A method and apparatus for determining network experience satisfaction

Abstract

The application discloses a kind of method and device for determining network experience satisfaction, to solve the problem of inaccurate network experience satisfaction evaluation result.The scheme provided in the present application includes: obtaining the network experience satisfaction score and network use information of multiple network users;According to the network use information of multiple network users, clustering analysis is performed on multiple network users to obtain multiple groups of network users based on network service type classification, and the network service type represents the network function use preference of network users;Determine the network score weight of multiple network users according to the network service type to which multiple network users belong;According to the network score weight and network experience satisfaction score of multiple network users, determine the network experience satisfaction evaluation result.The scheme of the embodiment of the present application generates the network experience satisfaction evaluation result according to the use of network service by network users, effectively improves the accuracy of the generated evaluation result.

Description

Technical Field

[0001] This invention relates to the field of communications, and more particularly to a method and apparatus for determining network experience satisfaction. Background Technology

[0002] With the development of wireless internet and 5G technology, internet users have increasingly higher demands for network and mobile internet products. To meet these diverse needs, network services are also diversifying. To understand user satisfaction with network services, user satisfaction surveys are typically conducted.

[0003] In practical applications, different network users have significantly different needs for network services. This makes the satisfaction evaluations made by network users rather one-sided. Such one-sided evaluations will seriously affect the network service satisfaction evaluation results, making the satisfaction evaluation results also rather one-sided and difficult to accurately reflect the user satisfaction of network services.

[0004] How to improve the accuracy of online experience satisfaction evaluation results is the technical problem that this application aims to solve. Summary of the Invention

[0005] The purpose of this application is to provide a method and apparatus for determining network experience satisfaction, so as to solve the problem of inaccurate network experience satisfaction evaluation results.

[0006] Firstly, a method for determining network experience satisfaction is provided, including:

[0007] Obtain network experience satisfaction ratings and network usage information from multiple network users;

[0008] Cluster analysis is performed on the multiple network users based on their network usage information to obtain multiple groups of network users classified according to network service type, where the network service type represents the network user's network function usage preference.

[0009] The network score weights of the multiple network users are determined based on the network service types to which the multiple network users belong;

[0010] The network experience satisfaction evaluation result is determined based on the network rating weights and network experience satisfaction scores of the multiple network users.

[0011] Secondly, an apparatus for determining network experience satisfaction is provided, comprising:

[0012] The acquisition module retrieves network experience satisfaction ratings and network usage information from multiple network users.

[0013] The clustering module performs clustering analysis on the multiple network users based on their network usage information to obtain multiple groups of network users classified according to network service type, where the network service type represents the network user's network function usage preference.

[0014] The first determining module determines the network score weights of the multiple network users based on the network service types to which the multiple network users belong;

[0015] The second determining module determines the network experience satisfaction evaluation result based on the network rating weights and network experience satisfaction scores of the multiple network users.

[0016] Thirdly, an electronic device is provided, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method of the first aspect.

[0017] Fourthly, a computer-readable storage medium is provided on which a computer program is stored, which, when executed by a processor, implements the steps of the method of the first aspect.

[0018] In this embodiment, network experience satisfaction scores and network usage information of multiple network users are obtained; cluster analysis is performed on the multiple network users based on their network usage information to obtain multiple groups of network users categorized by network service type, where the network service type represents the network user's network function usage preference; network score weights for the multiple network users are determined based on their respective network service types; and a network experience satisfaction evaluation result is determined based on the network score weights and network experience satisfaction scores of the multiple network users. This embodiment of the invention generates network experience satisfaction evaluation results based on network users' network service usage, effectively improving the accuracy of the generated evaluation results. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0020] Figure 1 This is one of the flowcharts illustrating a method for determining network experience satisfaction according to an embodiment of the present invention.

[0021] Figure 2 This is a second flowchart illustrating a method for determining network experience satisfaction according to an embodiment of the present invention.

[0022] Figure 3This is the third flowchart illustrating a method for determining network experience satisfaction according to an embodiment of the present invention.

[0023] Figure 4 This is the fourth flowchart illustrating a method for determining network experience satisfaction according to an embodiment of the present invention.

[0024] Figure 5 This is the fifth flowchart illustrating a method for determining network experience satisfaction according to an embodiment of the present invention.

[0025] Figure 6 This is a schematic diagram illustrating an embodiment of the present invention of removing network user samples based on outliers.

[0026] Figure 7 This is the sixth flowchart illustrating a method for determining network experience satisfaction according to an embodiment of the present invention.

[0027] Figure 8 This is a schematic diagram of a device for determining network experience satisfaction according to an embodiment of the present invention. Detailed Implementation

[0028] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. The drawing numbers in this application are only used to distinguish the various steps in the solution and are not used to limit the execution order of the various steps. The specific execution order is as described in the specification.

[0029] To address the problems existing in the prior art, embodiments of this application provide a method for determining network experience satisfaction, such as... Figure 1 As shown, it includes:

[0030] S11: Obtain network experience satisfaction ratings and network usage information from multiple network users.

[0031] In this step, network user satisfaction can be obtained through various methods such as telephone, SMS, and push notifications. For example, if a telephone survey is used to obtain network experience satisfaction, customer service personnel can randomly select network users for a telephone survey, inviting them to rate their satisfaction with their mobile network usage. For example, the score can be from 1 to 10, where 1 represents very dissatisfied and 10 represents very satisfied. The network experience satisfaction obtained in this way can characterize the network user's satisfaction with the network experience, thereby revealing network user loyalty, which is beneficial for optimizing network services and improving network quality.

[0032] The aforementioned network experience satisfaction rating can be directly assessed by network users, or it can be automatically generated based on user complaint records, star ratings, and comment content. Specifically, it can be expressed in the form of a numerical score, or in the form of rating levels such as "very satisfied," "somewhat satisfied," or "dissatisfied." This plan does not limit the specific format of the network experience satisfaction rating.

[0033] The aforementioned network usage information may specifically include call detail records (CDRs) related to the mobile network, key information records of network data traffic, traffic logs, and other information. Network usage information includes historical records of network users' use of network functions, which can characterize the types of network functions used, the duration of use, and the frequency of use.

[0034] S12: Perform cluster analysis on the multiple network users based on their network usage information to obtain multiple groups of network users classified according to network service type, whereby the network service type represents the network user's network function usage preference.

[0035] In this step, K-Means clustering or other clustering algorithms can be used to perform cluster analysis on multiple network users. Since the aforementioned network usage information includes historical information about network users' use of network functions, this step uses cluster analysis to cluster multiple network users based on their network function usage preferences, resulting in multiple groups of network users with different preferences. Each group of network users can correspond to a different network service type, which characterizes the network function usage preferences of the users. For example, group A of network users prefers using online games, while group B prefers watching online videos, etc.

[0036] S13: Determine the network score weights of the multiple network users based on the network service types to which they belong.

[0037] Different network users have different preferences for network functions. Some users tend to favor a single network function, while others may use multiple functions extensively. This results in varying user experiences with different network functions. Users who have in-depth experience with a network function tend to provide more objective and accurate evaluations, while those with only superficial experience tend to provide less accurate assessments. Therefore, by determining the network rating weights for multiple user groups based on their respective network service types obtained through clustering, the evaluations from users with in-depth experience can be given higher weight, improving the accuracy of the final evaluation results.

[0038] S14: Determine the network experience satisfaction evaluation result based on the network rating weights and network experience satisfaction scores of the multiple network users.

[0039] The steps described above involved clustering multiple network users based on their network usage information and determining network rating weights for each user based on the clustering results. In this step, a network experience satisfaction evaluation result is generated based on the network rating weights determined in the previous steps and the network experience satisfaction scores of multiple users. This effectively improves the accuracy of the generated evaluation result and avoids negative impacts from evaluations made by users with only a superficial experience of the network functions.

[0040] Based on the solution provided in the above embodiments, optionally, step S11 above, as follows: Figure 2 As shown, it includes:

[0041] S21: Obtain network experience satisfaction scores from multiple network users, wherein the network experience satisfaction scores include satisfaction scores generated based on the multiple network users' use of at least one network function within a preset time period.

[0042] In this step, we obtain the satisfaction ratings generated by network users after using network functions within a preset time period. Since network experience satisfaction is time-sensitive, and users' satisfaction with the network experience often varies at different times, this step specifically obtains user evaluations of the network functions used within the preset time period. This ensures that the obtained network experience satisfaction ratings correspond to the network functions used during the preset time period, improving the accuracy of the final evaluation results.

[0043] S22: Obtain network usage information of the multiple network users within the preset time period, the network usage information including the usage of at least one network function by the network users within the preset time period.

[0044] In this step, network usage information of network users within a preset time period is selectively obtained, which corresponds to the network experience satisfaction score obtained in step S21 above. This ensures that the weights of the network scores generated in subsequent steps are consistent with the actual network functions experienced by network users, thereby improving the accuracy of the final network experience satisfaction evaluation result.

[0045] Based on the solutions provided in the above embodiments, optionally, the network usage information includes the usage of at least one of the following network functions by a network user during the preset time period:

[0046] Duration of online instant messaging usage, duration of online information browsing, duration of online gaming, data usage for online music playback, and duration of online video playback.

[0047] The network usage information mentioned above can specifically include the usage of various network functions used by network users. To facilitate subsequent processing and analysis, the usage of these network functions can be recorded in a table, as shown in Table 1:

[0048] Table 1 Network Usage Information of Network Users

[0049]

[0050] The network usage information obtained through the solutions provided in this application can characterize the network user's use of at least one network function, and thus characterize the depth of the network user's use of the network function. This facilitates improving the accuracy of the network scoring weights determined in subsequent steps, and ensures that the determined network scoring weights accurately reflect the validity of the scores made by the network user.

[0051] Based on the solutions provided in the above embodiments, optionally, such as Figure 3 As shown, step S12 above includes:

[0052] S31: Perform cluster analysis on the multiple network users based on the usage of at least one network function in the network usage information of the multiple network users to obtain multiple groups of network users classified according to network service type.

[0053] For example, in this step, multiple network users can be divided into k categories according to the network service type. In this embodiment, the K-Means clustering algorithm can be used to perform cluster analysis on the above multiple network users. Assuming that the network usage information includes m network function usage items, an m-dimensional space can be constructed in this step. The value of each network function usage item corresponds to the position of the network user in that dimension, and the network users are clustered in the m-dimensional space according to distance.

[0054] Suppose we need to calculate the distance between network user X and network user Y in m-dimensional space, we can use the following formula (1) to calculate it:

[0055]

[0056] The K-Means clustering algorithm can be used to obtain k network user group classifications, namely Type1, Type2, Type3...Type k As shown in Table 2 below:

[0057] Table 2. Correspondence between Network Users and Network Service Types

[0058] Network user serial number i Network service types of network users Business requirements characteristics 1 <![CDATA[Type1]]> Online games are the main focus. 2 <![CDATA[Type2]]> Primarily news browsing and instant messaging 3 <![CDATA[Type3]]> Online video is the main source of information. 4 <![CDATA[Type6]]> Online games and online videos are the main products. …… …… ……

[0059] In Table 2, the network user serial number i is used to distinguish different network users. In practical applications, other forms of identification, such as user IP information, user mobile phone number, and user MAC address, can also be used. The network service types and service demand characteristics of the above network users correspond to each other. This correspondence can be marked in the table as shown in Table 2, or it can be stored in another table associated with the network user serial number or the network service type of the network user.

[0060] Step S13 above includes:

[0061] S32: Use principal component analysis to determine the experience factors corresponding to the network service types of the multiple network users, whereby the experience factors represent the functional indicators of the corresponding network services.

[0062] S33: Determine the network score weights of the multiple network users based on the experience factors corresponding to the network service types to which the multiple network users belong.

[0063] To assess how well users utilize network functions, this step selects various metrics that characterize the depth and breadth of user network usage as experience factors. Based on the k user group classifications obtained from the previous steps, an experience factor for each user group is selected from the total number of experience factors. This experience factor reflects the degree of network experience experienced by users in the corresponding group.

[0064] Specifically, the aforementioned experience factors can be determined through refined identification using Deep Packet Inspection (DPI). For example, selections can be made based on services such as instant messaging, mobile online games, online video, and online music. For instant messaging, these could include metrics such as "number of instant messaging text chat messages," "number of instant messaging voice chat messages," "number of instant messaging image sends," and "number of instant messaging image receives." For online video, these could include metrics such as "number of initial video streaming requests," "number of video streaming buffer requests," and "number of video streaming playback completed." The various metrics are recorded in Table 3 below.

[0065] Table 3 Correspondence between Network Service Types and Experience Factors

[0066]

[0067] Based on the solutions provided in the above embodiments, optionally, such as Figure 4 As shown, step S33 above includes:

[0068] S41: Determine the contribution of the multiple experience factors to the network services based on the correlation between the experience factors and the network services;

[0069] S42: Determine the network score weights of the multiple network users based on the contribution of the multiple experience factors to network services.

[0070] To accurately select the experience factors for network user groups corresponding to various network service types and narrow down the range of evaluation indicators, principal component analysis can be used to determine the experience factors that are most strongly correlated with various network services as the experience factors corresponding to the network service types.

[0071] Taking the calculation of the experience factor for a certain network service type as an example, assuming there are 'a' users of Type 1 network and 'b' selectable network experience factors, the experience factor corresponding to the network service type can be calculated as follows:

[0072] Organize the b experience factor data of a network users into a factor correlation matrix Z with b rows and a columns, as shown in the following formula (2);

[0073]

[0074] Any column in the matrix shown in formula (2) above can be understood as a b-dimensional variable matrix x = (x1, x2, ..., xn). b ) T Therefore, covariance can be used to measure the degree to which each dimension deviates from its mean. Since there are b dimensional factors, the b-dimensional covariance matrix is ​​defined as follows (3):

[0075]

[0076] Based on the eigenvalues ​​of C and the corresponding orthogonal eigenvectors, the principal components b1, b2, ..., b can be solved. b Furthermore, the contribution of each principal component to x is calculated. Based on the highest to lowest contribution, the principal components required when the cumulative contribution rate is ≥90% are selected as the minimum experience key factors for this requirement. These experience key factors are then determined as the experience factors corresponding to the network service type, as shown in Table 4 below:

[0077] Table 4. Correspondence between Experience Factor and Contribution

[0078]

[0079] Based on the above steps, the experience factors corresponding to each type of network service can be determined, and then the network score weights for each group of network users can be determined, effectively improving the accuracy of the subsequently generated network experience satisfaction evaluation results.

[0080] Optionally, based on the experience factors and contributions determined in the above steps, and according to the experience factors and contributions of the network user groups corresponding to each network service type, the calculation rules for the experience score P (0≤P≤10) of that group of network users are determined, where a larger P indicates a deeper level of experience with the network functions by the network user. This experience score can be directly used as the network rating weight for network users, or used to determine the network rating weight for network users.

[0081] For example, suppose a certain type of network service corresponds to a group of n network users, and the experience factors for this group are F1, F2, and F3. The index values ​​of each network user experience factor are recorded as follows: and

[0082] The contribution values ​​of the experience factors are FC1, FC2, and FC3. The weights of the three experience factors in the network user's experience score P, FW1, FW2, and FW3, can be calculated based on their contribution values, as shown in the following formula (4):

[0083]

[0084]

[0085] A network user rates the user experience on three factors: F1, F2, and F3, and records the scores as follows: and As shown in equation (5):

[0086]

[0087] After calculating the scores for each experience factor, multiply them by the relevant weighting coefficients FW1, FW2, and FW3 respectively, and then sum them to obtain the corresponding network user experience score P. n The calculation process is shown in equation (6) below:

[0088]

[0089] Subsequently, authoritative network users can be identified based on network user experience scores, and then the network rating weights of authoritative network users can be determined, thereby determining the network experience satisfaction evaluation results.

[0090] For example, user experience ratings (P) can be used to assess the performance of each network user. n Arrange them in ascending order, and calculate their first quartile, denoted as Q1. The rule for calculating the position of Q1 is as shown in the following formula (7):

[0091] Q1=(n+1)×0.25 (7)

[0092] The user experience score for network users at location Q1 is denoted as P. Q1 The user experience rating (P) of the network users n <P Q1 Network users who are deemed to have low authority are categorized as low-authority network users, and their experience rating (P) is determined accordingly. n ≥P Q1 The network users are denoted as high-authority network users, and each high-authority network user is assumed to have a satisfaction rating of S. n The final network experience satisfaction evaluation result S can be derived from the mean value. The calculation result is shown in the following formula (8):

[0093]

[0094] The solution provided in this application enables the evaluation of network satisfaction based on network user experience. This solution can more accurately and effectively assess network user satisfaction with the network, which is beneficial for guiding network operation and maintenance and marketing efforts.

[0095] Based on the solutions provided in the above embodiments, optionally, to further optimize the clustering effect, edge network users can be removed, such as... Figure 5 As shown, step S31 above includes:

[0096] S51: Determine the outlier value of the plurality of network users based on the usage of the at least one network function;

[0097] S52: Perform cluster analysis on network users whose outlier values ​​are less than a preset outlier value based on the usage of at least one network function in the network usage information of the multiple network users.

[0098] In the solution provided in this application embodiment, the network utility information of the network users to be clustered is cleaned before performing cluster analysis.

[0099] Assume there are n network users to be clustered, and the network usage information includes m items related to network function usage. To improve the accuracy of network service type clustering analysis, this step analyzes outliers in each network function usage item to remove outlier survey samples and ensure the clustering results are reasonable.

[0100] Specifically, taking the Standard Deviation method as an example, suppose we need to remove outliers based on call duration (CT):

[0101] Calculate the average score The standard deviation σ of the scores. The mean is calculated as shown in equation (9):

[0102]

[0103] The standard deviation is calculated as shown in equation (10):

[0104]

[0105] The overall distribution of network user call durations follows a normal distribution. Outliers with excessively long or short call durations need to be removed. A sample of network users within a standard deviation σ, such as Figure 6 As shown.

[0106] That is, removing the score CT i In this embodiment, for example, only 68% of the network user sample is retained within the following range:

[0107] 1)

[0108] 2)

[0109] The solution provided in this application allows for the screening of network usage information of network users to be clustered before cluster analysis, eliminating marginal network users and optimizing clustering results. This solution utilizes a clustering algorithm to classify network users' business needs based on XDR data. It accurately identifies the experience factors of each group of network users using a unified DPI business identification and component analysis algorithm. By calculating the network user experience scores, it ultimately identifies authoritative network users and comprehensively determines network user satisfaction with network functions. Specifically, this solution determines different experience factors based on different business need types, reasonably assessing the depth of network user experience. Furthermore, this solution determines the network rating weights of network users based on their depth of experience with network functions, effectively reducing the impact of subjective evaluations from less frequent network users on the overall results and improving the accuracy of satisfaction assessment.

[0110] Based on the solutions provided in the above embodiments, optionally, such as Figure 7 As shown, after step S14, the following steps are also included:

[0111] S71: When the network experience satisfaction evaluation result meets the preset compensation standard, perform compensation operations on network users whose network experience satisfaction scores are lower than the preset compensation scores; and / or,

[0112] S72: When the network experience satisfaction evaluation result meets the preset satisfaction standard, send network service promotion information to network users whose network experience satisfaction score is higher than the preset satisfaction score.

[0113] In the solution provided in this application embodiment, after determining the network experience satisfaction evaluation result, compensation or network service promotion is performed on different network users based on the network experience satisfaction. Specifically, the preset compensation standard can be manually preset or automatically generated by electronic devices. When the network experience satisfaction evaluation result meets the preset compensation standard, it indicates that the network user experience is generally poor. At this time, targeted compensation operations can be performed on network users whose network experience satisfaction score is lower than the preset compensation frequency score. For example, free network traffic or free network service memberships can be given to network users to compensate for the poor network experience of network users in historical periods.

[0114] The aforementioned preset satisfaction standards can be manually set or automatically generated by electronic devices. When the network experience satisfaction evaluation results meet the preset satisfaction standards, it indicates that the network users' experience is generally good. At this point, promotional information for network services can be pushed to network users whose network experience satisfaction scores are higher than the preset satisfaction scores. This can be done via SMS, telephone, or push notifications. This promotional information can specifically include information about new network services or network services that the user may not currently be using but might be interested in, increasing the likelihood of users subscribing to new network services and improving the success rate of network service promotion.

[0115] To address the problems existing in the prior art, embodiments of this application provide a device 80 for determining network experience satisfaction, such as... Figure 8 As shown, it includes:

[0116] Module 81 retrieves network experience satisfaction ratings and network usage information from multiple network users.

[0117] Clustering module 82 performs clustering analysis on the multiple network users based on their network usage information to obtain multiple groups of network users classified according to network service type, wherein the network service type represents the network user's network function usage preference;

[0118] The first determining module 83 determines the network score weight of the multiple network users based on the network service type to which the multiple network users belong;

[0119] The second determining module 84 determines the network experience satisfaction evaluation result based on the network rating weights and network experience satisfaction scores of the multiple network users.

[0120] The apparatus provided in this application provides the following steps: First, it acquires network experience satisfaction scores and network usage information from multiple network users. Then, it performs cluster analysis on these users based on their network usage information to obtain multiple groups of network users categorized by network service type, where each network service type represents a user's network function usage preference. Next, it determines the network score weights for each user based on their respective network service types. Finally, it determines the network experience satisfaction evaluation result based on the network score weights and network experience satisfaction scores. This invention generates network experience satisfaction evaluation results based on users' network service usage, effectively improving the accuracy of the generated evaluation results.

[0121] Preferably, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the various processes of the above-described method embodiment for determining network experience satisfaction and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0122] This invention also provides a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the various processes of the above-described method embodiment for determining network experience satisfaction, and achieves the same technical effect. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0123] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0124] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0125] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of the present invention.

Claims

1. A method for determining network experience satisfaction, characterized in that, include: Obtain network experience satisfaction ratings and network usage information from multiple network users; Cluster analysis is performed on the multiple network users based on their network usage information to obtain multiple groups of network users classified according to network service type, where the network service type represents the network user's network function usage preference. The network score weights of the multiple network users are determined based on the network service types to which the multiple network users belong; The network experience satisfaction evaluation result is determined based on the network rating weights and network experience satisfaction scores of the multiple network users. Cluster analysis is performed on the multiple network users based on their network usage information to obtain multiple groups of network users classified according to network service type, including: Based on the usage of at least one network function in the network usage information of the multiple network users, the K-Means clustering algorithm is used to perform clustering analysis on the multiple network users to obtain multiple groups of network users classified according to network service type; The determination of network score weights for the multiple network users based on their network service types includes: Principal component analysis is used to determine the experience factors corresponding to the network service types to which the multiple network users belong, and the experience factors represent the functional indicators of the corresponding network services. The network score weights of the multiple network users are determined based on the experience factors corresponding to the network service types to which the multiple network users belong.

2. The method as described in claim 1, characterized in that, Obtain network experience satisfaction ratings and network usage information from multiple network users, including: Obtain network experience satisfaction scores from multiple network users, including satisfaction scores generated based on the multiple network users' use of at least one network function within a preset time period; Obtain network usage information of the multiple network users within the preset time period, the network usage information including the amount of time each network user uses at least one network function within the preset time period.

3. The method as described in claim 2, characterized in that, The network usage information includes the amount of time a network user uses at least one of the following network functions during the preset time period: Duration of online instant messaging usage, duration of online information browsing, duration of online gaming, data usage for online music playback, and duration of online video playback.

4. The method as described in claim 3, characterized in that, Cluster analysis is performed on the multiple network users based on the usage of at least one network function from their network usage information, including: The outlier values ​​of the plurality of network users are determined based on the usage of at least one of the network functions; Cluster analysis is performed on network users whose outliers are less than a preset outlier value based on the usage of at least one network function from the network usage information of the multiple network users.

5. The method as described in claim 1, characterized in that, Based on the experience factors corresponding to the network service types to which the multiple network users belong, the network score weights of the multiple network users are determined, including: The contribution of the multiple experience factors to the network services is determined based on the correlation between the experience factors and the network services. The network score weights of the multiple network users are determined based on the contribution of the multiple experience factors to network services.

6. The method according to any one of claims 1 to 5, characterized in that, After determining the network experience satisfaction evaluation result based on the network rating weights and network experience satisfaction scores of the multiple network users, the process also includes: When the network experience satisfaction evaluation result meets the preset compensation standard, compensation is performed on network users whose network experience satisfaction score is lower than the preset compensation score; and / or, When the network experience satisfaction evaluation result meets the preset satisfaction standard, network service promotion information is sent to network users whose network experience satisfaction score is higher than the preset satisfaction score.

7. An apparatus for determining network experience satisfaction, characterized in that, include: The acquisition module retrieves network experience satisfaction ratings and network usage information from multiple network users. The clustering module performs clustering analysis on the multiple network users based on their network usage information to obtain multiple groups of network users classified according to network service type, where the network service type represents the network user's network function usage preference. The first determining module determines the network score weights of the multiple network users based on the network service types to which the multiple network users belong; The second determining module determines the network experience satisfaction evaluation result based on the network rating weights and network experience satisfaction scores of the multiple network users. In the clustering module, clustering analysis is performed on the multiple network users based on their network usage information to obtain multiple groups of network users classified according to network service type, including: Based on the usage of at least one network function in the network usage information of the multiple network users, the K-Means clustering algorithm is used to perform clustering analysis on the multiple network users to obtain multiple groups of network users classified according to network service type; The determination of network score weights for the multiple network users based on their network service types includes: Principal component analysis is used to determine the experience factors corresponding to the network service types to which the multiple network users belong, and the experience factors represent the functional indicators of the corresponding network services. The network score weights of the multiple network users are determined based on the experience factors corresponding to the network service types to which the multiple network users belong.

8. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 1 to 6.

Images