, and equipment failures). User ratings fluctuate significantly, and existing systems struggle to effectively mitigate external environmental influences on rating results, leading to unfair ratings for service personnel or organizations under uncontrollable conditions.

[0005] In addition, existing technologies fail to achieve dynamic adjustment and adaptive updating of user rating influence, and are unable to optimize the weight of users based on their long-term rating behavior, resulting in ineffective utilization of historical rating behavior and a lack of in-depth exploration and management of rating trends.

[0006] Furthermore, current power supply service evaluation methods fail to dynamically adjust evaluation weights based on actual business conditions, such as seasonal variations, peak and off-peak business hours, number of complaints, business volume, and other key indicators. For example, during the off-season, users expect higher service quality, but the existing system fails to increase the weight of service quality in the evaluation, making it difficult to accurately reflect actual service performance.

[0007] In summary, the existing power supply service evaluation technology has problems such as distorted scoring results, lack of scoring weight management, poor anti-interference ability and lack of adaptive adjustment ability. There is an urgent need for a power supply service evaluation method based on dynamic weight adjustment to achieve scientific management of scoring behavior, effective buffering of the external environment and dynamic optimization of user weights, thereby improving the objectivity, fairness and reliability of the evaluation results and enhancing the technical level of power supply service quality management. Summary of the Invention

[0008] In order to solve the problems in the prior art, the present invention provides a power supply service evaluation method based on dynamic weight adjustment, comprising the following steps:

[0009] Collecting user ratings of power supply services and storing the ratings in association with user information;

[0010] Divide the ratings into preset time periods and calculate the average scores of all users in each time period;

[0011] Determine the environmental impact factor based on the power supply status within a preset time period;

[0012] Determine the dynamic weight of the user in the current time period based on the average score, the environmental impact factor, and the user's basic weight;

[0013] Calculate each user's weighted score based on the user's dynamic weight and their score;

[0014] Calculate the total score of the power supply service based on the weighted score of each user;

[0015] After combining each time period, the basic weight is updated for each user according to the dynamic weight.

[0016] Furthermore, the determination of the environmental impact factors includes the following steps:

[0017] Collecting power supply status information within the preset time period;

[0018] The power supply status information is standardized and weighted calculation is performed according to the set weights of various indicators to obtain the initial value of the environmental impact factor of the time period;

[0019] Performing a gain adjustment on the initial value of the environmental impact factor according to the business busyness of the time period to obtain the adjusted environmental impact factor.

[0020] Furthermore, the method for determining the dynamic weight includes:

[0021] ,

[0022] in, represents the dynamic weight, Indicates the user's basic weight, Indicates the highest rating a user can give. represents the environmental impact factor, represents the average score, Represents a user 's rating.

[0023] Furthermore, the calculation of each user's weighted score based on the user's dynamic weight and score includes:

[0024] ,

[0025] in, Represents a user Rating, represents the dynamic weight, Represents a weighted score.

[0026] Furthermore, the updating of the basic weight for each user according to the dynamic weight after combining each time period includes one of the following strategies:

[0027] Update strategy 1:

[0028] ,

[0029] Update strategy 2:

[0030] ,

[0031] Update strategy three:

[0032] like , then trigger the update;

[0033] Update method: ,

[0034] Otherwise, keep the original basic weight unchanged;

[0035] Update strategy four:

[0036] ,

[0037] in: represents the new base weight; Indicates the basic weight of the previous period; represents the smoothing coefficient; Indicates the cumulative number of time periods in which the user participated in the evaluation; represents the dynamic weight in the 𝑡th time period; Indicates the highest score a user can give; Indicates the preset update trigger threshold; Represents the historical weight decay coefficient; Indicates the system preset neutral weight reference value.

[0038] The present invention also provides a power supply service evaluation system based on dynamic weight adjustment, comprising the following modules:

[0039] A collection module, configured to collect user ratings of power supply services and associate the ratings with user information for storage;

[0040] A statistics module is used to divide the ratings into preset time periods and calculate the average scores of all users in each time period;

[0041] A first determining module, configured to determine an environmental impact factor according to a power supply status within a preset time period;

[0042] A second determination module is configured to determine a dynamic weight of the user in the current time period based on the average score, the environmental impact factor, and the user's basic weight;

[0043] A first calculation module is used to calculate a weighted score for each user based on the user's dynamic weight and score;

[0044] The second calculation module is used to calculate the total score of the power supply service according to the weighted score of each user;

[0045] An updating module is used to update the basic weight of each user according to the dynamic weight after combining each time period.

[0046] Furthermore, the first determining module includes the following submodules:

[0047] A second acquisition module is used to collect power supply status information within the preset time period;

[0048] A weighted calculation module, configured to perform standardization processing on the power supply status information and perform weighted calculation according to the set weights of various indicators to obtain an initial value of the environmental impact factor for the time period;

[0049] The adjustment module is configured to perform a gain adjustment on the initial value of the environmental impact factor according to the business busyness of the time period to obtain the adjusted environmental impact factor.

[0050] Furthermore, the dynamic weight is determined using the following formula:

[0051] ,

[0052] in, represents the dynamic weight, Indicates the user's basic weight, Indicates the highest rating a user can give. represents the environmental impact factor, represents the average score, Represents a user 's rating.

[0053] Furthermore, the calculation of each user's weighted score based on the user's dynamic weight and score includes:

[0054] ,

[0055] in, Represents a user Rating, represents the dynamic weight, Represents a weighted score.

[0056] Furthermore, the updating of the basic weight for each user according to the dynamic weight after combining each time period includes one of the following strategies:

[0057] Update strategy 1:

[0058] ,

[0059] Update strategy 2:

[0060] ,

[0061] Update strategy three:

[0062] like , then trigger the update;

[0063] Update method: ,

[0064] Otherwise, keep the original basic weight unchanged;

[0065] Update strategy four:

[0066] ,

[0067] in: represents the new base weight; Indicates the basic weight of the previous period; represents the smoothing coefficient; Indicates the cumulative number of time periods in which the user participated in the evaluation; represents the dynamic weight in the 𝑡th time period; Indicates the highest score a user can give; Indicates the preset update trigger threshold; Represents the historical weight decay coefficient; Indicates the system preset neutral weight reference value.

[0068] The present invention provides a power supply service evaluation method and system based on dynamic weight adjustment. By introducing a dynamic weight calculation mechanism for user ratings, an adjustment strategy for environmental impact factors, and an adaptive update mechanism for user base weights, this method achieves a scientific evaluation of power supply service quality. Compared with existing technologies, it has the following beneficial effects:

[0069] The present invention dynamically calculates the weight of user ratings based on the deviation between the user ratings and the average ratings of the time period. If the rating is higher than the average, its weight is reduced; if the rating is lower than the average, its weight is increased, effectively weakening the impact of high-scoring cheating on the evaluation results, strengthening the weight of low-scoring feedback, and ensuring that the evaluation results truly reflect the power supply service level.

[0070] The present invention introduces environmental impact factors to quantify the power supply status (such as power outage frequency, number of complaints, business volume, etc.). When the external environment is abnormal, the contribution of the score deviation item to the weight is appropriately adjusted to buffer the score anomalies caused by environmental fluctuations and prevent service personnel or units from having their evaluation lowered due to abnormal scores under uncontrollable conditions.

[0071] By updating the basic weight of the user according to the current dynamic weight after each evaluation cycle, the system dynamically adjusts the influence of the user's rating in subsequent cycles, suppresses the long-term rating weight of users with abnormal ratings, and increases the weight of users with stable ratings, thereby achieving long-term adaptive optimization of the system's rating behavior.

[0072] The present invention combines seasonal changes and business processing peak and valley cycles, and dynamically adjusts the evaluation weight configuration according to key indicators such as business volume and service quality. For example, it appropriately increases the evaluation weight of service quality in the off-season to reflect users' higher service expectations and improve the sensitivity and accuracy of the evaluation results to the actual power supply service performance.

[0073] In summary, the present invention constructs a power supply service evaluation system with intelligent adaptability through dynamic scoring management and self-learning weight optimization mechanism, providing power companies with high-quality and quantifiable service quality evaluation basis, and promoting the continuous improvement and optimization of power supply services. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in 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.

[0075] Figure 1 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0076] Below, the invention is preferably described with reference to the accompanying drawings and specific embodiments.

[0077] like Figure 1 As shown, this embodiment describes a power supply service evaluation method based on dynamic weight adjustment, which periodically collects user rating data and determines the environmental impact factor in combination with the power supply status. According to the deviation between the user rating and the average score, the dynamic weight of the user rating is adjusted in real time, and then the score is weighted to ensure that the evaluation result is more objective and fair, and effectively suppresses the impact of abnormal rating or cheating behavior on the accuracy of power supply service evaluation, thereby improving the scientificity and accuracy of power supply service quality monitoring.

[0078] In this context, power supply service evaluation refers to a comprehensive, quantitative or qualitative assessment of the service provider (e.g., power supply company, service personnel, or related business departments) in terms of service quality, business processing efficiency, customer satisfaction, and business complexity handling capabilities during the provision of power services. This evaluation objectively reflects the overall quality of power supply services by collecting user feedback, particularly ratings, and combining it with actual business processing indicators.

[0079] The power supply service evaluation includes a digital evaluation of the service level demonstrated by the power supply enterprise in power supply, business acceptance, customer service, emergency response, fault handling and other aspects. The evaluation data includes at least user rating information, business processing information and auxiliary information related to the power supply status, which is used to reflect the service performance and customer satisfaction of the power supply service unit or personnel within a preset period.

[0080] Step S10: collecting the user's rating of the power supply service, and storing the rating in association with the user information.

[0081] After the power supply service is completed or phased out, the user's feedback information on the power supply service is collected through a preset service evaluation system. The feedback information includes but is not limited to numerical scores, textual evaluations, service satisfaction level selections, service time period selections, and service type identification information; wherein, the numerical score is a quantitative score given by the user to the power supply service quality based on subjective evaluation, and the score has a limited scoring range, preferably in the form of an integer or decimal between 1 and 10.

[0082] The user information includes, but is not limited to, user identification, user account information, electricity address information, service request number, scoring timestamp, terminal device identification, etc. The system associates the scoring data submitted by the user with the corresponding user information to form a data pair that can be used for subsequent statistical analysis, and stores the data pair in a preset data storage module to support subsequent operations such as time division, introduction of environmental factors, and weight calculation.

[0083] The scoring collection method can be manual input, mobile terminal submission, web page interactive submission or voice recognition conversion. The scoring collection system can be configured with a data verification module to verify the integrity and legality of the scoring data to ensure the accurate collection of the scoring data.

[0084] By associating and storing user scores with user information, structured management of score data is achieved, ensuring the traceability and integrity of the data in subsequent processing links, and providing reliable data support for dynamic weight adjustment and power supply service quality evaluation.

[0085] Step S20: Divide the ratings into preset time periods, and calculate the average scores of all users in each time period.

[0086] After collecting user rating data and completing the associated storage with user information, the rating data is divided into corresponding time intervals according to the set time division rules, and all user rating data in each time interval are statistically processed to obtain the average score in the time period for subsequent dynamic weight calculation.

[0087] Specifically, the time division rule is a system-preset evaluation period division strategy, preferably divided into units of days, weeks, months, or quarters. The time division granularity can also be adaptively adjusted based on business needs or service fluctuations. The time period is a contiguous, non-overlapping time interval used for time-series organization and batch processing of the scoring data.

[0088] In each time period, the system extracts all user ratings belonging to that time period , and calculate the value based on the following mathematical formula

[0089] Average score during the time period :

[0090] ,

[0091] in, Indicates the total number of users who participated in the rating in the current time period. Represents a user Submitted ratings.

[0092] The average score, as the statistical result of the current time period, will be used to adjust the dynamic weight of each user to determine the overall evaluation standard for the scoring behavior in the time period.

[0093] In addition, the system can optimize and adjust the time period division granularity and the statistical method of average score based on user activity, number of ratings and rating volatility in different time periods to improve data processing flexibility and evaluation accuracy. The optimization and adjustment specifically include the following methods:

[0094] Based on the user activity evaluation, the system counts the number of active users in each preset time period. The active users refer to users who submitted at least one rating in the time period. If the number of active users is higher than the first threshold, , the system will subdivide the time period into smaller sub-time periods; if the number of active users is lower than the second threshold , the system merges the current time period with the adjacent time period to form a time interval with larger granularity.

[0095] Based on the number of ratings, the system records the total number of ratings in each time period ; When the total number of ratings exceeds the preset rating threshold When calculating the average score, the system will give priority to using the median or weighted average method to reduce the impact of extreme scores; if the number of scores is small, the system will maintain the traditional arithmetic average calculation method, or perform weighted smoothing based on historical score data.

[0096] Based on the score volatility analysis, the system calculates the variance or standard deviation of the score in each time period, which is recorded as or ; If the score volatility is greater than the volatility threshold , the system introduces robust statistical methods, such as excluding the highest and lowest 5% scores before averaging, or using a weighted moving average method to reduce the impact of abnormal scores on the average score.

[0097] Adaptive adjustment based on time granularity, the system takes into account the number of active users , total number of ratings and rating volatility Afterwards, the time period division granularity is dynamically adjusted according to the following rules:

[0098] like and , then divide the current time period into multiple sub-time periods;

[0099] like and , then merge with the adjacent time period and extend the time interval;

[0100] The time period division results are fed back to the data processing module to optimize the subsequent scoring statistics and weight adjustment process in real time.

[0101] Furthermore, based on the rating distribution characteristics, the average value calculation method is dynamically selected, including arithmetic mean, weighted mean, truncated mean, median, etc. The weighting coefficient can be set according to the user's historical rating weight to ensure that the evaluation results are more real and reliable.

[0102] Through the above optimization process, the system realizes an adaptive adjustment mechanism for time period division and average score statistics, improves the processing ability of the power supply service evaluation system under the circumstances of diversified user rating behaviors and fluctuating business processing, enhances the ability to resist abnormal rating behaviors, and ensures that the evaluation results have statistical stability, computational robustness and universal application.

[0103] Step S30: determining an environmental impact factor according to the power supply status within a preset time period.

[0104] In each time period, the system collects status information related to power supply services, analyzes the objective power supply environment in that time period, quantifies environmental factors that may affect users' rating emotions and actual service experience, and generates an environmental impact factor 𝑘 for weight adjustment based on this. This factor is used to suppress rating deviations caused by environmental anomalies and enhance the fairness and robustness of the rating results.

[0105] Specifically, the power supply status includes but is not limited to the following information:

[0106] The number of planned power outages within the power supply area during this time period ;

[0107] Number of unplanned power outages or power outages due to faults ;

[0108] Average power outage duration for users ;

[0109] Grid load rate ;

[0110] Number of customer service complaints ;

[0111] Total business volume accepted by power supply units ;

[0112] Whether the time period is during business peak hours or special periods (such as holidays or severe weather).

[0113] Based on the power supply status parameters, the system determines the environmental impact factor according to the following calculation steps: :

[0114] Standardize each parameter and record the normalized value as:

[0115] ,

[0116] ,

[0117] ,

[0118] ,

[0119] ,

[0120] ,

[0121] Among them, the denominator of each indicator is the set maximum allowable value or historical maximum value, which is used for normalization processing.

[0122] Assume that the weight coefficients of each parameter are satisfy , then calculate the initial value of the environmental impact factor :

[0123] ,

[0124] If the time period is during business peak or emergency period, To make a gain adjustment:

[0125] ,

[0126] in, is the adjustment factor, the value range is , reflecting the degree of buffering against score fluctuations under special circumstances.

[0127] The final value range of the environmental impact factor 𝑘 is limited to:

[0128] ,

[0129] in, is 1, Can be set to MaxScore (such as 10) to avoid excessive adjustments affecting the score calculation.

[0130] Through the above calculation, the environmental impact factor 𝑘 can comprehensively reflect the stability of the power supply environment and the business pressure level in the current time period, effectively suppress the abnormal fluctuation of the score caused by the deterioration of the power supply status, make the dynamic weight adjustment more reasonable, and improve the adaptability and accuracy of the power supply service evaluation system in a complex power supply environment.

[0131] For example, in February 2025, a region experienced frequent power outages due to grid upgrades and renovations, resulting in a poor power supply environment and a significant increase in user complaints. The system collected power supply status data for this period, as shown in Table 1:

[0132] Table 1 Power supply status data

[0133] project Numerical Normalized maximum value (reference) Normalized results Planned power outages 9 times 10 times 0.9 Number of unplanned power outages 6 times 10 times 0.6 Average power outage duration 3 hours 4 hours 0.75 Grid load rate 90% 100% 0.9 Number of customer service complaints 180 items 200 items 0.9 Total business volume accepted 800 items 1000 pieces 0.8

[0134] The weight coefficient is:

[0135] ,

[0136] ,

[0137] ,

[0138] February is the peak period for power transformation, and the adjustment factor ,but:

[0139] ,

[0140] Step S40 , determining the dynamic weight of the user in the current time period according to the average score, the environmental impact factor and the user's basic weight.

[0141] The system retrieves the user from the database Ratings submitted during the current period , and get the average rating of all users in that time period , and the user base weights stored in previous cycles .

[0142] The system will use the user's basic weight Normalization is performed according to the maximum score MaxScore to eliminate the impact of differences in scoring standards on weight calculation, and the standardized basic weight is obtained:

[0143] ,

[0144] The system uses the following formula to determine the user Dynamic weight in the current time period :

[0145] ,

[0146] in:

[0147] The first term contributes to the base weight;

[0148] The second item is the contribution of score deviation, which is affected by environmental factors. After adjustment, the impact of score deviation on weight is controlled.

[0149] This step dynamically adjusts the influence of each user's rating by combining user rating behavior with the system's global statistical data to avoid unreasonable deviations in the evaluation results caused by a single rating. Its core calculation mechanism is based on the following principles:

[0150] User base weight This reflects the stability and authority of a user's historical ratings and is a historical dependency factor for dynamic weights. The system standardizes this to ensure uniform influence across different rating systems.

[0151] The difference between the user's rating and the average rating for the time period Measures the degree of deviation of individual scoring behavior, and the difference is affected by environmental factors Adjust to prevent excessive fluctuations in scores from amplifying the results.

[0152] The environmental impact factor 𝑘 reflects the interference of the external environment of the power supply service on the user's score. The worse the environment, the greater the 𝑘, thereby reducing the weight contribution of the score deviation item and achieving buffering and isolation of the external environment on the evaluation system.

[0153] Step S50: Calculate the weighted score of each user based on the user's dynamic weight and score.

[0154] After calculating the dynamic weight of the user's current time period, the system operates the dynamic weight with the user's original score to obtain the user's weighted score in the power supply service evaluation system, which is used to reflect the actual contribution value of the user's score to the service evaluation results.

[0155] Specifically, the weighted score reflects the actual influence of the user's score after the weight is corrected, which is used to eliminate the impact of score deviation and ensure the fairness and authenticity of the final score results. The system uses the following weighted calculation method:

[0156] Set user The original rating is , the dynamic weight is , the weighted score is ,but:

[0157] ,

[0158] Among them, if the weighted score exceeds the maximum score value MaxScore, MaxScore is taken as the upper limit of the weighted score of the user to prevent score distortion.

[0159] In this step, when the user rating is high and the weight is positive, the weighted rating is moderately increased to encourage positive rating behavior; when the user rating is low and the weight is negative, the weighted rating is moderately reduced to weaken the impact of malicious or abnormally low ratings on the evaluation results; the weighted rating does not exceed the preset maximum value to ensure the rationality of the rating and the stability of the results; the system quantifies the user rating contribution through weighted rating and improves the credibility of service evaluation.

[0160] Step S60: Calculate the total score of the power supply service according to the weighted score of each user.

[0161] After calculating each user's weighted score, the system aggregates all of the weighted scores to quantitatively reflect the overall service quality of the power supply service recipient (such as a service person, service unit, or business department) within a preset evaluation period. This total score serves as a reference for power supply service performance assessment, user satisfaction management, and service improvement.

[0162] The system obtains the weighted rating data of all valid users in the current evaluation cycle, recorded as ,in, is the total number of users who participated in the evaluation. is the weighted score of user 𝑖, which is the user's original score obtained by dynamic weight adjustment and has been capped according to the set upper limit.

[0163] In one selected method, the system uses arithmetic average to calculate the total score of the power supply service , the formula is as follows:

[0164] ,

[0165] In one option, based on actual needs, the system can assign higher weights to some user ratings and use a weighted average to calculate the total score:

[0166] ,

[0167] in, is the weighting coefficient of user 𝑖, which is used to enhance the rating influence of key users.

[0168] For example:

[0169] In a certain period, the highest score MaxScore = 10; the average score of the time period ; User base weight Environmental impact factor: Normal environment: k = 1, off-season (low volume, higher weight of business quality), k = 2

[0170] Table 2 User rating data

[0171] user User Rating illustrate User A 9 High score, satisfactory, slightly high User B 4 Low scores, dissatisfaction, mood swings User C 7 Equal to average, neutral User D 8 Slightly higher score, more positive User E 5 Lower scores may indicate emotional low evaluation User F 6 Slightly below average, normal fluctuation

[0172] Table 3 Dynamic weight (k = 1)

[0173] user Calculation process Dynamic Weight User A 0.05 + (7-9) / 10 = 0.05 - 0.2 -0.15 User B 0.05 + (7-4) / 10 = 0.05 + 0.3 0.35 User C 0.05 + 0 = 0.05 0.05 User D 0.05 + (7-8) / 10 = 0.05 - 0.1 -0.05 User E 0.05 + (7-5) / 10 = 0.05 + 0.2 0.25 User F 0.05 + (7-6) / 10 = 0.05 + 0.1 0.15

[0174] Table 4 Dynamic weight (k = 2, business off-season)

[0175] user Calculation process Dynamic Weight User A 0.05 + (7-9) / 20 = 0.05 - 0.1 -0.05 User B 0.05 + (7-4) / 20 = 0.05 + 0.15 0.2 User C 0.05 + 0 = 0.05 0.05 User D 0.05 + (7-8) / 20 = 0.05 - 0.05 0 User E 0.05 + (7-5) / 20 = 0.05 + 0.1 0.15 User F 0.05 + (7-6) / 20 = 0.05 + 0.05 0.10

[0176] Table 5 If there is no dynamic weight (traditional method)

[0177] user Original score Weighted score = original score User A 9 9 User B 4 4 User C 7 7 User D 8 8 User E 5 5 User F 6 6

[0178] Total score = (9+4+7+8+5+6) / 6 = 6.5

[0179] With dynamic weights, k = 1

[0180] user calculate Weighted Scoring User A 9 × 0.85 = 7.65 7.65 User B 4 × 1.35 = 5.4 5.4 User C 7 × 1.05 = 7.35 7.35 User D 8 × 0.95 = 7.6 7.6 User E 5 × 1.25 = 6.25 6.25 User F 6 × 1.15 = 6.9 6.9

[0181] Total score ≈ (7.65+5.4+7.35+7.6+6.25+6.9) / 6 ≈ 6.86

[0182] Table 6 Dynamic weight + environmental factor (k = 2, business off-season)

[0183] user calculate Weighted Scoring User A 9 × 0.95 = 8.55 8.55 User B 4 × 1.2 = 4.8 4.8 User C 7 × 1.05 = 7.35 7.35 User D 8 × 1 = 8 8 User E 5 × 1.15 = 5.75 5.75 User F 6 × 1.1 = 6.6 6.6

[0184] Total score ≈ (8.55+4.8+7.35+8+5.75+6.6) / 6 ≈ 6.84

[0185] Table 7 Comparison of several evaluation tests

[0186] Evaluation method Total score High-scoring user contributions Impact of low-scoring users System Advantages No dynamic weights (traditional) 6.5 Unregulated Directly pull down It is easily affected by low scores and cannot reflect the adjustment function of score deviation. Dynamic weight, no environmental factors 6.86 Weakened Buffered Suppress high scores, amplify real low scores, and improve fairness Dynamic weight + environmental factors (off-season) 6.84 Partial buffer Moderate buffer The weight of business quality in the off-season is increased, the system buffers score fluctuations, and has strong stability

[0187] As can be seen from the above examples, the present invention reduces the weighting of high scores to prevent fraudulent score manipulation, while increasing the weighting of low scores to highlight issues. During off-seasons, k increases, mitigating scoring bias and preventing low scores from concentrating on the overall rating in adverse environments. Dynamic weighting works in conjunction with environmental factors to ensure that the overall score both truly reflects the service and is resistant to abnormal interference.

[0188] Step S70: After combining each time period, the basic weight is updated for each user according to the dynamic weight.

[0189] After each evaluation cycle, the user's rating performance during that period and the calculated dynamic weight are used. , using the preset weight update strategy to update the user's basic weight Adjustments are made to dynamically reflect the long-term credibility of user rating behavior and to reasonably control its influence in subsequent ratings. The update strategy can be any of the following, or based on a fusion of multiple strategies.

[0190] Update strategy 1:

[0191] ,

[0192] in, Indicates the smoothing coefficient, which controls the ratio of historical weight to current weight. .

[0193] Strategy 1 can suppress the sharp fluctuations caused by short-term score anomalies; maintain the long-term continuity of score influence, and is suitable for stable user groups; the system operates smoothly, and is suitable for daily power supply service evaluation.

[0194] Update strategy 2:

[0195] ,

[0196] in, Indicates the cumulative number of time periods in which the user participated in the evaluation; Denotes the dynamic weight in the 𝑡th time period.

[0197] Strategy 2 is suitable for scenarios where users' rating behavior is continuously tracked over time and their weights are gradually adjusted. Each update considers the dynamic weights over the entire historical cycle to reflect the long-term trends of users' ratings. This suppresses the interference of abnormal weights in individual time periods and improves weight stability. It is suitable for system environments with long-term user activity and frequent ratings.

[0198] Update strategy three:

[0199] like , then trigger the update;

[0200] Update method: ,

[0201] Otherwise, keep the original base weight unchanged.

[0202] in, Indicates the preset update trigger threshold, preferably 0.2~0.3.

[0203] Strategy three is suitable for identifying and responding to sudden changes in rating behavior or users with obvious tendency to cheat. It can quickly respond to abnormal fluctuations in user ratings and adjust weights in a timely manner. It is suitable for identifying cheating behavior and forcibly correcting the influence of ratings. It also avoids frequent changes in normal rating users and improves the system's ability to resist attacks.

[0204] Update strategy four:

[0205] ,

[0206] in, Represents the historical weight decay coefficient, preferably ;

[0207] Indicates the system preset neutral weight reference value, such as 0.5~1.

[0208] Strategy 4 is used to control the long-term accumulation of historical weights and prevent old data from dominating the scoring process. It gradually brings historical weights closer to the system standard, eliminating long-term accumulated bias. It also helps maintain a balanced scoring influence by periodically resetting the system to zero. It also simplifies data requirements and is suitable for resource-constrained scenarios.

[0209] In another embodiment, the present invention further provides a power supply service evaluation system based on dynamic weight adjustment, comprising:

[0210] A collection module, configured to collect user ratings of power supply services and associate the ratings with user information for storage;

[0211] A statistics module is used to divide the ratings into preset time periods and calculate the average scores of all users in each time period;

[0212] A first determining module, configured to determine an environmental impact factor according to a power supply status within a preset time period;

[0213] A second determination module is configured to determine a dynamic weight of the user in the current time period based on the average score, the environmental impact factor, and the user's basic weight;

[0214] A first calculation module is used to calculate a weighted score for each user based on the user's dynamic weight and score;

[0215] The second calculation module is used to calculate the total score of the power supply service according to the weighted score of each user;

[0216] An updating module is used to update the basic weight of each user according to the dynamic weight after combining each time period.

[0217] It should be noted that the explanation of the above-mentioned embodiment of the power supply service evaluation method based on dynamic weight adjustment is also applicable to the device of the embodiment of the present application and will not be repeated here.

[0218] Those skilled in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented using a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0219] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0220] In the several embodiments provided in this application, if any 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 this application, 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 method described in each embodiment of this application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), magnetic disk or optical disk, and other media that can store program code.

[0221] The above is only a specific implementation method of the present application. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be included in the protection scope of this application. The protection scope of this application should be based on the protection scope of the claims. For some module structures that are not particularly clear in the present invention, the content recorded in the prior art shall prevail. The prior art mentioned in the above background technology section and the specific embodiment section of the present invention can be regarded as part of the present invention and used to understand the meaning of some technical features or parameters.

Claims

1. A power supply service evaluation method based on dynamic weight adjustment, characterized in that: The method comprises the following steps: Collect user ratings of power supply services and store the ratings in association with user information; divide the ratings into preset time periods and calculate the average scores of all users in each time period; Determine the environmental impact factor based on the power supply status within a preset time period; Determine the dynamic weight of the user in the current time period based on the average score, the environmental impact factor, and the user's basic weight; Calculate each user's weighted score based on the user's dynamic weight and their score; Calculate the total score of the power supply service based on the weighted score of each user; After combining each time period, the basic weight is updated for each user according to the dynamic weight; The determination of the environmental impact factors comprises the following steps: Collecting power supply status information within the preset time period; The power supply status information is standardized and weighted calculation is performed according to the set weights of various indicators to obtain the initial value of the environmental impact factor of the time period; Performing a gain adjustment on the initial value of the environmental impact factor according to the business busyness of the time period to obtain the adjusted environmental impact factor; Methods for determining dynamic weights include: Among them, W i Represents the dynamic weight, W i0 Represents the user's basic weight, MaxScore represents the highest score a user can give, and k represents the environmental impact factor. represents the average score, S i represents the rating of user i.

2. The power supply service evaluation method based on dynamic weight adjustment according to claim 1, characterized in that: Calculating each user's weighted score based on the user's dynamic weight and score includes: S i ′=S i ·(1+W i ) Among them, S i represents the rating of user i, W i represents the dynamic weight, S i ′ indicates weighted score.

3. The power supply service evaluation method based on dynamic weight adjustment according to claim 1, characterized in that: The updating of the basic weight for each user according to the dynamic weight after combining each time period includes one of the following strategies: Update strategy 1: Update strategy 2: Update strategy three: like Then trigger the update; Update method: Otherwise, keep the original basic weight unchanged; Update strategy four: in: represents the new base weight; represents the basic weight of the previous period; α represents the smoothing coefficient; T represents the cumulative number of time periods in which users have participated in the evaluation; W i (t) Represents the dynamic weight within the tth time period; MaxScore represents the highest score that the user can give; Δ represents the preset update trigger threshold; β represents the historical weight attenuation coefficient; μ represents the system preset neutral weight reference value.

4. A power supply service evaluation system based on dynamic weight adjustment, characterized in that: The system includes the following modules: A collection module, configured to collect user ratings of power supply services and associate the ratings with user information for storage; A statistics module is used to divide the ratings into preset time periods and calculate the average scores of all users in each time period; A first determining module, configured to determine an environmental impact factor according to a power supply status within a preset time period; A second determination module is configured to determine a dynamic weight of the user in the current time period based on the average score, the environmental impact factor, and the user's basic weight; A first calculation module is used to calculate a weighted score for each user based on the user's dynamic weight and score; The second calculation module is used to calculate the total score of the power supply service according to the weighted score of each user; An updating module, configured to update the basic weight for each user according to the dynamic weight after combining each time period; The first determination module includes the following submodules: a second collection module for collecting power supply status information within the preset time period; a weighted calculation module for standardizing the power supply status information and performing weighted calculation according to the set weights of various indicators to obtain an initial value of the environmental impact factor for the time period; The adjustment module is configured to perform a gain adjustment on the initial value of the environmental impact factor according to the business busyness of the time period to obtain the adjusted environmental impact factor; the dynamic weight is determined using the following formula: Among them, W i Represents the dynamic weight, W i0 Represents the user's basic weight, MaxScore represents the highest score a user can give, and k represents the environmental impact factor. represents the average score, S i represents the rating of user i.

5. The power supply service evaluation system based on dynamic weight adjustment according to claim 4, characterized in that: Calculating each user's weighted score based on the user's dynamic weight and score includes: S i ′=S i ·(1+W i ) Among them, S i represents the rating of user i, W i represents the dynamic weight, S i ′ indicates weighted score.

6. The power supply service evaluation system based on dynamic weight adjustment according to claim 4, characterized in that: The updating of the basic weight for each user according to the dynamic weight after combining each time period includes one of the following strategies: Update strategy 1: Update strategy 2: Update strategy three: like Then trigger the update; Update method: Otherwise, keep the original basic weight unchanged; Update strategy four: in: represents the new base weight; represents the basic weight of the previous period; α represents the smoothing coefficient; T represents the cumulative number of time periods in which users participated in the evaluation; W i (t) represents the dynamic weight in the tth time period; MaxScore represents the highest score a user can give; Δ represents the preset update trigger threshold; β represents the historical weight attenuation coefficient; μ represents the system preset neutral weight reference value.

Images