Complaint early warning method and device, electronic equipment, storage medium and product

Abstract

The invention provides a complaint early warning method and device, electronic equipment, a storage medium and a product, and the method comprises the steps: obtaining signaling XDR data, and recognizing an abnormal behavior event that a user restarts a terminal based on the signaling XDR data; comparing the service awareness indexes of the users before and after the abnormal behavior event occurs, and determining the users with unimproved service awareness indexes as high complaint risk users; performing clustering analysis on the abnormal behavior events to obtain a plurality of candidate cells, and determining the candidate cells with the cell performance indexes smaller than a preset performance index threshold as high complaint risk cells; and performing complaint early warning based on the high-complaint-risk user and the high-complaint-risk cell. According to the invention, the abnormal behavior of restarting the terminal by the user is associated with the network problem to carry out high complaint risk analysis, the high complaint risk user and the high complaint risk network area are identified, and complaint early warning is carried out, so that the analysis before complaint is realized, the network problem can be timely and comprehensively found, and the complaint rate of the user is reduced.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a complaint warning method, device, electronic equipment, storage medium and product. Background Art

[0002] With the continuous development of technology, users can use data and voice services through mobile networks. In order to improve user experience and reduce user complaint rates, operators usually adopt a variety of methods such as network performance indicator early warning, indicator degradation demarcation and positioning analysis, user satisfaction surveys, and user complaint diagnosis and analysis to achieve network quality problem analysis, discovery and optimization processing.

[0003] However, existing complaint warning schemes have difficulty identifying a variety of potential complaint risk issues, resulting in reduced user experience and, in turn, a high user complaint rate. Summary of the invention

[0004] The present invention provides a complaint warning method, device, electronic device, storage medium and product, which are used to solve the defect that it is difficult to identify multiple potential complaint risk problems in the prior art, resulting in reduced user experience, and to achieve the identification and timely warning of high complaint risk users and high complaint risk communities, so as to timely and comprehensively discover network problems and reduce user complaint rates.

[0005] The present invention provides a complaint early warning method, comprising: Acquire signaling XDR data, and identify an abnormal behavior event of a user restarting a terminal based on the signaling XDR data; Comparing the service perception indicators of users before and after the abnormal behavior event occurs, and determining that users whose service perception indicators have not improved are high complaint risk users; Performing cluster analysis on the abnormal behavior events to obtain multiple candidate cells, and determining that the candidate cells whose cell performance indicators are less than a preset performance indicator threshold are high complaint risk cells; Complaint warnings are issued based on the high complaint risk users and the high complaint risk cells.

[0006] According to a complaint warning method provided by the present invention, the identifying an abnormal behavior event of a user restarting a terminal based on the signaling XDR data includes: Based on the signaling XDR data, identifying a behavior event of a user restarting a terminal; If the user's service perception index before the behavior event occurs is less than a preset perception index threshold, the behavior event is determined to be an abnormal behavior event; If the occurrence frequency of the behavior event exceeds a preset frequency threshold, the behavior event is determined to be an abnormal behavior event.

[0007] According to a complaint warning method provided by the present invention, the identifying a behavior event of a user restarting a terminal based on the signaling XDR data includes: If the description field information of the first target event in the signaling XDR data meets the shutdown event identification condition, outputting the shutdown event information based on the basic field information of the first target event; If the description field information of the second target event in the signaling XDR data meets the power-on event identification condition, output the power-on event information based on the basic field information of the second target event; According to the information of the shutdown event and the information of the startup event, a behavior event of the user restarting the terminal is obtained.

[0008] According to a complaint warning method provided by the present invention, the information of the shutdown event includes at least a user number and a shutdown time; if the description field information of the second target event in the signaling XDR data meets the power-on event identification condition, then based on the basic field information of the second target event, outputting the power-on event information includes: According to the user number and shutdown time of the shutdown event, searching the signaling XDR data for a second target event of the user corresponding to the user number after the shutdown time; If the description field information of the second target event meets the power-on event identification condition, the power-on event information is output based on the basic field information of the second target event.

[0009] According to a complaint warning method provided by the present invention, the comparison of the service perception indicators of users before and after the abnormal behavior event occurs, and determining that users whose service perception indicators have not improved are high complaint risk users, includes: Calculate the year-on-year comparison of the service perception index based on the user service perception index within a first preset time range before the abnormal behavior event occurs and within a second preset time range after the abnormal behavior event occurs; If the service perception indicator is within a preset range on a month-on-month basis, the user is determined to be a high complaint risk user.

[0010] According to a complaint warning method provided by the present invention, clustering analysis is performed on the abnormal behavior events to obtain multiple candidate cells, and determining that the candidate cells whose cell performance indicators are less than a preset performance indicator threshold are high complaint risk cells includes: Based on the time dimension and the cell dimension, cluster analysis is performed on the abnormal behavior events to obtain multiple candidate cells; For each candidate cell, according to the number of abnormal behavior events and the cell ID of the candidate cell, associating the cell performance indicator of the candidate cell; If the cell performance index of the candidate cell is less than a preset performance index threshold, the cell is determined to be a high complaint risk cell.

[0011] The present invention also provides a complaint warning device, comprising: An acquisition module, used to acquire signaling XDR data, and identify an abnormal behavior event of a user restarting a terminal based on the signaling XDR data; A comparison module, used to compare the service perception indicators of users before and after the abnormal behavior event occurs, and determine that users whose service perception indicators have not improved are high complaint risk users; A clustering module, used to perform cluster analysis on the abnormal behavior events to obtain multiple candidate cells, and determine that the candidate cells whose cell performance indicators are less than a preset performance indicator threshold are high complaint risk cells; The early warning module is used to issue complaint early warning based on the high complaint risk users and the high complaint risk cells.

[0012] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the complaint warning method described above is implemented.

[0013] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the complaint warning method as described in any one of the above is implemented.

[0014] The present invention also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the computer program implements any of the above-mentioned complaint warning methods.

[0015] The complaint warning method, device, electronic device, storage medium and product provided by the present invention can realize the output of two types of warnings, namely, high complaint risk users and high complaint risk cells, by identifying abnormal behavior events of users restarting terminals based on signaling XDR data, comparing the service perception indicators of users before and after the abnormal behavior events, determining that users whose service perception indicators have not improved are high complaint risk users, and clustering analysis of abnormal behavior events to obtain multiple candidate cells, and determining that candidate cells whose cell performance indicators are less than preset performance indicator thresholds are high complaint risk cells. The present invention associates the abnormal behavior of users restarting terminals with network problems for high complaint risk analysis. On the one hand, users with poor service experience perception can be included in the potential complaint analysis scope, and high complaint risk users can be identified. On the other hand, potential problem network areas can be discovered and network areas with high complaint risks can be identified therefrom, thereby realizing pre-complaint analysis, which is conducive to timely and comprehensive discovery of network problems and reducing user complaint rates. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 This is one of the flow charts of the complaint warning method provided by the embodiment of the present invention.

[0018] Figure 2 This is the second flow chart of the complaint warning method provided by the embodiment of the present invention.

[0019] Figure 3 It is a flowchart of identifying abnormal terminal restart events provided by an embodiment of the present invention.

[0020] Figure 4 It is a structural diagram of a complaint warning device provided by an embodiment of the present invention.

[0021] Figure 5 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] In recent years, the use of DPI (Deep Packet Inspection) technology (i.e., behavioral pattern recognition technology) to achieve user behavior analysis has received increasing attention. Before implementing behavioral pattern recognition technology, operators usually study the various behaviors of end users and establish a behavior recognition model based on this. Based on the behavior recognition model, the user's ongoing or upcoming actions can be judged based on the user's existing behaviors.

[0024] At present, operators can use DPI technology to analyze user behavior characteristics to achieve differentiated services. Using DPI technology and cluster crawler technology, accurate analysis of LTE (Long Term Evolution, the fourth generation of mobile communication network) network user behavior based on protocol decoding and XDR (Extended Data Record, service call / transaction detail record) synthesis is achieved. The user behavior analysis model based on DPI technology is mainly divided into the following modules: (1) Data collection and sampling module: mainly responsible for implementing the interface with the user system and collecting user data.

[0025] (2) Data deep analysis module: mainly responsible for deep analysis of the user's HTTP (Hypertext Transfer Protocol).

[0026] (3) User ID (Identity document) definition library: Mainly responsible for generating a unique ID for a user and performing hash conversion based on the user's IP (Internet Protocol) address to keep the user's identity information confidential.

[0027] (4) User behavior definition library: defines known user behaviors with obvious characteristics.

[0028] (5) User behavior classification: The data input by the data deep-level module can be classified using the SVM (Support Vector Machine) algorithm, and combined with the user behavior definition library, a comprehensive analysis can be performed to obtain the user behavior classification. Through in-depth analysis of users, strategies can be formulated to enhance user perception and improve user satisfaction.

[0029] When users use data and voice services through mobile networks, in order to improve service perception, user experience and user satisfaction and reduce user complaint rates, operators usually adopt a variety of methods such as network performance indicator early warning, indicator degradation demarcation and positioning analysis, user satisfaction surveys, and user complaint diagnosis and analysis to analyze, discover and optimize network quality problems.

[0030] Among them, network performance indicator warning (hereinafter referred to as solution A) is based on network element-level performance indicator data, such as the cell-level attachment success rate indicator, and judges network problems according to indicator thresholds. User complaint analysis and diagnosis (hereinafter referred to as solution B) extracts the user number, problem time and location information of the complaint ticket, evaluates user service perception indicators based on historical DPI data, and analyzes and discovers network problems in combination with the poor quality network element library, terminal library, and user contract data.

[0031] However, Plan A and Plan B have the following shortcomings: (1) Shortcomings of Solution A: Since the granularity of the underlying data source - network management performance data - is only at the network element or network area level, it reflects the average service quality of all users in the entire network element. It is impossible to identify potential users with poor service quality and complain, and does not support proactive prevention of complaints.

[0032] (2) Shortcomings of Solution B: After a problem occurs, historical data is retrospectively analyzed, and only a portion of dissatisfied users (i.e., users who have made explicit complaints) are analyzed. This has limitations such as time lag, passive analysis, and inability to reduce the complaint rate. The details are as follows: (a) Post-analysis, poor timeliness: There is a lag from user experience degradation to complaint generation. Problems with existing solutions are discovered late and resolved late, resulting in an increase in complaints from other users in the poor quality network area. In addition, network problems change dynamically, with randomness and volatility. Service quality is related to network capacity load. Service quality deteriorates during busy hours, while network load changes dynamically. Post-analysis cannot be used for early warning analysis.

[0033] (b) Only analyzing explicitly dissatisfied users: Only the complaining users (and the network areas where their services are generated) are analyzed. However, the proportion of complaining users is very small, and the network problems discovered by existing solutions have limitations.

[0034] In response to the above problems, the embodiments of the present invention perform high complaint risk analysis by associating the abnormal behavior of users restarting their terminals with network problems. On the one hand, users with poor service experience perception can be included in the potential complaint analysis scope, and high complaint risk users can be identified. On the other hand, potential problem network areas can be discovered and network areas with high complaint risks can be identified therefrom, thereby achieving pre-complaint analysis, which is conducive to timely and comprehensive discovery of network problems and reducing user complaint rates.

[0035] Figure 1 This is one of the flow charts of the complaint warning method provided by the embodiment of the present invention. Figure 1 The embodiment of the present invention provides a complaint warning method, which may specifically include the following steps: Step 110: Acquire signaling XDR data, and identify an abnormal behavior event of a user restarting the terminal based on the signaling XDR data.

[0036] It should be noted that the execution subject of the complaint warning method provided in the embodiment of the present invention may be an electronic device, a component in an electronic device, an integrated circuit or a chip. The electronic device may be a mobile electronic device or a non-mobile electronic device. For example, the mobile electronic device may be a mobile phone, a tablet computer, a laptop computer, a PDA, a wearable device, an ultra-mobile personal computer (UMPC), a netbook or a personal digital assistant (PDA), etc., and the non-mobile electronic device may be a server, a network attached storage (NAS), a personal computer (PC), a television (TV), a teller machine or a self-service machine, etc., and the embodiment of the present invention does not specifically limit this.

[0037] The signaling XDR data may refer to an extended data record that records network signaling information in a mobile communication network.

[0038] There is a possibility of complaints from users who have restarted their terminals. In an embodiment of the present invention, signaling XDR data can be obtained, and abnormal behavior events of users restarting their terminals can be identified from multiple events recorded in the signaling XDR data, so that the occurrence of abnormal behavior events of users restarting their terminals can be used as a signal that a problem has occurred in the network, and the abnormal behavior of users restarting their terminals can be associated with network problems to perform potential complaint risk analysis, thereby achieving pre-complaint analysis, and more comprehensive network optimization, improving user service perception, and thus reducing user complaint rates.

[0039] Step 120 , comparing the service perception indicators of the users before and after the abnormal behavior event occurs, and determining that the users whose service perception indicators have not improved are high complaint risk users.

[0040] When the user's service experience is degraded, the user usually performs self-repair before complaining. If the problem is not solved, the user may complain. Among them, the user's service experience deterioration may include slow data service access, video freezes, and voice service dropped calls. Among them, the means of self-repair may include reopening the application, disconnecting and reconnecting, and even restarting the terminal.

[0041] When the terminal is restarted, a series of network signaling interaction processes may be generated, such as cell reselection, network re-registration, UE (User Equipment) context re-establishment, bearer resource reallocation establishment, etc. If the service quality is significantly improved after the terminal is restarted, it may reflect that there are problems with the network (including network-terminal compatibility issues).

[0042] The embodiment of the present invention compares the service perception indicators before and after the abnormal behavior event of the user restarting the terminal, and determines the users whose service perception indicators have not improved as high complaint risk users, thereby including users with poor service experience perception into the potential complaint analysis scope, which is conducive to more comprehensive discovery of potential network problems and reducing user complaint rates.

[0043] Step 130 , cluster analysis is performed on the abnormal behavior events to obtain multiple candidate cells, and candidate cells whose cell performance indicators are less than a preset performance indicator threshold are determined as high complaint risk cells.

[0044] A cell may refer to a user distribution area where an abnormal terminal restart event occurs. In an embodiment of the present invention, by clustering abnormal terminal restart events into cells, characteristics of the distribution area of ​​abnormal terminal restart users can be effectively collected, and users with similar abnormal terminal restart events in the same cell can be obtained according to the clustering results, and user indicators in each area can be analyzed.

[0045] Specifically, it is possible to analyze whether each candidate cell is a cell with a high complaint risk based on the cell performance index of each candidate cell. If the cell performance index of a candidate cell is less than a preset performance index threshold, the candidate cell can be determined as a cell with a high complaint risk.

[0046] The embodiment of the present invention can discover potential problem network areas by clustering analysis of abnormal behavior events of users restarting their terminals. By further combining cell performance indicators for secondary screening, network areas with high complaint risks are determined from potential problem network areas, thereby realizing pre-complaint analysis, which is conducive to timely discovery of network problems and thus reducing user complaint rates.

[0047] Step 140: Produce complaint warning based on the high complaint risk users and the high complaint risk cells.

[0048] Specifically, the embodiments of the present invention can output two types of warnings: high complaint risk users and high complaint risk network areas, thereby supporting complaint warnings, so that operators can proactively care for high complaint risk users and optimize networks in high complaint risk network areas, thereby improving network quality and enhancing user service perception, thereby reducing user complaint rates.

[0049] Figure 2 This is the second flow chart of the complaint warning method provided by the embodiment of the present invention. Figure 2 In an embodiment of the present invention, after the abnormal terminal restart event is identified, user analysis and cluster analysis can be performed based on the abnormal terminal restart event.

[0050] When conducting user analysis, you can compare the user's business perception indicators before and after the abnormal behavior event occurs. If the business perception indicators improve (indicating that the user experience has improved), the complaint warning analysis ends; if the business perception indicators do not improve (indicating that the user experience has not improved), the users whose business perception indicators have not improved are identified as high-complaint risk users, and high-complaint risk user warnings are issued, and active care is provided to high-complaint risk users.

[0051] When performing cluster analysis, abnormal behavior events can be clustered based on time and area dimensions, and associated with cell-level performance indicators for secondary screening, thereby identifying high complaint risk cells, issuing early warnings for high complaint risk areas, and optimizing the network in high complaint risk areas.

[0052] The embodiment of the present invention identifies abnormal behavior events of users restarting their terminals based on signaling XDR data, compares the service perception indicators of users before and after the abnormal behavior events, determines that users whose service perception indicators have not improved are high complaint risk users, and performs cluster analysis on abnormal behavior events to obtain multiple candidate cells, and determines that candidate cells whose cell performance indicators are less than a preset performance indicator threshold are high complaint risk cells, which can achieve the output of two types of early warnings: high complaint risk users and high complaint risk cells. The embodiment of the present invention associates the abnormal behavior of users restarting their terminals with network problems for high complaint risk analysis. On the one hand, users with poor service experience perception can be included in the potential complaint analysis scope and high complaint risk users can be identified. On the other hand, potential problem network areas can be discovered and network areas with high complaint risks can be identified, thereby achieving pre-complaint analysis, which is conducive to timely and comprehensive discovery of network problems and reducing user complaint rates.

[0053] Based on any of the above embodiments, the identifying the abnormal behavior event of the user restarting the terminal based on the signaling XDR data may specifically include: Step 111: Identify a behavior event of a user restarting the terminal based on the signaling XDR data.

[0054] In the embodiment of the present invention, the behavior event of the user restarting the terminal can be identified based on the signaling XDR data, and then the abnormal behavior event can be determined according to the user service experience and the terminal restart frequency before the terminal restart behavior event.

[0055] Step 112: If the service perception index of the user before the behavior event occurs is less than a preset perception index threshold, the behavior event is determined to be an abnormal behavior event.

[0056] In the embodiment of the present invention, the user service perception within a period of time before the terminal restart event occurs may be evaluated based on the user plane XDR data.

[0057] Specifically, at least one service perception indicator of the user before the behavior event occurs can be compared with the corresponding perception indicator thresholds. If at least one service perception indicator is less than the corresponding perception indicator threshold, that is, any service perception indicator is less than the corresponding perception indicator threshold, indicating that the user service experience before the restart is poor, it can be determined that the terminal restart behavior event is caused by the user's service perception degradation, and the terminal restart behavior event is marked as a terminal restart abnormal behavior event of type 1.

[0058] Among them, the service perception indicators may include large packet download rate, small packet delay, wireless side TCP handshake delay, etc.

[0059] Step 113: If the occurrence frequency of the behavior event exceeds a preset frequency threshold, the behavior event is determined to be an abnormal behavior event.

[0060] In an embodiment of the present invention, regardless of whether user plane XDR data is generated before restart and whether service perception is degraded, if the frequency of terminal restart behavior events exceeds a preset frequency threshold, indicating that the user terminal is frequently restarted, the terminal restart behavior event can be marked as a type 2 terminal restart abnormal behavior event.

[0061] The preset frequency threshold may be set according to actual needs, such as 5 times per hour, and the present invention is not limited thereto.

[0062] Figure 3 1 is a flow chart of identifying abnormal terminal restart events provided by an embodiment of the present invention. Figure 3 After identifying the shutdown event, it can be determined whether there is a startup event within the window time. If there is a startup event within the window time, the user service perception in the period before the terminal restart event occurs can be evaluated.

[0063] If the user service perception is poor before the restart, it can be determined as an abnormal behavior event of the terminal restart; if the user service perception is good before the restart, it can be determined whether the user terminal restarts frequently. If the user terminal restarts frequently, it can be determined as an abnormal behavior event of the terminal restart.

[0064] The embodiment of the present invention identifies the terminal restart behavior event based on the signaling XDR data, and then determines the abnormal behavior event according to the user service perception and the terminal restart frequency before the terminal restart behavior event, so as to associate the abnormal behavior event of the user restarting the terminal with the network problem for high complaint risk analysis, which can realize pre-complaint analysis and network optimization, and is conducive to reducing the user complaint rate.

[0065] Based on any of the foregoing embodiments, identifying a behavior event of a user restarting a terminal based on the signaling XDR data may specifically include: Step 1111: if the description field information of the first target event in the signaling XDR data meets the shutdown event identification condition, output the shutdown event information based on the basic field information of the first target event.

[0066] In an embodiment of the present invention, it can be determined whether the description field information of the first target event in the signaling XDR data meets the shutdown event identification condition. The description field information can at least include an interface type field, a process type field, a detach type keyword field, and a trigger mode keyword field.

[0067] According to the 3GPP (3rd Generation Partnership Project, standardization organization) specification, the 4th bit of the first byte (Octet) of the detach type information field (detach type IE) in the detach request message from the user terminal device to the network can indicate the detach type. 1 can indicate switch off. That is, the detach request message initiated by the UE due to shutdown must follow the above message coding.

[0068] Specifically, the shutdown event identification conditions can be as follows: interface type = 5 (i.e. S1-MME, Mobility Management Entity, mobile management entity) and process type = 6 (i.e. detachment) and bit 4 of keyword 1 (i.e. detachment type) is 1 (i.e. switch off) and keyword 2 (i.e. trigger mode) = 1 (i.e. initiated by the user terminal device).

[0069] In the embodiment of the present invention, if the description field information of the first target event in the signaling XDR data meets the shutdown event identification condition, the shutdown event information can be output based on the basic field information of the first target event. The basic field information can include at least a time field, a user number field, and a cell field, and the shutdown event information can include at least a shutdown time, a user number, and a shutdown cell.

[0070] Specifically, for an XDR that matches the shutdown event identification condition, the following shutdown event information may be output, as shown in Table 1: Table 1

[0071] Step 1112: If the description field information of the second target event in the signaling XDR data meets the power-on event identification condition, output the power-on event information based on the basic field information of the second target event.

[0072] In the embodiment of the present invention, it can be determined whether the description field information of the second target event in the signaling XDR data meets the shutdown event identification condition. The description field information can at least include an interface type field, a process type field, a user number field and a time field.

[0073] Specifically, the power-on event identification condition may be as follows: interface type=5 (S1-MME) and procedure type=1 (attachment) and msisdn=power-off event user number and (start_time-power-off time)<60 seconds.

[0074] In an embodiment of the present invention, if the description field information of the second target event in the signaling XDR data meets the power-on event identification condition, the power-on event information can be output based on the basic field information of the second target event. The basic field information may include a time field, a user number field, and a cell field, and the power-on event information may include at least a power-on event, a user number, and a power-on cell.

[0075] Specifically, for an XDR that matches the power-on event identification condition, the following power-on event information may be output, as shown in Table 2: Table 2

[0076] Step 1113: Obtain a behavior event of the user restarting the terminal according to the information of the shutdown event and the information of the startup event.

[0077] The information of the behavior event of the user restarting the terminal may at least include the user number, the shutdown time, the startup event, the shutdown cell and the startup cell.

[0078] In the embodiment of the present invention, a terminal restart behavior event can be formed based on a pair of information related to a shutdown event and a power-on event, and the following information of the user restarting the terminal behavior event can be output, as shown in Table 3: Table 3

[0079] In an embodiment of the present invention, after identifying a terminal restart abnormal event of type 1 and a terminal restart abnormal event of type 2, the terminal restart abnormal events of type 1 and type 2 may be merged to output abnormal behavior event information of terminal restart, as shown in Table 4: Table 4

[0080] The embodiment of the present invention identifies the terminal restart behavior event by utilizing the detachment request message initiated by the user terminal device due to shutdown and the message coding of startup, realizes the identification of terminal restart behavior events based on XDR data and then associates network problems to perform high complaint risk analysis, thereby realizing pre-complaint analysis, which is conducive to reducing user complaint rates.

[0081] Based on any of the above embodiments, the information of the shutdown event includes at least a user number and a shutdown time; if the description field information of the second target event in the signaling XDR data meets the power-on event identification condition, then based on the basic field information of the second target event, outputting the information of the power-on event may specifically include: Step 11121, according to the user number and shutdown time of the shutdown event, searching the signaling XDR data for a second target event of the user corresponding to the user number after the shutdown time; Step 11122: if the description field information of the second target event meets the power-on event identification condition, output the power-on event information based on the basic field information of the second target event.

[0082] In the embodiment of the present invention, after outputting the information of the shutdown event, it is possible to search, based on the user number and the shutdown time of the shutdown event, whether the user corresponding to the user number has an attach process within a certain time after the shutdown time, for example, within 1 minute. The first attach process that matches is the power-on event.

[0083] Specifically, the power-on event within a certain period of time after the shutdown time can be searched. If there is a second target event whose description field information meets the power-on event identification condition, the second target event can be determined to be a power-on event associated with the shutdown event. Based on the basic field information of the second target event, the power-on event information can be output.

[0084] The embodiment of the present invention searches for a power-on event associated with the power-off event according to the user number and power-off time of the power-off event, and then determines a terminal restart event, thereby performing complaint risk analysis based on the terminal restart behavior event and associating network problems, thereby achieving pre-complaint analysis and helping to reduce user complaint rates.

[0085] Based on any of the above embodiments, comparing the service perception indicators of users before and after the abnormal behavior event occurs and determining that users whose service perception indicators have not improved are high complaint risk users may specifically include: Step 121, calculating a year-on-year comparison of a service perception index based on user service perception indexes within a first preset time range before the abnormal behavior event occurs and within a second preset time range after the abnormal behavior event occurs; Step 122: If the service perception index is within a preset range on a month-on-month basis, the user is determined to be a high complaint risk user.

[0086] In an embodiment of the present invention, for a terminal restart abnormal event of type 1, it can be determined whether the user service perception index after the restart is significantly improved compared to before the restart. If there is no obvious improvement, the user can be determined as a high complaint risk user.

[0087] Specifically, based on the user plane XDR data, the user service perception within a first preset time range before the abnormal terminal restart event occurs and within a second preset time range after the abnormal terminal restart event occurs, that is, within a certain period of time before and after the abnormal terminal restart event, is evaluated, and each service perception indicator is calculated on a month-on-month basis. If at least one service perception indicator is within the corresponding preset month-on-month range, it means that at least one service perception indicator has not been significantly improved, and the user can be determined to be a high complaint risk user.

[0088] The service perception indicators may include at least the large packet download rate, small packet delay, wireless side TCP handshake delay, etc. of the service. The preset month-on-month range may be set to 90%-110%. The first preset time range and the second preset time range may be the same or different, for example, both may be set to 1h.

[0089] In the embodiment of the present invention, for type 2 terminal restart abnormal events, that is, terminals that restart frequently, they can be directly identified as high complaint risk users.

[0090] The related technical solution is to conduct complaint analysis through users with poor experience + unresolved problems + complained users, that is, users with explicit poor experience, which makes it difficult to discover potential complaint risks; the embodiment of the present invention uses user service perception indicators to identify users with poor experience + self-service resolution + no complaints, that is, users with implicit poor experience. By incorporating users with implicit poor experience into the scope of complaint risk analysis, the scope of analyzed users can be expanded, which is conducive to more comprehensive discovery of potential network problems.

[0091] Based on any of the above embodiments, clustering analysis of the abnormal behavior events to obtain multiple candidate cells, and determining that a candidate cell whose cell performance index is less than a preset performance index threshold is a high complaint risk cell may specifically include: Step 131, performing cluster analysis on the abnormal behavior events based on the time dimension and the cell dimension to obtain multiple candidate cells; Step 132, for each candidate cell, correlating a cell performance indicator of the candidate cell according to the number of abnormal behavior events and the cell ID of the candidate cell; Step 133: If the cell performance index of the candidate cell is less than a preset performance index threshold, the cell is determined to be a high complaint risk cell.

[0092] Specifically, a two-dimensional coordinate system can be constructed based on the time dimension and the cell dimension, abnormal behavior events can be mapped into the two-dimensional coordinate system, and cluster analysis can be performed on the abnormal behavior events in the two-dimensional coordinate system to obtain multiple candidate cells with complaint risks.

[0093] Specifically, the cell performance indicators, such as coverage, interference, capacity load and other cell-level indicators, can be associated with the number of abnormal behavior events and cell ID of the candidate cell. When at least one cell performance indicator is lower than the corresponding performance indicator threshold, it means that the cell not only has a high incidence of abnormal terminal restart events, but also has degraded cell performance, and the cell can be judged as a high complaint risk cell.

[0094] The results of cell clustering based on abnormal terminal restart events may be affected by user behavior factors; and judging based on the threshold of network element performance indicator degradation is also difficult to fully and truly reflect user perception. The embodiment of the present invention combines the results of the two analyses and performs secondary screening to ultimately determine the cells and network areas with high complaint risks, which is conducive to comprehensively discovering network problems and reducing user complaint rates.

[0095] Existing technical solutions usually conduct complaint risk analysis based on traditional user perception indicators, such as rate, latency, success rate, etc. When the user service experience deteriorates, such as slow Internet access, video freezes, voice drops, etc., users usually perform simple self-repairs such as restarting the terminal. The embodiment of the present invention incorporates abnormal user behavior events into network quality analysis, associates the user's self-troubleshooting behavior with network problems, and compares the service quality indicators before and after shutdown and after startup through XDR, or the frequency of abnormal events within a certain time sliding window, identifies high complaint risk users, and identifies high complaint risk areas based on time and cell clustering of terminal restart abnormal events, which can achieve pre-complaint analysis, facilitate timely and comprehensive discovery of network problems, and reduce user complaint rates.

[0096] The complaint warning device provided by the present invention is described below. The complaint warning device described below and the complaint warning method described above can be referenced to each other.

[0097] Figure 4 Schematic diagram of the structure of the complaint warning device provided by the embodiment of the present invention. Figure 3 The embodiment of the present invention provides a complaint warning device, which may specifically include the following modules: An acquisition module 410 is used to acquire signaling XDR data, and identify an abnormal behavior event of a user restarting a terminal based on the signaling XDR data; A comparison module 420 is used to compare the service perception indicators of users before and after the abnormal behavior event occurs, and determine that users whose service perception indicators have not improved are high complaint risk users; A clustering module 430 is used to perform cluster analysis on the abnormal behavior events to obtain multiple candidate cells, and determine that the candidate cells whose cell performance indicators are less than a preset performance indicator threshold are high complaint risk cells; The early warning module 440 is used to issue a complaint early warning based on the high complaint risk users and the high complaint risk cells.

[0098] In an optional embodiment, the acquisition module 410 is specifically configured to: Based on the signaling XDR data, identifying a behavior event of a user restarting a terminal; If the user's service perception index before the behavior event occurs is less than a preset perception index threshold, the behavior event is determined to be an abnormal behavior event; If the occurrence frequency of the behavior event exceeds a preset frequency threshold, the behavior event is determined to be an abnormal behavior event.

[0099] In an optional embodiment, the acquisition module 410 is specifically configured to: If the description field information of the first target event in the signaling XDR data meets the shutdown event identification condition, outputting the shutdown event information based on the basic field information of the first target event; If the description field information of the second target event in the signaling XDR data meets the power-on event identification condition, output the power-on event information based on the basic field information of the second target event; According to the information of the shutdown event and the information of the startup event, a behavior event of the user restarting the terminal is obtained.

[0100] In an optional embodiment, the information of the shutdown event includes at least a user number and a shutdown time; the acquisition module 410 is specifically used to: According to the user number and shutdown time of the shutdown event, searching the signaling XDR data for a second target event of the user corresponding to the user number after the shutdown time; If the description field information of the second target event meets the power-on event identification condition, the power-on event information is output based on the basic field information of the second target event.

[0101] In an optional embodiment, the comparison module 420 is specifically used for: Calculate the year-on-year comparison of the service perception index based on the user service perception index within a first preset time range before the abnormal behavior event occurs and within a second preset time range after the abnormal behavior event occurs; If the service perception indicator is within a preset range on a month-on-month basis, the user is determined to be a high complaint risk user.

[0102] In an optional embodiment, the clustering module 430 is specifically configured to: Based on the time dimension and the cell dimension, cluster analysis is performed on the abnormal behavior events to obtain multiple candidate cells; For each candidate cell, according to the number of abnormal behavior events and the cell ID of the candidate cell, associating the cell performance indicator of the candidate cell; If the cell performance index of the candidate cell is less than a preset performance index threshold, the cell is determined to be a high complaint risk cell.

[0103] The embodiment of the present invention identifies abnormal behavior events of users restarting their terminals based on signaling XDR data, compares the service perception indicators of users before and after the abnormal behavior events, determines that users whose service perception indicators have not improved are high complaint risk users, and performs cluster analysis on abnormal behavior events to obtain multiple candidate cells, and determines that candidate cells whose cell performance indicators are less than a preset performance indicator threshold are high complaint risk cells, which can achieve the output of two types of early warnings: high complaint risk users and high complaint risk cells. The embodiment of the present invention associates the abnormal behavior of users restarting their terminals with network problems for high complaint risk analysis. On the one hand, users with poor service experience perception can be included in the potential complaint analysis scope and high complaint risk users can be identified. On the other hand, potential problem network areas can be discovered and network areas with high complaint risks can be identified, thereby achieving pre-complaint analysis, which is conducive to timely and comprehensive discovery of network problems and reducing user complaint rates.

[0104] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5 As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530 and a communication bus 540, wherein the processor 510, the communication interface 520 and the memory 830 complete mutual communication through the communication bus 540. The processor 510 may call the logic instructions in the memory 530 to execute the complaint warning method, which includes: obtaining signaling XDR data, identifying the abnormal behavior event of the user restarting the terminal based on the signaling XDR data; comparing the service perception index of the user before and after the abnormal behavior event occurs, and determining that the user whose service perception index has not improved is a high complaint risk user; performing cluster analysis on the abnormal behavior event to obtain multiple candidate cells, and determining that the candidate cell whose cell performance index is less than the preset performance index threshold is a high complaint risk cell; and performing complaint warning based on the high complaint risk user and the high complaint risk cell.

[0105] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0106] On the other hand, the present invention also provides a computer program product, which includes a computer program, and the computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the complaint warning method provided by the above methods, which includes: obtaining signaling XDR data, and identifying abnormal behavior events of users restarting their terminals based on the signaling XDR data; comparing the service perception indicators of users before and after the abnormal behavior event occurs, and determining that users whose service perception indicators have not improved are high complaint risk users; performing cluster analysis on the abnormal behavior events to obtain multiple candidate cells, and determining that the candidate cells whose cell performance indicators are less than a preset performance indicator threshold are high complaint risk cells; and performing complaint warnings based on the high complaint risk users and the high complaint risk cells.

[0107] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the complaint warning method provided by the above-mentioned methods, the method comprising: obtaining signaling XDR data, and identifying an abnormal behavior event of a user restarting a terminal based on the signaling XDR data; comparing the service perception indicators of the user before and after the abnormal behavior event occurs, and determining that the user whose service perception indicators have not improved is a high complaint risk user; performing a cluster analysis on the abnormal behavior event to obtain multiple candidate cells, and determining that the candidate cell whose cell performance indicator is less than a preset performance indicator threshold is a high complaint risk cell; and performing a complaint warning based on the high complaint risk user and the high complaint risk cell.

[0108] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0109] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A complaint early warning method, characterized in that: include: Acquire signaling XDR data, and identify an abnormal behavior event of a user restarting a terminal based on the signaling XDR data; Comparing the service perception indicators of users before and after the abnormal behavior event occurs, and determining that users whose service perception indicators have not improved are high complaint risk users; Performing cluster analysis on the abnormal behavior events to obtain multiple candidate cells, and determining that the candidate cells whose cell performance indicators are less than a preset performance indicator threshold are high complaint risk cells; Complaint warnings are issued based on the high complaint risk users and the high complaint risk cells.

2. The complaint warning method according to claim 1, characterized in that: The identifying, based on the signaling XDR data, an abnormal behavior event of a user restarting a terminal includes: Based on the signaling XDR data, identifying a behavior event of a user restarting a terminal; If the user's service perception index before the behavior event occurs is less than a preset perception index threshold, the behavior event is determined to be an abnormal behavior event; If the occurrence frequency of the behavior event exceeds a preset frequency threshold, the behavior event is determined to be an abnormal behavior event.

3. The complaint warning method according to claim 2, characterized in that: The identifying, based on the signaling XDR data, a behavior event of a user restarting the terminal includes: If the description field information of the first target event in the signaling XDR data meets the shutdown event identification condition, outputting the shutdown event information based on the basic field information of the first target event; If the description field information of the second target event in the signaling XDR data meets the power-on event identification condition, output the power-on event information based on the basic field information of the second target event; According to the information of the shutdown event and the information of the startup event, a behavior event of the user restarting the terminal is obtained.

4. The complaint warning method according to claim 3, characterized in that: The information of the shutdown event includes at least a user number and a shutdown time; if the description field information of the second target event in the signaling XDR data meets the power-on event identification condition, then based on the basic field information of the second target event, outputting the information of the power-on event includes: According to the user number and shutdown time of the shutdown event, searching the signaling XDR data for a second target event of the user corresponding to the user number after the shutdown time; If the description field information of the second target event meets the power-on event identification condition, the power-on event information is output based on the basic field information of the second target event.

5. The complaint warning method according to claim 1, characterized in that: The comparing the service perception indicators of the users before and after the abnormal behavior event occurs, and determining that the users whose service perception indicators have not improved are high complaint risk users, includes: Calculate the year-on-year comparison of the service perception index based on the user service perception index within a first preset time range before the abnormal behavior event occurs and within a second preset time range after the abnormal behavior event occurs; If the service perception indicator is within a preset range on a month-on-month basis, the user is determined to be a high complaint risk user.

6. The complaint warning method according to claim 1, characterized in that: The clustering analysis of the abnormal behavior events to obtain a plurality of candidate cells, and determining that a candidate cell having a cell performance index less than a preset performance index threshold is a high complaint risk cell, includes: Based on the time dimension and the cell dimension, cluster analysis is performed on the abnormal behavior events to obtain multiple candidate cells; For each candidate cell, according to the number of abnormal behavior events and the cell ID of the candidate cell, associating the cell performance indicator of the candidate cell; If the cell performance index of the candidate cell is less than a preset performance index threshold, the cell is determined to be a high complaint risk cell.

7. A complaint warning device, characterized in that: include: An acquisition module, used to acquire signaling XDR data, and identify an abnormal behavior event of a user restarting a terminal based on the signaling XDR data; A comparison module, used to compare the service perception indicators of users before and after the abnormal behavior event occurs, and determine that users whose service perception indicators have not improved are high complaint risk users; A clustering module, used to perform cluster analysis on the abnormal behavior events to obtain multiple candidate cells, and determine that the candidate cells whose cell performance indicators are less than a preset performance indicator threshold are high complaint risk cells; The early warning module is used to issue complaint early warning based on the high complaint risk users and the high complaint risk cells.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, it implements the complaint warning method as described in any one of claims 1 to 7.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the complaint warning method as described in any one of claims 1 to 7 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the complaint warning method as described in any one of claims 1 to 7 is implemented.

Images