CN116777043A

CN116777043A - Complaint user prediction method, complaint user prediction device and storage medium

Info

Publication number: CN116777043A
Application number: CN202310520035.7A
Authority: CN
Inventors: 张永杰; 熊金州; 范娟; 陈孟香; 何春霞; 邓闻韬; 林秋爽; 贾君凯; 沈涛
Original assignee: China United Network Communications Group Co Ltd
Current assignee: China United Network Communications Group Co Ltd
Priority date: 2023-05-09
Filing date: 2023-05-09
Publication date: 2023-09-19

Abstract

The application provides a complaint user prediction method, a complaint user prediction device and a storage medium, relates to the field of big data prediction, and can solve the problem that whether a complaint possibility exists for a user cannot be predicted in advance at present. The method comprises the following steps: determining a score for a plurality of first metrics for the target user, the first metrics comprising at least one of: network quality index, praise evaluation index, service perception index, and user type index; determining a weight value of each first index in the plurality of first indexes; and predicting the complaint index of the target user based on the scores of the first indexes and the weight value of each first index. The application can predict the complaint index of the user so as to predict whether the user has complaint possibility.

Description

Complaint user prediction method, complaint user prediction device and storage medium

Technical Field

The present application relates to the field of big data prediction, and in particular, to a method and apparatus for predicting a complaint user, and a storage medium.

Background

The rapid development of internet technology has shown an explosive growth in the demand for mobile data, and the demands of users for communication service quality have been gradually increased. When the mobile network is not good in use experience, a user can solve the problem 'complaint' through a customer hotline of an operator. However, if the user is still not satisfied with the post-complaint processing result, the user may further solve the problem "complaint" to the units such as the letter department (ministry of industry and information technology) and the communication administration (communications administration). The complaints have great influence on annual client public praise perception evaluation of operators, so that the complaint quantity of the clients is reduced, and the complaint method has great significance for the operators.

At present, a communication operator can only passively receive network-moving complaints of users, and can not predict whether the users have complaints or not in advance, so that the complaint quantity can not be effectively reduced.

Disclosure of Invention

The application provides a complaint user prediction method, a complaint user prediction device and a storage medium, which solve the problem that whether a user has a complaint possibility or not cannot be predicted in advance at present, and can predict a complaint index of the user so as to predict whether the user has the complaint possibility or not.

In order to achieve the above purpose, the application adopts the following technical scheme:

in a first aspect, the present application provides a complaint user prediction method comprising: determining a score for a plurality of first metrics for the target user, the first metrics comprising at least one of: network quality index, praise evaluation index, service perception index, and user type index; wherein the network quality index is used for representing the network quality of the network used by the user; the public praise evaluation index is used for representing the satisfaction degree of the user on the network; the service perception index is used for representing service perception of a user on a service; the user type index is used for representing the service type used by the user; determining a weight value of each first index in the plurality of first indexes; and predicting the complaint index of the target user based on the scores of the first indexes and the weight value of each first index.

With reference to the first aspect, in one possible implementation manner, the method further includes: determining a sample value of at least one second index corresponding to each first index; a score for each first indicator is determined based on the sample value of at least one second indicator corresponding to each first indicator.

With reference to the first aspect, in one possible implementation manner, the plurality of first indexes include network quality indexes; the at least one second index corresponding to the network quality index comprises: fault alert indicator, coverage level indicator, network load indicator, and signal to noise ratio indicator; the method further comprises the steps of: determining a fault alarm duration value, a coverage level intensity value, a network load value and a signal to noise value; determining a first weight value of a fault alarm duration value, a second weight value of a coverage level intensity value, a third weight value of a network load value and a fourth weight value of a signal to noise value; and based on the first weight value, the second weight value, the third weight value and the fourth weight value, the fault alarm duration value, the coverage level intensity value, the network load value and the signal to noise ratio value are weighted and summed to determine the score of the network quality index.

With reference to the first aspect, in one possible implementation manner, the plurality of first indexes include a public praise evaluation index; the at least one second index corresponding to the public praise evaluation index comprises: complaint frequency index, net recommendation value NPS scoring index and satisfaction scoring index; the method further comprises the steps of: determining complaint times, NPS scores and satisfaction scores; determining a fifth weight value of complaint times, a sixth weight value of complaint times, a seventh weight value of NPS scores and an eighth weight value of satisfaction scores; and based on the fifth weight value, the sixth weight value, the seventh weight value and the eighth weight value, the complaint times, the NPS scores and the satisfaction scores are weighted and summed to determine the scores of the public praise evaluation indexes.

With reference to the first aspect, in one possible implementation manner, the plurality of first indexes include service awareness indexes; the at least one second index corresponding to the service awareness index comprises: a downlink rate index, an uplink rate index, a call drop frequency index, a game click-through rate index and a video click-through rate index; the method further comprises the steps of: determining downlink speed, uplink speed, call drop times, game jamming rate and video jamming rate; determining a ninth weight value of a downlink rate, a tenth weight value of an uplink rate, an eleventh weight value of call drop times, a twelfth weight value of a game click-through rate and a thirteenth weight value of a video click-through rate; and based on the ninth weight value, the tenth weight value, the eleventh weight value, the twelfth weight value and the thirteenth weight value, the downlink speed, the uplink speed, the call drop times, the game click-through rate and the video click-through rate are weighted and summed to determine the score of the service perception index.

With reference to the first aspect, in a possible implementation manner, the plurality of first indexes include a user type index; the at least one second index corresponding to the user type index comprises: big packet download service index, video service index, game service index, voice service index, webpage service index and WeChat service index; the method further comprises the steps of: determining a large packet downloading flow rate, a video use flow rate, a game use flow rate, a voice use flow rate, a web page use flow rate and a micro messenger use flow rate; determining a fourteenth weight value of a large packet downloading flow rate, a fifteenth weight value of a video use flow rate, a sixteenth weight value of a game use flow rate, a seventeenth weight value of a voice use flow rate, an eighteenth weight value of a web page use flow rate and a nineteenth weight value of a micro messenger use flow rate; and determining a score of the user type index based on a weighted sum of the fourteenth weight value, the fifteenth weight value, the sixteenth weight value, the seventeenth weight value, the eighteenth weight value and the nineteenth weight value for the large packet download traffic ratio, the video use traffic ratio, the game use traffic ratio, the voice use traffic ratio, the web page use traffic ratio and the micro messenger use traffic ratio.

In a second aspect, the present application provides a complaint user prediction apparatus comprising: a processing unit; a processing unit for determining scores of a plurality of first indicators of the target user, the first indicators comprising at least one of: network quality index, praise evaluation index, service perception index, and user type index; wherein the network quality index is used for representing the network quality of the network used by the user; the public praise evaluation index is used for representing the satisfaction degree of the user on the network; the service perception index is used for representing service perception of a user on a service; the user type index is used for representing the service type used by the user; the processing unit is further used for determining a weight value of each first index in the plurality of first indexes; the processing unit is further used for predicting the complaint index of the target user based on the scores of the first indexes and the weight value of each first index.

With reference to the second aspect, in a possible implementation manner, the apparatus further includes: a communication unit; a communication unit, configured to determine a sample value of at least one second index corresponding to each first index; the processing unit is specifically configured to determine a score of each first indicator based on a sample value of at least one second indicator corresponding to each first indicator.

With reference to the second aspect, in one possible implementation manner, the plurality of first indexes include network quality indexes; the at least one second index corresponding to the network quality index comprises: fault alert indicator, coverage level indicator, network load indicator, and signal to noise ratio indicator; the communication unit is specifically used for determining a fault alarm duration value, a coverage level intensity value, a network load value and a signal to noise value; the processing unit is specifically used for determining a first weight value of a fault alarm duration value, a second weight value of a coverage level intensity value, a third weight value of a network load value and a fourth weight value of a signal to noise value; the processing unit is specifically configured to determine a score of the network quality indicator by weighted summation of the fault alarm duration value, the coverage level intensity value, the network load value and the signal-to-noise ratio value based on the first weight value, the second weight value, the third weight value and the fourth weight value.

With reference to the second aspect, in one possible implementation manner, the plurality of first indexes include a public praise evaluation index; the at least one second index corresponding to the public praise evaluation index comprises: complaint frequency index, net recommendation value NPS scoring index and satisfaction scoring index; the communication unit is specifically used for determining complaint times, NPS scores and satisfaction scores; the processing unit is specifically used for determining a fifth weight value of the complaint times, a sixth weight value of the complaint times, a seventh weight value of the NPS score and an eighth weight value of the satisfaction score; the processing unit is specifically configured to determine a score of the public praise evaluation index based on the fifth weight value, the sixth weight value, the seventh weight value and the eighth weight value by weighted summation of the complaint times, the NPS scores and the satisfaction scores.

With reference to the second aspect, in one possible implementation manner, the plurality of first indexes include service awareness indexes; the at least one second index corresponding to the service awareness index comprises: a downlink rate index, an uplink rate index, a call drop frequency index, a game click-through rate index and a video click-through rate index; the communication unit is specifically used for determining the downlink rate, the uplink rate, the call drop times, the game jamming rate and the video jamming rate; the processing unit is specifically configured to determine a ninth weight value of the downlink rate, a tenth weight value of the uplink rate, an eleventh weight value of the call drop number, a twelfth weight value of the game click-through rate, and a thirteenth weight value of the video click-through rate; the processing unit is specifically configured to determine a score of the service perception indicator by performing weighted summation on the downlink rate, the uplink rate, the call drop number, the game click-through rate and the video click-through rate based on the ninth weight value, the tenth weight value, the eleventh weight value, the twelfth weight value and the thirteenth weight value.

With reference to the second aspect, in one possible implementation manner, the plurality of first indexes include user type indexes; the at least one second index corresponding to the user type index comprises: big packet download service index, video service index, game service index, voice service index, webpage service index and WeChat service index; the communication unit is specifically used for determining the large packet downloading flow rate, the video use flow rate, the game use flow rate, the voice use flow rate, the web page use flow rate and the micro messenger use flow rate; the processing unit is specifically configured to determine a fourteenth weight value of a large packet downloading traffic ratio, a fifteenth weight value of a video usage traffic ratio, a sixteenth weight value of a game usage traffic ratio, a seventeenth weight value of a voice usage traffic ratio, an eighteenth weight value of a web page usage traffic ratio, and a nineteenth weight value of a micro messenger usage traffic ratio; the processing unit is specifically configured to determine a score of the user type indicator by weighted summation of the packet download traffic ratio, the video use traffic ratio, the game use traffic ratio, the voice use traffic ratio, the web page use traffic ratio, and the micro messenger use traffic ratio based on a fourteenth weight value, a fifteenth weight value, a sixteenth weight value, a seventeenth weight value, an eighteenth weight value, and a nineteenth weight value.

In a third aspect, the present application provides a complaint user prediction apparatus comprising: a processor and a communication interface; the communication interface is coupled to a processor for running a computer program or instructions to implement the complaint user prediction method as described in any one of the possible implementations of the first aspect and the first aspect.

In a fourth aspect, the present application provides a computer readable storage medium having instructions stored therein which, when run on a terminal, cause the terminal to perform a complaint user prediction method as described in any one of the possible implementations of the first aspect and the first aspect.

In a fifth aspect, the present application provides a computer program product comprising instructions which, when run on a complaint user prediction device, cause the complaint user prediction device to perform the complaint user prediction method as described in any one of the possible implementations of the first aspect and the first aspect.

In a sixth aspect, the present application provides a chip comprising a processor and a communications interface, the communications interface and the processor being coupled, the processor being for running a computer program or instructions to implement the complaint user prediction method as described in any one of the possible implementations of the first aspect and the first aspect.

In particular, the chip provided in the present application further includes a memory for storing a computer program or instructions.

It should be noted that the above-mentioned computer instructions may be stored in whole or in part on a computer-readable storage medium. The computer readable storage medium may be packaged together with the processor of the apparatus or may be packaged separately from the processor of the apparatus, which is not limited in this respect.

In a seventh aspect, the present application provides a complaint user prediction system comprising: a complaint user prediction apparatus and a data server, wherein the complaint user prediction apparatus is for performing a complaint user prediction method as described in any one of the possible implementations of the first aspect and the first aspect.

The description of the second to seventh aspects of the present application may refer to the detailed description of the first aspect; also, the advantageous effects described in the second aspect to the seventh aspect may refer to the advantageous effect analysis of the first aspect, and are not described herein.

In the present application, the names of the above-described complaint user predicting means do not constitute limitations on the devices or function modules themselves, but in actual implementation, these devices or function modules may appear under other names. Insofar as the function of each device or function module is similar to that of the present application, it falls within the scope of the claims of the present application and the equivalents thereof.

These and other aspects of the application will be more readily apparent from the following description.

The scheme at least brings the following beneficial effects: based on the technical scheme, the complaint user prediction method provided by the application has the advantages that the complaint prediction device determines the network quality index, the public praise evaluation index, the service perception index and the grading of the user type index of the target user. The network quality index is used for representing the network quality of the network used by the user, the public praise evaluation index is used for representing the satisfaction degree of the user on the network, the service perception index is used for representing the service perception of the user on the service, the user type index is used for representing the service type used by the user, and the complaint prediction device fully combines subjective factors and objective factors and analyzes and obtains the index scores of the subjective and objective aspects of the target user. And the complaint predicting device predicts the complaint index of the target user based on the scores of the network quality index, the public praise evaluation index, the business perception index and the user type index, and the weight values of the indexes. Compared with the prior art that the communication carrier cannot predict whether the user has the possibility of complaining in advance, the technical scheme can accurately obtain the complaint index of the user, so that whether the user has the possibility of complaining is predicted.

Drawings

FIG. 1a is a schematic flow chart of a complaint handling method according to an embodiment of the present application;

FIG. 1b is a schematic flow chart of another complaint handling method according to an embodiment of the present application;

FIG. 2 is a schematic diagram of a system for predicting a complaint user according to an embodiment of the present application;

FIG. 3 is a schematic diagram of a device for predicting a complaint user according to an embodiment of the present application;

FIG. 4 is a schematic diagram of a hardware structure of a complaint user prediction device according to an embodiment of the present application;

FIG. 5 is a flowchart of a method for complaint user prediction according to an embodiment of the present application;

FIG. 6 is a flowchart of a complaint index determining method according to an embodiment of the present application;

FIG. 7 is a flowchart of another method for complaint user prediction provided by an embodiment of the present application;

FIG. 8 is a flowchart of another complaint user prediction method provided by an embodiment of the present application;

FIG. 9 is a flowchart of another complaint user prediction method provided by an embodiment of the present application;

FIG. 10 is a flowchart of another method for complaint user prediction provided by an embodiment of the present application;

FIG. 11 is a flowchart of another method for complaint user prediction provided by an embodiment of the present application;

FIG. 12 is a flowchart of another complaint index determining method according to an embodiment of the present application;

Fig. 13 is a schematic structural diagram of another complaint user prediction device according to an embodiment of the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

The term "and/or" is herein merely an association relationship describing an associated object, meaning that there may be three relationships, e.g., a and/or B, may represent: a exists alone, A and B exist together, and B exists alone.

The terms "first" and "second" and the like in the description and in the drawings are used for distinguishing between different objects or between different processes of the same object and not for describing a particular order of objects.

Furthermore, references to the terms "comprising" and "having" and any variations thereof in the description of the present application are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed but may optionally include other steps or elements not listed or inherent to such process, method, article, or apparatus.

It should be noted that, in the embodiments of the present application, words such as "exemplary" or "such as" are used to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" or "e.g." in an embodiment should not be taken as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "such as" is intended to present related concepts in a concrete fashion.

In the description of the present application, unless otherwise indicated, the meaning of "a plurality" means two or more.

Internet technology is silently changing the life style of society, and mobile phone terminals are derivatives of internet technology. In recent years, under the background of dual-growth of mobile network users and traffic, the demand of mobile data is explosively increased, new services and application layers of mobile phone terminals are endless, and the requirements of communication operators are also developed from the original speed-up and cost-reduction to network perception dual-guarantee. With the increasing high demands of the country and society on operators and the increasing degree of sensitivity and awareness of the rights of users to network awareness, as shown in fig. 1a, when the experience of users using the mobile network is poor, the problem of "complaint" is solved by the client hotline of operators. However, if the user is still not satisfied with the post-complaint processing result, the user can further solve the problem of complaint to the work department, the communication management office and other units, and the operator receives the complaint list and then carries out follow-up processing through the network department. Compared with hot line complaints, the complaints are from supervision of a superior management unit, have higher timeliness requirements and resolution effects, and have great influence on annual client public praise perception evaluation of operators, so that the client complaint amount is reduced, and the method has great significance for the operators.

However, in statistics of the complaint amount, as long as the user has complained about the behavior, that is, the statistics number, the subsequent solution is not considered, and the behavior of the user is more uncertain. At present, a communication operator can only passively receive network-moving complaints of users, and can not predict whether the users have complaints or not in advance, so that the complaint quantity can not be effectively reduced.

Based on the above, the application provides a complaint user prediction method, as shown in fig. 1b, by outputting a list of potential complaint users, the early warning of the potential complaint users is realized. The method is convenient for the follow-up operators to intervene in the network in advance to solve the requirements of potential complaint users, so that the users are prevented from complaining to the letter department, the communication management office and the like.

At present, the communication operator carries out internal flow upgrading processing aiming at individual cases of complaint tendency mentioned by the user so as to avoid the follow-up complaint of the user. However, even if some users do not mention the tendency of complaints in the complaint process, there is a possibility that complaints will be generated later, and therefore, there is a case that the complaint population is monitored for a lack.

The current complaint user prediction analysis method cannot realize the accurate output of batch potential complaint user groups, and consumes human resources.

In view of this, the complaint user prediction method provided by the present application, the complaint prediction apparatus determines the scores of the network quality index, the public praise evaluation index, the business perception index, and the user type index of the target user. The network quality index is used for representing the network quality of the network used by the user, the public praise evaluation index is used for representing the satisfaction degree of the user on the network, the service perception index is used for representing the service perception of the user on the service, the user type index is used for representing the service type used by the user, and the complaint prediction device fully combines subjective factors and objective factors and analyzes and obtains the index scores of the subjective and objective aspects of the target user. And the complaint predicting device predicts the complaint index of the target user based on the scores of the network quality index, the public praise evaluation index, the business perception index and the user type index, and the weight values of the indexes. Compared with the prior art that the communication carrier cannot predict whether the user has the possibility of complaining in advance, the technical scheme can accurately obtain the complaint index of the user, so that whether the user has the possibility of complaining is predicted.

The following describes embodiments of the present application in detail with reference to the drawings.

Fig. 2 is a block diagram of a complaint user prediction system 20 according to an embodiment of the present application. As shown in fig. 2, the complaint user prediction system 20 includes: complaint user prediction apparatus 201 and data server 202.

The number of the complaint user prediction device 201 and the data server 202 may be one or plural, and only one is shown in fig. 2 for the sake of understanding.

The complaint user prediction device 201 and the data server 202 are connected by a communication link. The communication link may be a wired communication link or a wireless communication link, which is not limited in this regard by the present application.

In one possible implementation, the complaint user predicting means 201 receives a plurality of sample values from the network quality side, a plurality of sample values from the public praise evaluation side, a plurality of sample values from the service awareness side, and a plurality of sample values from the user type side of the target user in the data server 202. The complaint user prediction device 201 determines a score of the network quality index according to the plurality of sample values of the network quality end and the weight of the network quality end; determining scores of the public praise evaluation indexes according to a plurality of sample values of the public praise evaluation end and weights of the public praise evaluation end; determining the score of the index of the service perception end according to a plurality of sample values of the service perception end and the weight of the service perception end; and determining the score of the user type index according to the plurality of sample values of the user type end and the weight of the user type end. Then, the complaint user predicting device 201 determines the sum of the score of the network quality index, the score of the public praise evaluation index, the score of the service perception end index, and the score of the user type index as the complaint index of the target user.

In another possible implementation, as shown in fig. 3, the complaint user prediction apparatus 201 includes a data acquisition unit 2011, a complaint early warning unit 2012, and a problem location processing unit 2013. The data collection unit 2011 obtains a plurality of sample values of the network quality end, a plurality of sample values of the public praise evaluation end, a plurality of sample values of the service perception end and a plurality of sample values of the user type end of the target user from the data server 202. The complaint early warning unit 2012 distributes the training samples and the test samples 7:3 to the sample values through a machine learning (light gradient boosting machine, lightGBM) algorithm model, and processes sparse data by a gradient-based one-side sampling (GOSS) algorithm and a mutual exclusion feature binding (exclusive feature bundling, EFB) based decision tree algorithm based on a histogram (histogram), so as to realize high-precision and large-scale data processing. The problem positioning processing unit 2013 outputs a user list with possibility of complaints, combines user core network signaling and resident cell information, positions user priorities based on the network list and an algorithm, realizes user network problem positioning, improves network quality by planning stand and newly creating open sites aiming at resource problems, and achieves the purposes of improving user perception, accurately pre-controlling from the source and solving the generation of potential complaints of users.

In one possible implementation, the data server 202 is configured to send the sample value of the network quality end to the complaint user prediction device 201. Optionally, the sample values at the network quality end include, but are not limited to, at least one of: fault alarm duration value, coverage level intensity value, network load value and signal to noise value, and sample value of the public praise evaluation end: complaint times, NPS scores and satisfaction scores, sample values of business perception ends: downlink speed, uplink speed, call drop times, game jamming rate and video jamming rate; sample value at user type end: the large packet download traffic ratio, the video usage traffic ratio, the game usage traffic ratio, the voice usage traffic ratio, the web page usage traffic ratio, and the micro messenger usage traffic ratio.

When implemented in hardware, the individual modules in the complaint user prediction apparatus 201 may be integrated into the hardware configuration of the complaint user prediction apparatus as shown in fig. 4. Specifically, as shown in fig. 4, a basic hardware configuration of the complaint user prediction apparatus is introduced.

Fig. 4 is a schematic structural diagram of a complaint user prediction device according to an embodiment of the present application. As shown in fig. 4, the complaint user prediction device includes at least one processor 401, a communication line 402, and at least one communication interface 404, and may further include a memory 403. The processor 401, the memory 403, and the communication interface 404 may be connected by a communication line 402.

The processor 401 may be a central processing unit (central processing unit, CPU), an application specific integrated circuit (application specific integrated circuit, ASIC), or one or more integrated circuits configured to implement embodiments of the present application, such as: one or more digital signal processors (digital signal processor, DSP), or one or more field programmable gate arrays (field programmable gate array, FPGA).

Communication line 402 may include a path for communicating information between the components described above.

The communication interface 404, for communicating with other devices or communication networks, may use any transceiver-like device, such as ethernet, radio access network (radio access network, RAN), wireless local area network (wireless local area networks, WLAN), etc.

The memory 403 may be, but is not limited to, a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (random access memory, RAM) or other type of dynamic storage device that can store information and instructions, or an electrically erasable programmable read-only memory (electrically erasable programmable read-only memory, EEPROM), a compact disc (compact disc read-only memory) or other optical disc storage, optical disc storage (including compact disc, laser disc, optical disc, digital versatile disc, blu-ray disc, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to contain or store the desired program code in the form of instructions or data structures and that can be accessed by a computer.

In a possible design, the memory 403 may exist independent of the processor 401, i.e. the memory 403 may be a memory external to the processor 401, where the memory 403 may be connected to the processor 401 by a communication line 402, for storing execution instructions or application codes, and the execution is controlled by the processor 401, to implement the complaint user prediction method provided by the following embodiment of the present application. In yet another possible design, the memory 403 may be integrated with the processor 401, i.e., the memory 403 may be an internal memory of the processor 401, e.g., the memory 403 may be a cache, and may be used to temporarily store some data and instruction information, etc.

As one possible implementation, processor 401 may include one or more CPUs, such as CPU0 and CPU1 in fig. 4. As another possible implementation, the complaint user prediction means may include a plurality of processors, such as the processor 401 and the processor 407 in fig. 4. As yet another possible implementation, the complaint user prediction means may further comprise an output device 405 and an input device 406.

It should be noted that, the embodiments of the present application may refer to or refer to each other, for example, the same or similar steps, and the method embodiment, the system embodiment and the device embodiment may refer to each other, which is not limited.

Fig. 5 is a flowchart of a method for predicting a complaint user according to an embodiment of the present application, which can be applied to the apparatus for predicting a complaint user shown in fig. 4. As shown in fig. 5, the method includes the following: S501-S503.

S501, complaint user prediction means determines scores of a plurality of first indexes of a target user.

Wherein the first indicator comprises at least one of: network quality metrics, praise evaluation metrics, traffic awareness metrics, and user type metrics. The network quality indicator is used to characterize the network quality of the network used by the user. The public praise rating index is used to characterize the satisfaction of the user with the network. The service awareness index is used for representing service awareness of the user on the service. The user type index is used to characterize the type of service used by the user.

In one possible implementation, as shown in fig. 6, the complaint user prediction apparatus determines a complaint index of a target user through a network quality index, a public praise evaluation index, a traffic perception index, a user type index, and weights corresponding to the respective indexes.

It should be noted that, in fig. 6, specific parameters for determining the complaint index of the target user are only described as an example, and in practical application, the complaint user prediction device may also determine the complaint index of the target user through other parameters, which is not limited in the present application.

S502, complaint user prediction device determines the weight value of each first index in the plurality of first indexes.

In one possible implementation manner, the complaint user prediction device determines the scales between the first indexes by an expert scoring method, and determines the weight value of each first index respectively.

Illustratively, as shown in table 1, the complaint user prediction apparatus determines that the importance degree of the network quality with respect to the network quality is 1 by the expert scoring method; determining that the public praise evaluation is 1/3 of the importance degree relative to the network quality; determining the importance degree of service perception relative to the network quality as 1/5; the importance of the user type with respect to the network quality is determined to be 1/7. The first weight of the complaint user prediction means to determine the network quality is 1/(1+1/3+1/5+1/7). By analogy, determining the second weight of the network quality as 3/(3+1+1/3+1/5) respectively; the third weight of the network quality is 5/(5+3+1+1/3); the fourth weight of the network mass is 7/(7+5+3+1). Then, the complaint user prediction device determines that the average weight of the network quality is 0.5579 by an arithmetic average method.

TABLE 1 first index expert scoring Table

Expert scoring	Network quality	Public praise evaluation	Service awareness	User type
					Network quality	1	3	5	7
Public praise evaluation	1/3	1	3	5
					Service awareness	1/5	1/3	1	3
User type	1/7	1/5	1/3	1

It should be noted that, the manner of determining the weight in the present application is not limited to one type. The weight calculation may be performed by any one of a factor analysis method, a principal component analysis method, an Analytic Hierarchy Process (AHP), an order graph method, an entropy value method (entropy weight method), an objective weighting method, an independent weighting method, and an information amount weighting method.

S503, the complaint user prediction device predicts the complaint index of the target user based on the scores of the plurality of first indexes and the weight value of each first index.

In one possible implementation, the complaint user prediction device determines a complaint index of the target user based on the score of each first index and the weight of each first index by weighted summation of the scores of the plurality of first indexes.

Illustratively, as shown in fig. 6, the complaint user prediction apparatus determines a product of the network quality index score and the corresponding weight, a product of the public place name evaluation index score and the corresponding weight, and a product of the service perception index score and the corresponding weight, and a product of the user type index score and the corresponding weight based on the network quality index score, the public place name evaluation index score, the service perception index score, the user type index score, and the weight of each index. The complaint index of the target user is determined as the sum of a plurality of products by the complaint user prediction means.

In another possible implementation manner, the complaint user prediction device performs algorithm combination by combining a K-means clustering algorithm (K-means clustering algorithm, K-means), a neural network algorithm or a decision tree algorithm, and optimizes the weight value of the first index to obtain a plurality of target weight values. Based on the first indexes and the target weight values, the complaint user prediction device obtains a complaint index of the target user through a linear regression algorithm.

Based on the technical scheme, the complaint user prediction method provided by the application has the advantages that the complaint prediction device determines the network quality index, the public praise evaluation index, the service perception index and the grading of the user type index of the target user. The network quality index is used for representing the network quality of the network used by the user, the public praise evaluation index is used for representing the satisfaction degree of the user on the network, the service perception index is used for representing the service perception of the user on the service, the user type index is used for representing the service type used by the user, and the complaint prediction device fully combines subjective factors and objective factors and analyzes and obtains the index scores of the subjective and objective aspects of the target user. And the complaint predicting device predicts the complaint index of the target user based on the scores of the network quality index, the public praise evaluation index, the business perception index and the user type index, and the weight values of the indexes. Compared with the prior art that the communication carrier cannot predict whether the user has the possibility of complaining in advance, the technical scheme can accurately obtain the complaint index of the user, so that whether the user has the possibility of complaining is predicted.

As a possible embodiment of the present application, as shown in fig. 7 in conjunction with fig. 5, the above-mentioned process of determining scores of a plurality of first indexes of the target user in S501 may also be implemented by the following S701-S702.

S701, complaining user prediction device determines sample value of at least one second index corresponding to each first index.

One possible implementation manner, the complaint user prediction device obtains a sample value of at least one second index corresponding to the network quality index, wherein the sample value comprises a fault alarm, a coverage level, a network load and a signal to noise ratio; obtaining sample values of at least one second index corresponding to the public praise evaluation index, wherein the sample values comprise complaint times, NPS scores and satisfaction scores; determining that the sample value of at least one second index corresponding to the service perception index comprises an uplink rate, a downlink rate, call drop times, a game click-through rate and a video click-through rate; the sample value of at least one second index corresponding to the user type index is determined to comprise a large packet downloading flow rate duty ratio, a video use flow rate duty ratio, a game use flow rate duty ratio, a voice use flow rate duty ratio, a webpage use flow rate duty ratio and a micro messenger use flow rate duty ratio.

S702, complaint user prediction means determines the score of each first index based on the sample value of at least one second index corresponding to each first index.

One possible implementation complaints that the user prediction device determines a network quality indicator score based on fault alert, coverage level, network load, and signal to noise ratio; determining a public praise rating index score based on the number of complaints, the NPS score, and the satisfaction score; determining a service perception index score based on the uplink rate, the downlink rate, the call drop times, the game click-through rate and the video click-through rate; the user type indicator score is determined based on the big packet download traffic duty cycle, the video usage traffic duty cycle, the game usage traffic duty cycle, the voice usage traffic duty cycle, the web page usage traffic duty cycle, and the micro messenger usage traffic duty cycle.

Based on the technical scheme, the complaint user prediction device firstly determines the sample value of at least one second index corresponding to each first index through the technical scheme, and determines the score of each first index based on the sample value of at least one second index corresponding to each first index, so that the complaint index of the user can be determined through the scores of a plurality of first indexes.

As a possible embodiment of the present application, when the plurality of first indexes includes network quality indexes, as shown in fig. 8, at least one second index corresponding to the network quality indexes includes: fault alert indicator, coverage level indicator, network load indicator, and signal to noise ratio indicator. The process of determining the score of the first index in S702 described above may also be implemented by the following S801 to S803.

S801, complaint user prediction device determines a fault alarm duration value, a coverage level intensity value, a network load value and a signal to noise value.

S802, complaining of the user prediction device to determine a first weight value of a fault alarm duration value, a second weight value of a coverage level intensity value, a third weight value of a network load value and a fourth weight value of a signal to noise value.

In one possible implementation manner, the complaint user prediction device determines the scales of the fault alarm index, the coverage level index, the network load index and the signal to noise ratio index by using an expert scoring method, and determines the weight values of the fault alarm index, the coverage level index, the network load index and the signal to noise ratio index respectively.

Illustratively, as shown in table 2, the complaint user prediction apparatus determines that the importance degree of the fault alarm with respect to the fault alarm is 1 by expert scoring; determining that the importance degree of the public praise evaluation relative to the fault alarm is 1/2; determining the importance degree of service perception relative to fault alarm as 1/5; the importance degree of the user type relative to the fault alarm is determined to be 1/7. The complaint user prediction means determines that the first weight of the fault alert is 1/(1+1/2+1/5+1/7). And by analogy, determining that the second weight of the fault alarm is 2/(2+1+1/3+1/5) respectively; the third weight of the fault alarm is 5/(5+3+1+1/4); the fourth weight of the fault alarm is 7/(7+5+4+1). Then, the complaint user prediction means determines the average weight of the fault alarm to be 0.5192 by an arithmetic average method.

Table 2 network quality index expert scoring table

S803, the complaint user prediction device performs weighted summation on the fault alarm duration value, the coverage level intensity value, the network load value and the signal to noise ratio value based on the first weight value, the second weight value, the third weight value and the fourth weight value to determine the score of the network quality index.

Illustratively, the complaint user predicting means determines the product of the fault alert duration value and the corresponding weight, the product of the coverage level intensity value and the corresponding weight, and the product of the network load value and the corresponding weight, based on the fault alert duration value, the coverage level intensity value, the network load value, and the signal-to-noise value, and the weight of each index. The complaint user prediction means determines the sum of the products as a score of the network quality indicator.

Based on the technical scheme, the complaint user prediction device determines a fault alarm time length value, a coverage level intensity value, a network load value and a signal to noise value, and determines a first weight value of the fault alarm time length value, a second weight value of the coverage level intensity value, a third weight value of the network load value and a fourth weight value of the signal to noise value. And then the complaint user prediction device performs weighted summation on the fault alarm duration value, the coverage level intensity value, the network load value and the signal to noise ratio value based on the first weight value, the second weight value, the third weight value and the fourth weight value to determine the score of the network quality index. The complaint user prediction device obtains the scores of the network quality indexes through the technical scheme, so as to judge the influence condition of the network quality on the complaint of the user.

As a possible embodiment of the present application, when the plurality of first indexes includes a pragmatic evaluation index, as shown in fig. 9 in combination with fig. 7, at least one second index corresponding to the pragmatic evaluation index includes: complaint frequency index, net recommendation value NPS scoring index and satisfaction scoring index. The process of determining the score of the first index in S702 described above may also be implemented by the following S901 to S903.

S901, the complaint user prediction device determines complaint times, NPS scores and satisfaction scores.

S902, the complaint user prediction device determines a fifth weight value of the complaint times, a sixth weight value of the complaint times, a seventh weight value of the NPS score and an eighth weight value of the satisfaction score.

In one possible implementation manner, the complaint user prediction device determines the complaint frequency index, the net recommended value NPS scoring index and the satisfaction scoring index by using an expert scoring method, and determines the weight values of the complaint frequency index, the net recommended value NPS scoring index and the satisfaction scoring index respectively.

Illustratively, as shown in table 3, the complaint user predicting apparatus determines that the degree of importance of the number of complaints with respect to the number of complaints is 1 by expert scoring; determining the importance degree of complaint times relative to complaint times to be 1/4; determining that the importance degree of the NPS score of the net recommended value relative to the complaint times is 1/5; the degree of importance of the satisfaction score with respect to the number of complaints was determined to be 1/7. The first weight of the complaint number of times determined by the complaint user predicting means is 1/(1+1/4+1/5+1/7). And by analogy, determining that the second weight of the complaint times is 4/(4+1+1/3+1/5) respectively; the third weight of the complaint times is 5/(5+3+1+1/3); the fourth weight of the complaint times is 7/(7+5+3+1). Then, the complaint user prediction device determines the average weight of the complaint times as 0.5954 by the arithmetic average method.

Table 3 public praise evaluation index expert scoring table

Expert scoring	Number of complaints	Number of complaints	NPS scoring	Satisfaction scoring
					Number of complaints	1	4	5	7
Number of complaints	1/4	1	3	5
					NPS scoring	1/5	1/3	1	3
Satisfaction scoring	1/7	1/5	1/3	1

S903, the complaint user prediction device determines the score of the public praise evaluation index by weighted summation of the complaint times, the NPS score and the satisfaction degree score based on the fifth weight value, the sixth weight value, the seventh weight value and the eighth weight value.

Illustratively, the complaint user predicting means determines a product of the number of complaints and the corresponding weight, and a product of the NPS score and the corresponding weight, based on the number of complaints, the NPS score, and the satisfaction score, and the weight of each index. The complaint user prediction means determines the sum of the products as the score of the public praise rating index.

Based on the technical scheme, the complaint user prediction device determines the complaint times, the NPS scores and the satisfaction scores, and determines a fifth weight value of the complaint times, a sixth weight value of the complaint times, a seventh weight value of the NPS scores and an eighth weight value of the satisfaction scores. The complaint user prediction means then performs weighted summation of the complaint times, NPS scores, and satisfaction scores based on the fifth weight value, the sixth weight value, the seventh weight value, and the eighth weight value, and determines scores of the public praise evaluation index. The complaint user prediction device obtains the scores of the public praise evaluation indexes through the technical scheme, so as to judge the influence condition of the public praise evaluation on the complaint of the user.

As a possible embodiment of the present application, when the plurality of first indexes includes a traffic perception index, as shown in fig. 10, at least one second index corresponding to the traffic perception index includes: a downlink rate index, an uplink rate index, a call drop frequency index, a game click-through rate index and a video click-through rate index. The process of determining the score of the first index in S702 described above may also be implemented by the following S1001 to S1003.

S1001, the complaint user prediction device determines a downlink rate, an uplink rate, call drop times, a game click-through rate and a video click-through rate.

S1002, the complaint user prediction device determines a ninth weight value of the downlink rate, a tenth weight value of the uplink rate, an eleventh weight value of the call drop number, a twelfth weight value of the game click-through rate and a thirteenth weight value of the video click-through rate.

In one possible implementation manner, the complaint user prediction device determines the downlink rate index, the uplink rate index, the call drop frequency index, the game click-through rate index and the video click-through rate index by using an expert scoring method, and determines the weight values of the downlink rate index, the uplink rate index, the call drop frequency index, the game click-through rate index and the video click-through rate index respectively.

Illustratively, as shown in table 4, the complaint user prediction apparatus determines that the importance of the downstream rate with respect to the downstream rate is 1 by the expert scoring method; determining the importance degree of the uplink rate relative to the downlink rate to be 1/2; determining the importance degree of the call drop times relative to the downlink speed to be 1/4; determining the importance degree of the game jamming rate relative to the downlink rate to be 1/7; the importance degree of the video jamming rate relative to the downlink rate is determined to be 1/8. The complaint user prediction means determines that the first weight of the downstream rate is 1 +.

(1+1/2+1/4+1/7+1/8). And so on, respectively determining that the second weight of the uplink rate is 2 +.

(2+1+1/3+1/5+1/6); the third weight of the number of dropped calls is 4/(4+3+1+1/3+1/4); the fourth weight of the game click-through rate is 7/(7+5+3+1+1/2); the fifth weight of video clip rate is 8 +.

(8+6+4+2+1). Then, the complaint user prediction device determines the average weight of the downstream rate as 0.4642 by the arithmetic average method.

TABLE 4 Business awareness index expert scoring List

S1003, the complaint user prediction device performs weighted summation on the downlink rate, the uplink rate, the call drop times, the game jamming rate and the video jamming rate based on the ninth weight value, the tenth weight value, the eleventh weight value, the twelfth weight value and the thirteenth weight value, and determines the score of the service perception index.

Illustratively, the complaint user prediction apparatus determines a product of the downlink rate and the corresponding weight, a product of the uplink rate and the corresponding weight, a product of the number of times of call drops and the corresponding weight, a product of the game click-through rate and the corresponding weight, and a product of the video click-through rate and the corresponding weight based on the downlink rate, the uplink rate, the number of times of call drops, the game click-through rate, and the video click-through rate, and the weight of each index. The complaint user prediction means determines the sum of the products as the score of the traffic perception index.

Based on the technical scheme, the complaint user prediction device determines the downlink rate, the uplink rate, the call drop times, the game click-through rate and the video click-through rate, and determines a ninth weight value of the downlink rate, a tenth weight value of the uplink rate, an eleventh weight value of the call drop times, a twelfth weight value of the game click-through rate and a thirteenth weight value of the video click-through rate. And then the complaint user prediction device performs weighted summation on the downlink rate, the uplink rate, the call drop times, the game click-through rate and the video click-through rate based on the ninth weight value, the tenth weight value, the eleventh weight value, the twelfth weight value and the thirteenth weight value, and determines the score of the service perception index. The complaint user prediction device obtains the score of the service perception index through the technical scheme, so as to judge the influence condition of the service perception on the complaint of the user.

As a possible embodiment of the present application, when the plurality of first indexes includes a user type index, as shown in fig. 11, at least one second index corresponding to the user type index includes: big packet download service index, video service index, game service index, voice service index, webpage service index and WeChat service index. The process of determining the score of the first index in S702 described above may also be implemented by the following S1101 to S1103.

S1101, the complaint user prediction device determines a big packet downloading flow rate, a video use flow rate, a game use flow rate, a voice use flow rate, a web page use flow rate and a micro messenger use flow rate.

S1102, the complaint user prediction means determines a fourteenth weight value of the packet download traffic ratio, a fifteenth weight value of the video use traffic ratio, a sixteenth weight value of the game use traffic ratio, a seventeenth weight value of the voice use traffic ratio, an eighteenth weight value of the web page use traffic ratio, and a nineteenth weight value of the micro messenger use traffic ratio.

In one possible implementation manner, the complaint user prediction device determines the scales of the big packet downloading service index, the video service index, the game service index, the voice service index, the webpage service index and the WeChat service index by using an expert scoring method, and determines the weight values of the big packet downloading service index, the video service index, the game service index, the voice service index, the webpage service index and the WeChat service index respectively.

Illustratively, as shown in table 5, the complaint user prediction means determines that the importance of the big package download with respect to the big package download is 1 by expert scoring; determining the importance degree of the video relative to the downloading of the big package to be 1/2; determining the importance degree of the game relative to the download of the big package to be 1/3; determining the importance degree of the voice relative to the large packet downloading to be 1/5; determining the importance degree of the webpage relative to the download of the big package to be 1/7; the importance of WeChat with respect to big packet download is determined to be 1/8. The complaint user prediction means determines that the first weight of the big packet download is 1/(1+1/2+1/3+1/5+1/7+1/8). And so on, determining that the second weight of the big packet download is 2/(2+1+1/2+1/4+1/6+1/7) respectively; the third weight of the big package download is 3 +.

(3+2+1+1/2+1/5+1/6); the fourth weight of the big packet download is 5/(5+4+2+1+1/4+1/5); the fifth weight of the big packet download is 7/(7+6+5+4+1+1/2); the sixth weight of the big packet download is 8/(8+7+6+5+2+1). Then, the complaint user prediction device determines that the average weight of the large package download is 0.3948 by the arithmetic average method.

TABLE 5 user type index expert scoring List

Expert scoring	Bale download	Video frequency	Number of games	Speech sound	Web page	WeChat
							Bale download	1	2	3	5	7	8
Video frequency	1/2	1	2	4	6	7
							Game machine	1/3	1/2	1	2	5	6
Speech sound	1/5	1/4	1/2	1	4	5
							Web page	1/7	1/6	1/5	1/4	1	2
WeChat	1/8	1/7	1/6	1/5	1/2	1

S1103, the complaint user prediction device determines the score of the user type index based on the weighted summation of the fourteenth weight value, the fifteenth weight value, the sixteenth weight value, the seventeenth weight value, the eighteenth weight value and the nineteenth weight value on the large packet downloading flow rate, the video use flow rate, the game use flow rate, the voice use flow rate, the web page use flow rate and the micro messenger use flow rate.

Illustratively, the complaint user predicting means determines a product of the large packet download flow rate and the corresponding weight, a product of the video use flow rate and the corresponding weight, a product of the game use flow rate and the corresponding weight, a product of the voice use flow rate and the corresponding weight, a product of the web page use flow rate and the corresponding weight, and a product of the micro messenger use flow rate and the corresponding weight based on the large packet download flow rate, the video use flow rate, the game use flow rate, the voice use flow rate, the web page use flow rate, the micro messenger use flow rate, and the weight of each index. The complaint user prediction means determines the sum of the plurality of products as a score of the user type indicator.

Illustratively, as shown in FIG. 12, the complaint user prediction means determines a network quality indicator score from the fault alert, coverage level, network load, and signal to noise ratio; determining a public praise evaluation index score by complaint times, NPS scores and satisfaction scores; determining a service perception index score through the uplink speed, the downlink speed, the call drop times, the game jamming rate and the video jamming rate; user type indicator scores are determined by the big package download users, video users, game users, voice users, web page users, and micro-credit users.

It should be noted that, in fig. 12, specific parameters for determining the network quality index, the public praise evaluation index, the service perception index, and the user type index are only described as an example, and in practical application, the complaint user prediction device may also determine the network quality index, the public praise evaluation index, and the service perception index through other parameters, which is not limited in the present application.

Based on the technical scheme, the complaint user prediction device determines a big packet downloading flow rate, a video use flow rate, a game use flow rate, a voice use flow rate, a web page use flow rate and a micro messenger use flow rate, and determines a fourteenth weight value of the big packet downloading flow rate, a fifteenth weight value of the video use flow rate, a sixteenth weight value of the game use flow rate, a seventeenth weight value of the voice use flow rate, an eighteenth weight value of the web page use flow rate and a nineteenth weight value of the micro messenger use flow rate. Then, the complaint user prediction means determines a score of the user type index based on a weighted sum of a fourteenth weight value, a fifteenth weight value, a sixteenth weight value, a seventeenth weight value, an eighteenth weight value, and a nineteenth weight value for the large packet download traffic ratio, the video use traffic ratio, the game use traffic ratio, the voice use traffic ratio, the web page use traffic ratio, and the micro messenger use traffic ratio. The complaint user prediction device obtains the scores of the user type indexes through the technical scheme, so as to judge the influence condition of the user type on the complaint of the user.

The embodiment of the application can divide the functional modules or functional units of the complaint user prediction device according to the method example, for example, each functional module or functional unit can be divided corresponding to each function, and two or more functions can be integrated in one processing module. The integrated modules may be implemented in hardware, or in software functional modules or functional units. The division of the modules or units in the embodiment of the present application is schematic, which is merely a logic function division, and other division manners may be implemented in practice.

As shown in fig. 13, a schematic structural diagram of a complaint user prediction device 130 according to an embodiment of the present application is provided, where the device includes: a processing unit 1301.

A processing unit 1301 configured to determine scores of a plurality of first indicators of the target user, where the first indicators include at least one of: network quality index, praise evaluation index, service perception index, and user type index; wherein the network quality index is used for representing the network quality of the network used by the user; the public praise evaluation index is used for representing the satisfaction degree of the user on the network; the service perception index is used for representing service perception of a user on a service; the user type index is used to characterize the type of service used by the user.

Processing unit 1301 is further configured to determine a weight value of each first indicator in the plurality of first indicators.

The processing unit 1301 is further configured to predict a complaint index of the target user based on the scores of the plurality of first indexes and the weight value of each first index.

Complaint user prediction apparatus 130 further includes: a communication unit 1302; a communication unit 1302, configured to determine a sample value of at least one second index corresponding to each first index; processing unit 1301 is specifically configured to determine a score of each first indicator based on a sample value of at least one second indicator corresponding to each first indicator.

The plurality of first indicators includes a network quality indicator; the at least one second index corresponding to the network quality index comprises: fault alert indicator, coverage level indicator, network load indicator, and signal to noise ratio indicator. A communication unit 1302, specifically configured to determine a fault alert duration value, a coverage level strength value, a network load value, and a signal to noise ratio value; a processing unit 1301, configured to determine a first weight value of a fault alert duration value, a second weight value of a coverage level strength value, a third weight value of a network load value, and a fourth weight value of a signal to noise value; the processing unit 1301 is specifically configured to determine a score of the network quality indicator by performing weighted summation on the fault alert duration value, the coverage level strength value, the network load value, and the signal-to-noise ratio value based on the first weight value, the second weight value, the third weight value, and the fourth weight value.

The plurality of first indicators includes a public praise evaluation indicator; the at least one second index corresponding to the public praise evaluation index comprises: complaint frequency index, net recommendation value NPS scoring index and satisfaction scoring index. A communication unit 1302, specifically configured to determine complaint times, NPS scores, and satisfaction scores; a processing unit 1301, configured to determine a fifth weight value of the number of complaints, a sixth weight value of the number of complaints, a seventh weight value of the NPS score, and an eighth weight value of the satisfaction score; the processing unit 1301 is specifically configured to determine the score of the public praise evaluation index by weighted summation of the complaint times, the NPS score and the satisfaction score based on the fifth weight value, the sixth weight value, the seventh weight value and the eighth weight value.

The plurality of first metrics includes a traffic awareness metric; the at least one second index corresponding to the service awareness index comprises: a downlink rate index, an uplink rate index, a call drop frequency index, a game click-through rate index and a video click-through rate index. The communication unit 1302 is specifically configured to determine a downlink rate, an uplink rate, a call drop number, a game click-through rate, and a video click-through rate; a processing unit 1301, configured to determine a ninth weight value of a downlink rate, a tenth weight value of an uplink rate, an eleventh weight value of a call drop number, a twelfth weight value of a game click-through rate, and a thirteenth weight value of a video click-through rate; the processing unit 1301 is specifically configured to determine a score of the service perception indicator by performing weighted summation on the downlink rate, the uplink rate, the call drop number, the game click-through rate, and the video click-through rate based on the ninth weight value, the tenth weight value, the eleventh weight value, the twelfth weight value, and the thirteenth weight value.

The plurality of first indicators includes a user type indicator; the at least one second index corresponding to the user type index comprises: big packet download service index, video service index, game service index, voice service index, webpage service index and WeChat service index. A communication unit 1302, configured to determine a packet download traffic ratio, a video usage traffic ratio, a game usage traffic ratio, a voice usage traffic ratio, a web page usage traffic ratio, and a micro messenger usage traffic ratio; a processing unit 1301, configured to determine a fourteenth weight value of a packet download traffic ratio, a fifteenth weight value of a video usage traffic ratio, a sixteenth weight value of a game usage traffic ratio, a seventeenth weight value of a voice usage traffic ratio, an eighteenth weight value of a web page usage traffic ratio, and a nineteenth weight value of a micro messenger usage traffic ratio; treatment sheet

And a unit 1301, configured to determine a score of the user type indicator based on a weighted sum of the packet download traffic ratio, the video usage traffic ratio, the game usage traffic ratio, the voice usage traffic ratio, the web page usage traffic ratio, and the micro messenger usage traffic ratio, based on a fourteenth weight value, a fifteenth weight value, a sixteenth weight value, a seventeenth weight value, an eighteenth weight value, and a nineteenth weight value.

In a possible implementation manner, the complaint user prediction apparatus 130 may further include a storage unit 1303 (shown in a dashed box in fig. 13), where the storage unit 1303 stores a program or an instruction, and when the processing unit 1301 executes the program or the instruction, the complaint user prediction apparatus 130 may perform the complaint user prediction method described in the above method embodiment.

From the foregoing description of the embodiments, it will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of functional modules is illustrated, and in practical application, the above-described functional allocation may be implemented by different functional modules according to needs, i.e. the internal structure of the apparatus is divided into different functional modules to implement all or part of the functions described above. The specific working processes of the above-described systems, devices and units may refer to the corresponding processes in the foregoing method embodiments, which are not described herein.

Embodiments of the present application provide a computer program product comprising instructions which, when run on a computer, cause the computer to perform the complaint user prediction method of the method embodiment described above.

The embodiment of the application also provides a computer readable storage medium, wherein the computer readable storage medium stores instructions which, when run on a computer, cause the computer to execute the complaint user prediction method in the method flow shown in the method embodiment.

The computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access Memory (Random Access Memory, RAM), a Read-Only Memory (ROM), an erasable programmable Read-Only Memory (Erasable Programmable Read Only Memory, EPROM), a register, a hard disk, an optical fiber, a portable compact disc Read-Only Memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing, or any other form of computer readable storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an application specific integrated circuit (Application Specific Integrated Circuit, ASIC). In embodiments of the present application, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

Since the complaint user prediction apparatus, the computer-readable storage medium, and the computer program product in the embodiments of the present application can be applied to the above-mentioned method, the technical effects that can be obtained by the method can also refer to the above-mentioned method embodiments, and the embodiments of the present application are not described herein again.

In the several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interface, indirect coupling or communication connection of devices or units, electrical, mechanical, or other form.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.

The present application is not limited to the above embodiments, and any changes or substitutions within the technical scope of the present application should be covered by the scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method of complaint user prediction, the method comprising:

determining a score for a plurality of first metrics of a target user, the first metrics comprising at least one of: network quality index, praise evaluation index, service perception index, and user type index; wherein the network quality index is used for representing the network quality of a network used by a user; the public praise evaluation index is used for representing the satisfaction degree of a user on the network; the service perception index is used for representing service perception of a user on a service; the user type index is used for representing the service type used by the user;

Determining a weight value of each first index of the plurality of first indexes;

and predicting the complaint index of the target user based on the scores of the first indexes and the weight value of each first index.

2. The method of claim 1, wherein determining the scores of the plurality of first indicators for the target user comprises:

determining a sample value of at least one second index corresponding to each first index;

and determining the score of each first index based on the sample value of at least one second index corresponding to each first index.

3. The method of claim 2, wherein the plurality of first metrics include the network quality metrics; the at least one second index corresponding to the network quality index comprises: fault alert indicator, coverage level indicator, network load indicator, and signal to noise ratio indicator; the determining the score of each first index based on the sample value of at least one second index corresponding to each first index comprises:

determining a fault alarm duration value, a coverage level intensity value, a network load value and a signal to noise value;

determining a first weight value of a fault alarm duration value, a second weight value of a coverage level intensity value, a third weight value of a network load value and a fourth weight value of a signal to noise value;

And based on the first weight value, the second weight value, the third weight value and the fourth weight value, weighting and summing the fault alarm duration value, the coverage level intensity value, the network load value and the signal to noise ratio value, and determining the score of the network quality index.

4. The method of claim 2, wherein the plurality of first metrics include the praise rating metrics; at least one second index corresponding to the public praise evaluation index comprises: complaint frequency index, net recommendation value NPS scoring index and satisfaction scoring index; the determining the score of each first index based on the sample value of at least one second index corresponding to each first index comprises:

determining complaint times, NPS scores and satisfaction scores;

determining a fifth weight value of complaint times, a sixth weight value of complaint times, a seventh weight value of NPS scores and an eighth weight value of satisfaction scores;

and based on the fifth weight value, the sixth weight value, the seventh weight value and the eighth weight value, the complaint times, the NPS scores and the satisfaction scores are weighted and summed to determine the scores of the public praise evaluation indexes.

5. The method of claim 2, wherein the plurality of first metrics include the traffic awareness metrics; the at least one second index corresponding to the service awareness index comprises: a downlink rate index, an uplink rate index, a call drop frequency index, a game click-through rate index and a video click-through rate index; the determining the score of each first index based on the sample value of at least one second index corresponding to each first index comprises:

determining downlink speed, uplink speed, call drop times, game jamming rate and video jamming rate;

determining a ninth weight value of a downlink rate, a tenth weight value of an uplink rate, an eleventh weight value of call drop times, a twelfth weight value of a game click-through rate and a thirteenth weight value of a video click-through rate;

and based on the ninth weight value, the tenth weight value, the eleventh weight value, the twelfth weight value and the thirteenth weight value, the downlink speed, the uplink speed, the call drop times, the game click-through rate and the video click-through rate are weighted and summed, and the score of the service perception index is determined.

6. The method of any of claims 2-5, wherein the plurality of first metrics include the user type metrics; the at least one second index corresponding to the user type index comprises: big packet download service index, video service index, game service index, voice service index, webpage service index and WeChat service index; the determining the score of each first index based on the sample value of at least one second index corresponding to each first index comprises:

Determining a large packet downloading flow rate, a video use flow rate, a game use flow rate, a voice use flow rate, a web page use flow rate and a micro messenger use flow rate;

determining a fourteenth weight value of a large packet downloading flow rate, a fifteenth weight value of a video use flow rate, a sixteenth weight value of a game use flow rate, a seventeenth weight value of a voice use flow rate, an eighteenth weight value of a web page use flow rate and a nineteenth weight value of a micro messenger use flow rate;

and determining a score of the user type index based on the fourteenth weight value, the fifteenth weight value, the sixteenth weight value, the seventeenth weight value, the eighteenth weight value and the nineteenth weight value, and a weighted sum of the packet download traffic ratio, the video usage traffic ratio, the game usage traffic ratio, the voice usage traffic ratio, the web page usage traffic ratio and the micro messenger usage traffic ratio.

7. A complaint user prediction apparatus, the apparatus comprising: a processing unit;

the processing unit is configured to determine scores of a plurality of first indexes of the target user, where the first indexes include at least one of the following: network quality index, praise evaluation index, service perception index, and user type index; wherein the network quality index is used for representing the network quality of a network used by a user; the public praise evaluation index is used for representing the satisfaction degree of a user on the network; the service perception index is used for representing service perception of a user on a service; the user type index is used for representing the service type used by the user;

The processing unit is further configured to determine a weight value of each first index of the plurality of first indexes;

the processing unit is further configured to predict a complaint index of the target user based on the scores of the plurality of first indexes and the weight value of each first index.

8. The apparatus of claim 7, wherein the apparatus further comprises: a communication unit;

the communication unit is used for determining sample values of at least one second index corresponding to each first index;

the processing unit is specifically configured to determine a score of each first indicator based on a sample value of at least one second indicator corresponding to each first indicator.

9. A complaint user prediction apparatus comprising: a processor and a communication interface; the communication interface being coupled to the processor for executing a computer program or instructions to implement the complaint user prediction method of any one of claims 1-6.

10. A computer readable storage medium having instructions stored therein which, when executed by a computer, perform the complaint user prediction method of any one of claims 1-6.