Method, device and equipment for wideband network quality demarcation and storage medium

Abstract

The present disclosure provides a broadband network quality difference demarcation method, device, equipment and storage medium, and relates to the technical field of broadband testing. In some embodiments of the present disclosure, the complaint work order information of a complaint user in a set period is obtained; the complaint work order information is input into a user complaint frequency identification model to obtain the user type of the complaint user; a target quality difference demarcation model corresponding to the user type is determined; and a target broadband network operation parameter corresponding to the user type is obtained; the target broadband network operation parameter is input into the target quality difference demarcation model to obtain a quality difference demarcation result; different broadband network operation parameters are used for different user types in the present disclosure, the precision of broadband network quality difference demarcation is improved, and a quality difference demarcation model is used to improve the accuracy of broadband network quality difference demarcation.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of broadband testing, in particular to a broadband network quality difference determination method and device, equipment and a storage medium. BACKGROUND

[0002] With the rapid development of the Internet, more and more users use broadband to surf the Internet. In the process of using broadband network, users may encounter poor network speed, and operators may receive complaints about network quality from users. Poor network speed is caused by many factors, and broadband network quality difference determination technology emerges as the times require.

[0003] At present, the network quality difference determination technology collects home network soft probes, set-top box soft probes, and deep packet analysis and deep packet detection data to determine the network quality difference. The accuracy of broadband network quality difference determination is low, and the precision is low. SUMMARY

[0004] The present disclosure provides a broadband network quality difference determination method, device, equipment and storage medium to at least solve the problem of low accuracy and low precision of existing broadband network quality difference determination.

[0005] The technical solution of the present disclosure is as follows:

[0006] The present disclosure provides a broadband network quality difference determination method, which comprises:

[0007] Obtaining complaint work order information of a complaint user in a set period;

[0008] Inputting the complaint work order information into a user complaint frequency identification model to obtain a user type of the complaint user;

[0009] Determining a target quality difference determination model corresponding to the user type; and

[0010] Obtaining target broadband network operation parameters corresponding to the user type;

[0011] Inputting the target broadband network operation parameters into the target quality difference determination model to obtain a quality difference determination result.

[0012] Optionally, the obtaining of the complaint work order information of the complaint user in the set period comprises:

[0013] Obtaining at least one of the following complaint work order information of the complaint user in the set period: total complaint times, average complaint time interval, minimum complaint time interval, maximum complaint times of the same complaint fact, average complaint times of the same complaint fact, minimum complaint time interval of the same complaint fact, maximum continuous duration of the same complaint fact, repeated complaint times before processing is completed, and repeated complaint times after processing is completed.

[0014] Optionally, the determining the target quality difference bounding model corresponding to the user type comprises:

[0015] In a case where the user type is a first complaint frequency user, determining a first quality difference bounding model of a first time granularity as the target quality difference bounding model;

[0016] In a case where the user type is a second complaint frequency user, determining a second quality difference bounding model of a second time granularity as the target quality difference bounding model;

[0017] The complaint frequency of the first complaint frequency user is greater than the complaint frequency of the second complaint frequency user, and the first time granularity is less than the second time granularity.

[0018] Optionally, the acquiring the target wideband network running parameter corresponding to the user type comprises:

[0019] In a case where the user type is the first complaint frequency user, and the current complaint of the complaint user corresponds to complaint analysis, acquiring a network running parameter of a first time granularity in a first time period before a user complaint time as the target wideband network running parameter;

[0020] In a case where the user type is the first complaint frequency user, and the current complaint of the complaint user corresponds to evaluation analysis after complaint processing, acquiring a network running parameter of the first time granularity in a second time period before a current time as the target wideband network running parameter;

[0021] In a case where the user type is the second complaint frequency user, and the current complaint of the complaint user corresponds to complaint analysis, acquiring a network running parameter of a second time granularity in a third time period before the user complaint time as the target wideband network running parameter;

[0022] In a case where the user type is the second complaint frequency user, and the current complaint of the complaint user corresponds to evaluation analysis after complaint processing, acquiring a network running parameter of the second time granularity in a fourth time period before the current time as the target wideband network running parameter.

[0023] Optionally, the method further comprises:

[0024] In a case where the user type is the second complaint frequency user, and the user type is a user randomly selected by daily inspection, acquiring a network running parameter of the second time granularity in a fifth time period before the current time as the target wideband network running parameter.

[0025] Optionally, the target broadband network operation parameter comprises: a home side network operation parameter, a service side network operation parameter, an access side network operation parameter and a metropolitan area network side network operation parameter; the target broadband network operation parameter is input into the target quality defect boundary model to obtain a quality defect boundary result, comprising:

[0026] The home side network operation parameter, the service side network operation parameter, the access side network operation parameter and the metropolitan area network side network operation parameter are input into the target quality defect boundary model to obtain a quality defect boundary result.

[0027] Optionally, before using the target quality defect boundary model, the method further comprises:

[0028] Obtaining a sample broadband network operation parameter corresponding to a sample user type;

[0029] Obtaining a sample quality defect boundary result corresponding to the sample broadband network operation parameter;

[0030] According to the sample quality defect boundary result, the sample broadband network operation parameter and a focal loss function, an initial model is trained to obtain the target quality defect boundary model.

[0031] Optionally, the obtaining of the sample broadband network operation parameter corresponding to the sample user type comprises:

[0032] Collecting an original broadband network operation parameter corresponding to a sample user type;

[0033] Using a random forest, an importance degree of each original broadband network operation parameter is calculated;

[0034] From the original broadband network operation parameter, the sample broadband network operation parameter satisfying an importance degree condition is selected.

[0035] The present disclosure further provides a broadband network quality defect boundary device, comprising:

[0036] An information acquisition module is configured to acquire complaint work order information of a complaint user in a set period;

[0037] A type identification module is configured to input the complaint work order information into a user complaint frequency identification model to obtain a user type of the complaint user;

[0038] A model determination module is configured to determine a target quality defect boundary model corresponding to the user type;

[0039] A parameter acquisition module is configured to acquire a target broadband network operation parameter corresponding to the user type;

[0040] The quality difference determination module is configured to input the target wideband network operation parameter into the target quality difference determination model to obtain a quality difference determination result.

[0041] Optionally, the information acquisition module is configured to, when acquiring the complaint work order information of the complaint user in a set period, acquire at least one of the following complaint work order information of the complaint user in the set period: total complaint times, average complaint time interval, minimum complaint time interval, maximum complaint times of the same complaint fact, average complaint times of the same complaint fact, minimum complaint time interval of the same complaint fact, maximum duration of the same complaint fact, repeated complaint times before processing completion, and repeated complaint times after processing completion.

[0042] Optionally, the information acquisition module is configured to, when acquiring the complaint work order information of the complaint user in a set period, acquire at least one of the following complaint work order information of the complaint user in the set period: total complaint times, average complaint time interval, minimum complaint time interval, maximum complaint times of the same complaint fact, average complaint times of the same complaint fact, minimum complaint time interval of the same complaint fact, maximum duration of the same complaint fact, repeated complaint times before processing completion, and repeated complaint times after processing completion.

[0043] Optionally, the model determination module is configured to, when determining the target quality difference determination model corresponding to the user type, determine, in a case where the user type is a first complaint frequency user, a first quality difference determination model of a first time granularity as the target quality difference determination model.

[0044] Optionally, the model determination module is configured to, when determining the target quality difference determination model corresponding to the user type, determine, in a case where the user type is a second complaint frequency user, a second quality difference determination model of a second time granularity as the target quality difference determination model.

[0045] Optionally, the model determination module is configured to, when determining the target quality difference determination model corresponding to the user type, determine, in a case where the user type is a second complaint frequency user, a second quality difference determination model of a second time granularity as the target quality difference determination model.

[0046] Optionally, the model determination module is configured to, when determining the target quality difference determination model corresponding to the user type, determine, in a case where the user type is a second complaint frequency user, a second quality difference determination model of a second time granularity as the target quality difference determination model.

[0047] Optionally, the parameter acquisition module is configured to, when acquiring the target wideband network operation parameter corresponding to the user type, acquire, in a case where the user type is the first complaint frequency user and the current complaint of the complaint user corresponds to complaint analysis, a network operation parameter of the first time granularity in a first time period before a user complaint time as the target wideband network operation parameter.

[0048] Optionally, the parameter acquisition module is configured to, when acquiring the target wideband network operation parameter corresponding to the user type, acquire, in a case where the user type is the first complaint frequency user and the current complaint of the complaint user corresponds to complaint analysis, a network operation parameter of the first time granularity in a first time period before a user complaint time as the target wideband network operation parameter.

[0049] Optionally, the parameter acquisition module is configured to, when acquiring the target wideband network operation parameter corresponding to the user type, acquire, in a case where the user type is the first complaint frequency user and the current complaint of the complaint user corresponds to complaint analysis, a network operation parameter of the first time granularity in a first time period before a user complaint time as the target wideband network operation parameter.

[0050] Optionally, the parameter acquisition module is configured to, when acquiring the target wideband network operation parameter corresponding to the user type, acquire, in a case where the user type is the second complaint frequency user and the current complaint of the complaint user corresponds to complaint analysis, a network operation parameter of the second time granularity in a third time period before the user complaint time as the target wideband network operation parameter.

[0051] In a case where the user type is the second complaint frequency user, and the current complaint of the complaint user corresponds to an evaluation analysis after complaint processing, a network operation parameter of the second time granularity in a fourth time period before the current time is obtained as the target broadband network operation parameter.

[0052] Optionally, the parameter obtaining module is further configured to:

[0053] In a case where the user type is the second complaint frequency user, and the user type is a user randomly selected by daily inspection, a network operation parameter of the second time granularity in a fifth time period before the current time is obtained as the target broadband network operation parameter.

[0054] Optionally, the target broadband network operation parameter comprises a home side network operation parameter, a business side network operation parameter, an access side network operation parameter and a metropolitan area network side network operation parameter; and the quality difference boundary determination module, when inputting the target broadband network operation parameter into the target quality difference boundary determination model to obtain a quality difference boundary determination result, is configured to:

[0055] input the home side network operation parameter, the business side network operation parameter, the access side network operation parameter and the metropolitan area network side network operation parameter into the target quality difference boundary determination model to obtain the quality difference boundary determination result.

[0056] Optionally, before using the target quality difference boundary determination model, the quality difference boundary determination module is further configured to:

[0057] obtain a sample broadband network operation parameter corresponding to a sample user type;

[0058] obtain a sample quality difference boundary determination result corresponding to the sample broadband network operation parameter;

[0059] train an initial model according to the sample quality difference boundary determination result, the sample broadband network operation parameter and a focal loss function to obtain the target quality difference boundary determination model.

[0060] Optionally, when obtaining the sample broadband network operation parameter corresponding to the sample user type, the quality difference boundary determination module is configured to:

[0061] collect an original broadband network operation parameter corresponding to a sample user type;

[0062] calculate an importance degree of each original broadband network operation parameter by using a random forest;

[0063] select the sample broadband network operation parameter from the original broadband network operation parameter, wherein the importance degree of the sample broadband network operation parameter satisfies an importance degree condition.

[0064] The embodiments of the present disclosure further provide an electronic device, comprising:

[0065] a processor;

[0066] a memory for storing the processor-executable instructions;

[0067] The processor is configured to execute the instructions to implement each step in the above method.

[0068] The embodiments of the present disclosure further provide a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement each step in the above method.

[0069] The embodiments of the present disclosure further provide a computer program product comprising computer programs / instructions, the computer programs / instructions being executed by a processor to implement each step in the above method.

[0070] The embodiments of the present disclosure provide at least the following beneficial effects:

[0071] In some embodiments of the present disclosure, the complaint user's complaint work order information in a set period is obtained; the complaint work order information is input into a user complaint frequency identification model to obtain the user type of the complaint user; a target quality difference boundary model corresponding to the user type is determined; and a target broadband network operation parameter corresponding to the user type is obtained; the target broadband network operation parameter is input into the target quality difference boundary model to obtain a quality difference boundary result; different broadband network operation parameters are used for different user types in the present disclosure, which improves the precision of broadband network quality difference boundary determination, and the quality difference boundary model is used to improve the accuracy of broadband network quality difference boundary determination.

[0072] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and do not limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0073] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present disclosure and, together with the specification, serve to explain the principles of the present disclosure, and do not constitute an undue limitation on the present disclosure.

[0074] Figure 1 A flowchart of a broadband network quality difference boundary determination method according to an exemplary embodiment of the present disclosure is shown in the figure;

[0075] Figure 2 A schematic diagram of a communication network according to an exemplary embodiment of the present disclosure is shown in the figure;

[0076] Figure 3 A comparison diagram of a broadband network quality difference boundary range according to an exemplary embodiment of the present disclosure is shown in the figure;

[0077] Figure 4 A structural schematic diagram of a wideband network quality difference demarcation device provided for an exemplary embodiment of the present disclosure is shown in the figure;

[0078] Figure 5 A structural schematic diagram of an electronic device provided for an exemplary embodiment of the present disclosure is shown in the figure. DETAILED DESCRIPTION

[0079] In order for those skilled in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be described clearly and completely below in conjunction with the accompanying drawings.

[0080] It should be noted that the terms "first", "second", and the like in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Rather, they are merely examples of devices and methods consistent with some aspects of the present disclosure.

[0081] It should be noted that the user information involved in the present disclosure includes but is not limited to user equipment information and user personal information; the collection, storage, use, processing, transmission, provision and disclosure of user information in the present disclosure comply with the provisions of relevant laws and regulations, and do not violate public order and good customs.

[0082] At present, the network quality difference demarcation technology collects home network soft probes, set-top box soft probes, and deep packet analysis and deep packet detection data to demarcate network quality differences. The above network operation parameter statistical period is a day, the granularity is relatively coarse, the radix is relatively large, and the quality difference of individual period is easy to be lost, resulting in low accuracy. For example, a user has a lot of failures for 10 minutes in a day, which greatly affects the user's perception, but the success rate observed in a day may not reach the threshold of quality difference, and it cannot be identified as quality difference.

[0083] In order to solve the above technical problems, in some embodiments of the present disclosure, the complaint work order information of the complaint user in a set period is obtained; the complaint work order information is input into a user complaint frequency identification model to obtain the user type of the complaint user; a target quality difference demarcation model corresponding to the user type is determined; and target wideband network operation parameters corresponding to the user type are obtained; the target wideband network operation parameters are input into the target quality difference demarcation model to obtain a quality difference demarcation result; different user types of the present disclosure use different wideband network operation parameters, which improves the accuracy of wideband network quality difference demarcation, and uses a quality difference demarcation model, which improves the accuracy of wideband network quality difference demarcation.

[0084] The technical solutions provided by the embodiments of the present disclosure are described in detail below with reference to the drawings.

[0085] Figure 1 A flowchart of a broadband network quality boundary determination method provided by an exemplary embodiment of the present disclosure is shown in FIG. 1. As shown in FIG. 1, the method comprises the following steps. Figure 1

[0086] S101: obtaining complaint work order information of a complaint user in a set period;

[0087] S102: inputting the complaint work order information into a user complaint frequency identification model to obtain a user type of the complaint user;

[0088] S103: determining a target quality boundary determination model corresponding to the user type; and

[0089] S104: obtaining target broadband network operation parameters corresponding to the user type;

[0090] S105: inputting the target broadband network operation parameters into the target quality boundary determination model to obtain a quality boundary determination result.

[0091] In the present embodiment, the execution subject of the above method can be a server or a terminal device.

[0092] ​The terminal device includes, but is not limited to, a mobile station (MS), a mobile terminal, a mobile telephone, a handset, a portable equipment, and the like. The terminal device can communicate with one or more core networks through a radio access network (RAN). For example, the terminal device can be a mobile phone (also referred to as a "cellular" phone), a computer with wireless communication functions, and the like. The terminal device can also be a computer with wireless transceiver functions, a virtual reality (VR) terminal device, an AR terminal device, a wireless terminal in industrial control, a wireless terminal in self driving, a wireless terminal in remote medical treatment, a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, a wireless terminal in a smart home, and the like. The operating system installed on the terminal device includes, but is not limited to, an IOS, an Android, a windows, a linux, a Mac OS, and the like. The terminal device can be referred to by different names in different networks, such as a user equipment, a mobile station, a user unit, a station, a cellular phone, a personal digital assistant, a wireless modem, a wireless communication device, a handheld device, a laptop computer, a cordless phone, a wireless local loop station, a television, and the like. For the convenience of description, the terminal device is referred to as a terminal device in this embodiment.

[0093] In this embodiment, the implementation form of the server is described. For example, the server can be a conventional server, a cloud server, a cloud host, a virtual center, and the like. The server mainly includes a processor, a hard disk, a memory, a system bus, and the like, and a general computer architecture type.

[0094] In this embodiment, the complaint user's complaint work order information in a set period is obtained; the complaint work order information is input into a user complaint frequency identification model to obtain the user type of the complaint user; a target quality difference boundary model corresponding to the user type is determined; a target broadband network operation parameter corresponding to the user type is obtained; the target broadband network operation parameter is input into the target quality difference boundary model to obtain a quality difference boundary result; different broadband network operation parameters are used for different user types in this disclosure, the precision of broadband network quality difference boundary is improved, and the accuracy of broadband network quality difference boundary is improved by using the quality difference boundary model.

[0095] I. Data collection. This part introduces the collection of training data of the quality difference bounding model, and the data collection in the model using stage.

[0096] Obtain the home side network running parameter, the service side network running parameter, the access side network running parameter and the metropolitan area network side network running parameter.

[0097] The home side network running parameter is the intelligent gateway soft probe data, the service side network running parameter is the DPI (Deep Packet Inspection) data, and the access side network running parameter and the metropolitan area network side network running parameter are the QCA (Quality of Connection Analysis) data. Figure 2 A schematic diagram of a communication network is provided for the exemplary embodiments of the present disclosure. The collection positions of the above three kinds of data in the whole network are shown in FIG. 1. Figure 2 Figure 2 In FIG. 1, STB represents a digital television receiving device, which is usually used to receive digital television signals in a home; ONT is a network interface terminal, which is used to connect the network to a home; LSW is a local switching workstation, which is used for voice communication in a communication network; CR refers to a cable system for transmitting television signals; and PB is a private branch exchange, which is usually used for internal communication in an enterprise or organization.

[0098] From the collection positions in the figure, it can be seen that the three kinds of data are complementary to each other and indispensable, and each has its own advantages. The intelligent gateway soft probe data is closer to the real situation of user use, the DPI data is closer to the real situation of the service side, and the QCA data reflects the real situation of a certain OLT (Optical Line Terminal) or BNG (Broadband Network Gateway) in the network. Taking TCP packet loss as an example, each kind of data can detect whether the data received by the node has packet loss, but cannot determine the position where the packet loss occurs. At the same time, collecting the three kinds of data, for example, the following situations may occur:

[0099] The DPI data shows no downlink packet loss.

[0100] The QCA data of the BNG shows no downlink packet loss.

[0101] The QCA data of the OLT shows downlink packet loss.

[0102] The probe data of the ONT shows downlink packet loss.

[0103] Figure 3 A broadband network quality difference bounding range comparison schematic diagram is provided for the exemplary embodiments of the present disclosure. As shown in FIG. 2, Figure 3 ​As shown, by the network node topology relationship and the representation of the above data, the packet loss problem can be clearly located between the BNG and the OLT. If one or more of the data is missing, the delimited range will become larger and the accuracy will become smaller. For example, only with DPI data and ONT data, it can only be delimited as a service side problem, a problem between the DPI collection point and the ONT, a problem below the ONT, but the nodes between the DPI collection point and the ONT include the OLT, the BNG, the CR, etc., and only with the DPI data, it cannot be determined which section has the problem.

[0104] Preferably, the access side and the metro network side user data are acquired by collecting data through a northbound interface of a mobile broadband network QCA device.

[0105] Preferably, the access side and the metro network side user data are QCA data of OLT and BNG network elements, including user key indicators and user service key indicators. The access side and the metro network side user data are as follows:

[0106]

[0107]

[0108]

[0109]

[0110] Preferably, the home side network running parameters are data of an online optical network terminal in a broadband user's home, including average freezing and screen flashing frequency data, average freezing frequency of on-demand, average screen flashing frequency of live broadcast, single user resource data, single user period data, single user viewing data, and single user alarm data.

[0111] Preferably, the service side network running parameters are quality indicator data of user use of online services acquired by using application deep packet inspection technology, including request frequency and success frequency of instant communication, reading, microblog community, video, game, payment, animation, P2P service, browsing and downloading, security and antivirus, shopping, travel, video live broadcast, cloud game, and unclassified service.

[0112] The embodiments of the present disclosure use more basic network and user indicator data, cover a more comprehensive network topology, cover a more extensive user perception indicator, more accurately and precisely reflect the relationship between home broadband network performance quality and customer quality difference, finely distinguish high-frequency complaint users and occasional complaint users, and establish different granularity models accordingly, find specific problems of user complaints about broadband services in time, and improve troubleshooting efficiency.

[0113] Second, data processing. This part introduces the processing of training data of the quality difference boundary model, and the data processing in the model use stage.

[0114] The collected user data of the home side, the business side, the access side and the metropolitan area network side are uniformly gathered and processed into 15-minute granularity and day granularity key indicators. A small part of users are sensitive to broadband performance, and the quality difference in a short time will cause complaints. For this part of users, it is more appropriate to use 15-minute granularity data for analysis and prediction; in addition, most users consider resource occupation and performance consumption, and it is more appropriate to use day granularity data for analysis and prediction.

[0115] Specifically, taking 15-minute granularity as an example: the data is divided into 15-minute intervals of natural time, that is, 1 hour is divided into 4 time intervals, which are 0-14:59, 15-29:59, 30-44:59 and 45-59:59. Assuming that a user only performs 15-minute business interaction from 5 minutes to 20 minutes, the key indicators of the user should be divided into 2 time intervals, which are the key indicators in the time interval from 5 minutes to 14:59 and the key indicators in the time interval from 15 minutes to 20 minutes.

[0116] For count type data such as the number of stalls, the values in the time interval are accumulated. For average type data such as the average delay of TCP handshake, the measurement values in the time interval such as the number of TCP handshakes and the total delay of TCP handshake are accumulated, and then the average value is calculated.

[0117] In some embodiments of the present disclosure, complaint user information of a complaint user in a set period is obtained; the complaint work order information is input into a user complaint frequency identification model to obtain the user type of the complaint user. The embodiments of the present disclosure identify high-frequency complaint users and occasional complaint users and establish different granularity models accordingly, support subsequent differentiated finding of specific problems of user complaints about broadband services, and improve troubleshooting efficiency. The present disclosure identifies the user type by using the user complaint frequency identification model, which helps to fine-grained user management, improves the accuracy of quality difference boundary, and improves customer satisfaction.

[0118] Third, establishing a target quality difference boundary model.

[0119] The target quality difference demarcation model in the present disclosure includes a first quality difference demarcation model and a second quality difference demarcation model, wherein the first quality difference demarcation model is a model of a first time granularity for a first complaint frequency user, the second quality difference demarcation model is a model of a second time granularity for a second complaint frequency user, the complaint frequency of the first complaint frequency user is greater than the complaint frequency of the second complaint frequency user, and the first time granularity is smaller than the second time granularity. The first complaint frequency user is a high-frequency complaint user, and the second complaint frequency user is an occasional complaint user. It should be noted that the present embodiment does not limit the first time granularity and the second time granularity, and the first time granularity can be 5 minutes, 10 minutes, 15 minutes, 30 minutes, etc.; the second time granularity can be day, week, month, etc.

[0120] According to a certain time range, the historical data of the historical quality difference user is extracted, the historical quality difference user refers to the user who has ever complained about the broadband quality difference, which is divided into two cases. One case is a high-frequency complaint user predicted by a high-frequency complaint user identification model, and usually such a user is more sensitive to service usage, and short-time service unavailability can trigger a complaint. The other case is an occasional complaint user predicted by the high-frequency complaint user identification model.

[0121] In the present embodiment, the quality difference demarcation result includes but is not limited to: home side quality difference, ONT optical network device quality difference, OLT optical line terminal quality difference, BNG gateway quality difference, metropolitan area network quality difference, provincial network quality difference and service side quality difference.

[0122] For example, in the case of a high-frequency complaint user, 15-minute granularity data is used for analysis, 15-minute granularity data of the user within 3 hours before each complaint is extracted, which is identified as quality difference historical data, and the quality difference demarcation result is given after analysis. In addition, a same order of magnitude of 15-minute granularity historical data of non-complaint users is randomly extracted, which is identified as non-quality difference data. These performance indicators are used as input data, and the user quality difference demarcation category is used as output data. The improved CatBoost algorithm is used to train the first quality difference demarcation model to determine the relationship between the 15-minute granularity performance indicators and the user quality difference demarcation category.

[0123] For another example, in the case of an occasional complaint user, day granularity data is used for analysis, day granularity data of the user within 7 days before each complaint is extracted, which is identified as quality difference historical data, and the quality difference demarcation result is given after analysis. In addition, a same order of magnitude of day granularity historical data of non-complaint users is randomly extracted, which is identified as non-quality difference data. These performance indicators are used as input data, and the user quality difference demarcation category is used as output data. The improved CatBoost algorithm is used to train the second quality difference demarcation model to determine the relationship between the day granularity performance indicators and the user quality difference demarcation category.

[0124] In some embodiments of the present disclosure, sample broadband network operation parameters corresponding to a sample user type are obtained. One implementable way is to collect original broadband network operation parameters corresponding to the sample user type, calculate the importance of each original broadband network operation parameter using a random forest, and select sample broadband network operation parameters whose importance satisfies an importance condition from the original broadband network operation parameters.

[0125] Specifically, the importance of each data is calculated using a random forest on the performance index data obtained above, and is sorted from large to small. The features with the smallest importance are deleted by a specific ratio to obtain a new feature set. A new random forest is established using the new feature set, the importance of each feature in the feature set is calculated and sorted, and the features with the smallest importance are deleted by a specific ratio to obtain a new feature set. The above steps are repeated to obtain a batch of the most important feature variable data. For example, the most important feature variable data to be retained is set to 20. Then, the retained feature variable data is divided into a training set and a test set, and is input into an improved CatBoost classification model for training.

[0126] In some embodiments of the present disclosure, a focal loss function is used instead of the default cross-entropy loss function of CatBoost to solve the problem of extremely unbalanced number of partial classification samples, and to increase the weight of small number of target categories and classification error samples. The commonly used loss functions of CatBoost include root mean square error, logarithmic loss, mean absolute error, cross-entropy, quantile, and mean absolute percentage error. These loss functions are more suitable for the case where the samples are roughly balanced. For broadband user quality boundary application, it is difficult to ensure the balance of each classification sample. Generally, the sample of non-quality users is the largest, among the quality samples, the sample of family side quality is the largest, followed by the sample of business side quality, and the samples of BNG quality and metropolitan network quality account for a very small proportion. This sample imbalance will seriously affect the precision during the training of the classifier, resulting in a decrease in the accuracy of the model prediction. However, the improved use of the focal loss function in the present application can better eliminate the negative effects of sample imbalance.

[0127] The focal loss function formula is as follows:

[0128]

[0129] where L f1 is the loss value, N is the total number of samples, y i is the true classification of the i-th sample, is the probability of the prediction being correct for the i-th sample, a is a weight to reduce the imbalance of positive and negative samples, used to increase the weight proportion of small number of samples, and g is a weight to reduce the imbalance of difficult and easy samples, used to increase the weight proportion of classification error samples. Preferably, a takes a value of 0.25, and g takes a value of 2.

[0130] Fourth, model running prediction.

[0131] In some embodiments of the present disclosure, complaint user's complaint work order information in a set period is obtained. One realizable way is to obtain at least one of the following complaint work order information of the complaint user in the set period: total complaint times, average complaint time interval, minimum complaint time interval, maximum complaint times of the same complaint fact, average complaint times of the same complaint fact, minimum complaint time interval of the same complaint fact, maximum duration of the same complaint fact, repeated complaint times before processing completion, and repeated complaint times after processing completion.

[0132] In some embodiments of the present disclosure, target broadband network running parameters corresponding to the user type are obtained. One realizable way is that, in the case that the user type is a first complaint frequency user, and the current complaint of the complaint user corresponds to complaint analysis, the network running parameters of a first time granularity in a first time period before the user complaint time are obtained as the target broadband network running parameters; in the case that the user type is the first complaint frequency user, and the current complaint of the complaint user corresponds to evaluation analysis after complaint processing, the network running parameters of the first time granularity in a second time period before the current time are obtained as the target broadband network running parameters; in the case that the user type is a second complaint frequency user, and the current complaint of the complaint user corresponds to complaint analysis, the network running parameters of a second time granularity in a third time period before the user complaint time are obtained as the target broadband network running parameters; in the case that the user type is the second complaint frequency user, and the current complaint of the complaint user corresponds to evaluation analysis after complaint processing, the network running parameters of the second time granularity in a fourth time period before the current time are obtained as the target broadband network running parameters. In particular, in the case that the user type is the second complaint frequency user, and the user type is a user randomly selected by daily inspection, the network running parameters of the second time granularity in a fifth time period before the current time are obtained as the target broadband network running parameters.

[0133] In an example embodiment, the previous stage is grouped according to the complaint work order in a period of time, preferably one year, according to the complaint user number, and the total complaint times, the average complaint time interval, the minimum complaint time interval, the maximum complaint times of the same complaint fact, the average complaint times of the same complaint fact, the minimum complaint time interval of the same complaint fact, the maximum duration of the same complaint fact, the repeated complaint times before the completion of processing, the repeated complaint times after the completion of processing and other key indicators are counted as the input layer, and the marked high-frequency complaint user is classified as the output layer, a three-layer BP neural network is established and model training is performed to obtain a better high-frequency complaint user identification model; when receiving a demand to be analyzed, the total complaint times, the average complaint time interval, the minimum complaint time interval, the maximum complaint times of the same complaint fact, the average complaint times of the same complaint fact, the minimum complaint time interval of the same complaint fact, the maximum duration of the same complaint fact, the repeated complaint times before the completion of processing, the repeated complaint times after the completion of processing and other key indicators of the user to be analyzed are taken as inputs, and the above-mentioned better BP neural network model is operated to output the prediction result.

[0134] For example, if the prediction result of the user to be analyzed belongs to a high-frequency complaint user and this time is also a complaint analysis, 15-minute granularity data of the user 3 hours before the complaint is extracted; if the user to be analyzed belongs to a high-frequency complaint user but this time is an evaluation analysis after the complaint processing, 15-minute granularity data of the current time 3 hours before is extracted. The above-mentioned data is input into the 15-minute granularity model to directly output the quality difference boundary prediction result of the user. If the user to be analyzed belongs to an accidental complaint user and this time is also a complaint analysis, day-minute granularity data of the user 7 days before the complaint is extracted; if the user to be analyzed belongs to an accidental complaint user but this time is an evaluation analysis after the complaint processing or is a user randomly selected by daily inspection, day granularity data of the current time 3 days before is extracted. The above-mentioned data is input into the day granularity model to directly output the quality difference boundary prediction result of the user.

[0135] The quality difference boundary model is trained by using the improved CatBoost algorithm, the deep relationship between the user performance data and the quality difference boundary is mined, the collected user data is substituted into the quality difference boundary model, and then the quality difference boundary is output to evaluate whether the user has quality difference in real time, and the weak link of network operation and the optimization direction are found in time.

[0136] Figure 4 A structure schematic diagram of a broadband network quality difference boundary device 40 provided by an example embodiment of the present disclosure is shown in FIG. 1. Figure 4 As shown in the figure, the broadband network quality difference boundary device 40 includes an information acquisition module 41, a type identification module 42, a model determination module 43, a parameter acquisition module 44 and a quality difference boundary module 45.

[0137] The information acquisition module 41 is configured to acquire complaint work order information of the complaint user in a set period.

[0138] The type identification module 42 is configured to input the complaint work order information into a user complaint frequency identification model to obtain a user type of the complaint user.

[0139] The model determination module 43 is configured to determine a target quality difference boundary model corresponding to the user type.

[0140] The parameter acquisition module 44 is configured to acquire a target broadband network operation parameter corresponding to the user type.

[0141] The quality difference boundary module 45 is configured to input the target broadband network operation parameter into the target quality difference boundary model to obtain a quality difference boundary result.

[0142] Optionally, when acquiring the complaint work order information of the complaint user in the set period, the information acquisition module 41 is configured to:

[0143] acquire at least one of the following complaint work order information of the complaint user in the set period: total complaint times, average complaint time interval, minimum complaint time interval, maximum complaint times of the same complaint fact, average complaint times of the same complaint fact, minimum complaint time interval of the same complaint fact, maximum continuous duration of the same complaint fact, repeated complaint times before completion of processing, and repeated complaint times after completion of processing.

[0144] Optionally, when determining the target quality difference boundary model corresponding to the user type, the model determination module 43 is configured to:

[0145] in a case where the user type is a first complaint frequency user, determine a first quality difference boundary model of a first time granularity as the target quality difference boundary model;

[0146] in a case where the user type is a second complaint frequency user, determine a second quality difference boundary model of a second time granularity as the target quality difference boundary model;

[0147] wherein the complaint frequency of the first complaint frequency user is greater than the complaint frequency of the second complaint frequency user, and the first time granularity is less than the second time granularity.

[0148] Optionally, when acquiring the target broadband network operation parameter corresponding to the user type, the parameter acquisition module 44 is configured to:

[0149] in a case where the user type is the first complaint frequency user, and the current complaint of the complaint user corresponds to complaint analysis, acquire a network operation parameter of the first time granularity in a first time period before the user complaint time as the target broadband network operation parameter;

[0150] In a case where the user type is the first complaint frequency user, and the current complaint of the complaint user corresponds to the evaluation analysis after complaint processing, a network operation parameter of a first time granularity in a second time period before a current time is obtained as a target broadband network operation parameter;

[0151] In a case where the user type is the second complaint frequency user, and the current complaint of the complaint user corresponds to complaint analysis, a network operation parameter of a second time granularity in a third time period before a user complaint time is obtained as a target broadband network operation parameter;

[0152] In a case where the user type is the second complaint frequency user, and the current complaint of the complaint user corresponds to the evaluation analysis after complaint processing, a network operation parameter of the second time granularity in a fourth time period before the current time is obtained as the target broadband network operation parameter.

[0153] Optionally, the parameter obtaining module 44 can also be configured to:

[0154] In a case where the user type is the second complaint frequency user, and the user type is a user randomly selected by daily inspection, a network operation parameter of the second time granularity in a fifth time period before the current time is obtained as the target broadband network operation parameter.

[0155] Optionally, the target broadband network operation parameter comprises a home side network operation parameter, a service side network operation parameter, an access side network operation parameter and a metropolitan area network side network operation parameter; and the quality difference boundary determination module 45, when inputting the target broadband network operation parameter into a target quality difference boundary determination model to obtain a quality difference boundary determination result, is configured to:

[0156] input the home side network operation parameter, the service side network operation parameter, the access side network operation parameter and the metropolitan area network side network operation parameter into the target quality difference boundary determination model to obtain the quality difference boundary determination result.

[0157] Optionally, the quality difference boundary determination module 45, before using the target quality difference boundary determination model, can also be configured to:

[0158] obtain a sample broadband network operation parameter corresponding to a sample user type;

[0159] obtain a sample quality difference boundary determination result corresponding to the sample broadband network operation parameter;

[0160] train an initial model according to the sample quality difference boundary determination result, the sample broadband network operation parameter and a focal loss function to obtain the target quality difference boundary determination model.

[0161] Optionally, the quality difference boundary determination module 45, when obtaining the sample broadband network operation parameter corresponding to the sample user type, is configured to:

[0162] Collecting original broadband network operation parameters corresponding to a sample user type;

[0163] Using the random forest, calculating the importance of each original broadband network operation parameter;

[0164] From the original broadband network operation parameters, selecting sample broadband network operation parameters whose importance satisfies the importance condition.

[0165] As to the apparatus in the above embodiments, the specific manners in which the respective modules perform operations have been described in detail in the embodiments of the method, and thus will not be described in detail here.

[0166] Figure 5 A structural schematic diagram of an electronic device is provided for the exemplary embodiments of the present disclosure. As shown in the figure, the electronic device includes a memory 51 and a processor 52. In addition, the electronic device also includes a power supply component 53 and a communication component 54. Figure 5

[0167] The memory 51 is configured to store computer programs and can be configured to store other various data to support operations on the electronic device. Examples of these data include instructions for operating any application or method on the electronic device.

[0168] The memory 51 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0169] The communication component 54 is configured to perform data transmission with other devices.

[0170] The processor 52 can execute computer instructions stored in the memory 51, so as to: acquire complaint work order information of a complaint user in a set period; input the complaint work order information into a user complaint frequency identification model to obtain a user type of the complaint user; determine a target quality difference boundary model corresponding to the user type; acquire target broadband network operation parameters corresponding to the user type; and input the target broadband network operation parameters into the target quality difference boundary model to obtain a quality difference boundary result.

[0171] Correspondingly, the embodiments of the present disclosure also provide a computer readable storage medium storing a computer program. When the computer readable storage medium stores the computer program and the computer program is executed by one or more processors, the one or more processors are caused to perform the steps of the method embodiments. Figure 1 The steps in the method embodiments.

[0172] ​Accordingly, the embodiments of the present disclosure also provide a computer program product, which comprises computer programs / instructions Figure 1 to perform the steps of the method embodiments.

[0173] The communication component in the above-described Figure 5 is configured to facilitate wired or wireless communication between the device where the communication component is located and other devices. The device where the communication component is located can access a wireless network based on a communication standard, such as WiFi, a 2G, 3G, 4G / LTE, 5G, or the like mobile communication network, or a combination thereof. In an exemplary embodiment, the communication component receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component further includes a Near Field Communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0174] The power component in the above-described Figure 5 provides power to various components of the device where the power component is located. The power component can include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to the device where the power component is located.

[0175] The electronic device described above further includes a display screen and an audio component.

[0176] The display screen includes a screen, which can include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense a touch, a slide, and a gesture on the touch panel. The touch sensor can not only sense a boundary of a touch or a slide action, but also detect a duration and a pressure associated with a touch or a slide operation.

[0177] The audio component can be configured to output and / or input an audio signal. For example, the audio component includes a microphone (MIC) configured to receive an external audio signal when the device where the audio component is located is in an operation mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal can be further stored in a memory or transmitted via a communication component. In some embodiments, the audio component further includes a speaker to output an audio signal.

[0178] In the device, the equipment, the storage medium and the program product embodiments of the present disclosure, the complaint work order information of the complaint user in a set period is acquired; the complaint work order information is input into a user complaint frequency identification model to obtain a user type of the complaint user; a target quality difference boundary model corresponding to the user type is determined; and a target broadband network operation parameter corresponding to the user type is acquired; the target broadband network operation parameter is input into the target quality difference boundary model to obtain a quality difference boundary result; different broadband network operation parameters are used for different user types of the present disclosure, the precision of broadband network quality difference boundary is improved, and the accuracy of broadband network quality difference boundary is improved by using the quality difference boundary model.

[0179] Those skilled in the art will appreciate that embodiments of the application can be supplied as a method, a system, or a computer program product. Thus, the application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer-readable program code.

[0180] The present application is described in reference to flowcharts and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce an apparatus that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in the flow or flows and / or blocks Figure 1 The functions specified in the flow or flows and / or blocks

[0181] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction apparatus, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in the flow or flows and / or blocks Figure 1 The functions specified in the flow or flows and / or blocks

[0182] These computer program instructions can also be loaded into a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide a process for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1one or more processes and / or blocks Figure 1 steps of a function specified in one or more blocks.

[0183] In one typical configuration, the computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0184] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) about which the computer stores information about the operating environment. This media can also include non-volatile memory, such as read-only memory (ROM) or Flash memory (flash RAM). Memory is an example of computer readable media.

[0185] Computer readable media includes permanent and non-permanent, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile discs (DVDs) or other optical storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer readable media does not include transitory media, such as modulated data signals and carrier waves.

[0186] It should be noted that, in the present document, relational terms such as "first" and "second", and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without limitation, an element preceded by "comprises... a" does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0187] The foregoing is merely illustrative of the various implementations of the present disclosure and the general principles thereof. Numerous modifications can be made to these illustrations, and equivalents can be substituted therefor, without departing from the scope of the present disclosure. The specific embodiments commensurate with the specific application are intended to be illustrative only and not limiting of the scope of the application as set forth in the following claims.

Claims

1. A method for wideband network quality demarcation, the method comprising: The method comprises: obtaining complaint work order information of a complaint user in a set period; inputting the complaint work order information into a user complaint frequency identification model to obtain a user type of the complaint user; determining a target quality difference boundary model corresponding to the user type; in a case where the user type is a first complaint frequency user, determining a first quality difference boundary model of a first time granularity as the target quality difference boundary model; in a case where the user type is a second complaint frequency user, determining a second quality difference boundary model of a second time granularity as the target quality difference boundary model; the complaint frequency of the first complaint frequency user is greater than the complaint frequency of the second complaint frequency user, and the first time granularity is less than the second time granularity; obtaining a target broadband network operation parameter corresponding to the user type; inputting the target broadband network operation parameter into the target quality difference boundary model to obtain a quality difference boundary result.

2. The method of claim 1, wherein, The method comprises: obtaining at least one complaint work order information of the complaint user in the set period, including total complaint times, average complaint time interval, minimum complaint time interval, maximum complaint times of the same complaint fact, average complaint times of the same complaint fact, minimum complaint time interval of the same complaint fact, maximum continuous duration of the same complaint fact, repeated complaint times before processing is completed, and repeated complaint times after processing is completed.

3. The method of claim 1, wherein, The method comprises: in a case where the user type is the first complaint frequency user and the current complaint of the complaint user corresponds to complaint analysis, obtaining a network operation parameter of the first time granularity in a first time period before the user complaint time as the target broadband network operation parameter; in a case where the user type is the first complaint frequency user and the current complaint of the complaint user corresponds to evaluation analysis after complaint processing, obtaining a network operation parameter of the first time granularity in a second time period before the current time as the target broadband network operation parameter; in a case where the user type is the second complaint frequency user and the current complaint of the complaint user corresponds to complaint analysis, obtaining a network operation parameter of the second time granularity in a third time period before the user complaint time as the target broadband network operation parameter; in a case where the user type is the second complaint frequency user and the current complaint of the complaint user corresponds to evaluation analysis after complaint processing, obtaining a network operation parameter of the second time granularity in a fourth time period before the current time as the target broadband network operation parameter.

4. The method of claim 3, wherein, The method further comprises: in a case where the user type is the second complaint frequency user and the user type is a user randomly selected by daily inspection, obtaining a network operation parameter of the second time granularity in a fifth time period before the current time as the target broadband network operation parameter.

5. The method of claim 1, wherein, The target broadband network operation parameters include: home side network operation parameters, service side network operation parameters, access side network operation parameters and metropolitan area network side network operation parameters; the target broadband network operation parameters are input into the target quality difference boundary model to obtain a quality difference boundary result, including: The home side network operation parameters, the service side network operation parameters, the access side network operation parameters and the metropolitan area network side network operation parameters are input into the target quality difference boundary model to obtain a quality difference boundary result.

6. The method of claim 1, wherein, Before using the target quality difference boundary model, the method further includes: obtaining sample broadband network operation parameters corresponding to a sample user type; obtaining sample quality difference boundary results corresponding to the sample broadband network operation parameters; training an initial model according to the sample quality difference boundary results, the sample broadband network operation parameters and a focal loss function to obtain the target quality difference boundary model.

7. The method of claim 6, wherein, The obtaining of the sample broadband network operation parameters corresponding to the sample user type includes: collecting original broadband network operation parameters corresponding to a sample user type; calculating the importance of each original broadband network operation parameter by using a random forest; selecting, from the original broadband network operation parameters, the sample broadband network operation parameters whose importance satisfies an importance condition.

8. A broadband network quality demarcation apparatus, comprising: It includes: an information acquisition module, configured to acquire complaint work order information of a complaint user in a set period; a type identification module, configured to input the complaint work order information into a user complaint frequency identification model to obtain a user type of the complaint user; a model determination module, configured to determine a target quality difference boundary model corresponding to the user type; in a case where the user type is a first complaint frequency user, a first quality difference boundary model of a first time granularity is determined as the target quality difference boundary model; in a case where the user type is a second complaint frequency user, a second quality difference boundary model of a second time granularity is determined as the target quality difference boundary model; the complaint frequency of the first complaint frequency user is greater than that of the second complaint frequency user, and the first time granularity is less than the second time granularity; a parameter acquisition module, configured to acquire target broadband network operation parameters corresponding to the user type; a quality difference boundary module, configured to input the target broadband network operation parameters into the target quality difference boundary model to obtain a quality difference boundary result.

9. The apparatus of claim 8, wherein, When acquiring the complaint work order information of the complaint user in the set period, the information acquisition module is configured to: acquire at least one of the following complaint work order information of the complaint user in the set period: total complaint times, average complaint time interval, minimum complaint time interval, maximum complaint times of the same complaint fact, average complaint times of the same complaint fact, minimum complaint time interval of the same complaint fact, maximum continuous duration of the same complaint fact, repeated complaint times before processing is completed, and repeated complaint times after processing is completed.

10. The apparatus of claim 8, wherein, When acquiring the target broadband network operation parameters corresponding to the user type, the parameter acquisition module is configured to: In a case where the user type is a first complaint frequency user and the current complaint of the complaint user corresponds to complaint analysis, a network operation parameter of a first time granularity in a first time period before a user complaint time is acquired as the target broadband network operation parameter; In a case where the user type is the first complaint frequency user and the current complaint of the complaint user corresponds to evaluation analysis after complaint processing, a network operation parameter of the first time granularity in a second time period before a current time is acquired as the target broadband network operation parameter; In a case where the user type is a second complaint frequency user and the current complaint of the complaint user corresponds to complaint analysis, a network operation parameter of a second time granularity in a third time period before the user complaint time is acquired as the target broadband network operation parameter; In a case where the user type is the second complaint frequency user and the current complaint of the complaint user corresponds to evaluation analysis after complaint processing, a network operation parameter of the second time granularity in a fourth time period before the current time is acquired as the target broadband network operation parameter.

11. The apparatus of claim 10, wherein, The parameter acquisition module can also be used for: In a case where the user type is the second complaint frequency user and the user type is a user randomly selected by daily inspection, a network operation parameter of the second time granularity in a fifth time period before the current time is acquired as the target broadband network operation parameter.

12. The apparatus of claim 8, wherein, The target broadband network operation parameter includes a home side network operation parameter, a service side network operation parameter, an access side network operation parameter and a metropolitan area network side network operation parameter; when the target broadband network operation parameter is input into the target quality difference boundary determination model to obtain a quality difference boundary determination result, the quality difference boundary determination module is configured to: input the home side network operation parameter, the service side network operation parameter, the access side network operation parameter and the metropolitan area network side network operation parameter into the target quality difference boundary determination model to obtain the quality difference boundary determination result.

13. The apparatus of claim 8, wherein, Before using the target quality difference boundary determination model, the quality difference boundary determination module can also be used for: acquiring sample broadband network operation parameters corresponding to a sample user type; acquiring sample quality difference boundary determination results corresponding to the sample broadband network operation parameters; training an initial model according to the sample quality difference boundary determination results, the sample broadband network operation parameters and a focal loss function to obtain the target quality difference boundary determination model.

14. The apparatus of claim 13, wherein, When acquiring sample broadband network operation parameters corresponding to a sample user type, the quality difference boundary determination module is configured to: collect original broadband network operation parameters corresponding to the sample user type; calculate an importance degree of each original broadband network operation parameter by using a random forest; select the sample broadband network operation parameters whose importance degrees satisfy an importance degree condition from the original broadband network operation parameters.

15. An electronic device, comprising: comprise: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement each step in the method of any one of claims 1-7.

16. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program, which when executed by the processor, implements the steps of the method of any one of claims 1-7.

17. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instructions, which when executed by the processor, implements the steps of the method of any one of claims 1-7.

Images