Outbound Call Fault Analysis Method, Device, Computer Equipment and Storage Medium

Abstract

The present application relates to an outbound call failure analysis method, apparatus, computer equipment, and storage medium, and relates to the field of artificial intelligence technology. The method comprises: inputting a time period to be predicted into a preset neural network model, predicting the outbound call failure rate within the time period and the number of outbound call failures corresponding to various fault types caused by faults in the outbound call system itself within the time period; when the outbound call failure rate is greater than a preset first threshold, obtaining a weighted number corresponding to each fault type based on the number of outbound call failures and the preset weighted values corresponding to each fault type; and determining the target fault type of the outbound call failure that occurred within the time period based on the number of outbound call failures and the weighted numbers. This method can accurately locate the fault type of the outbound call failure.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to an outbound call fault analysis method, apparatus, computer equipment, and storage medium. Background Art

[0002] With the development of society, outbound call systems are becoming increasingly common. For example, banks can use outbound call systems to simulate manual phone calls and make calls to a large number of users simultaneously. However, outbound call systems can be blocked due to high concurrency, or even experience system failures that prevent calls from going through. Therefore, analyzing the causes of outbound call failures is crucial for the proper functioning of outbound call systems.

[0003] In traditional technology, the availability of each module that makes up the outbound call system is mainly monitored. For example, when some components of the outbound call system cannot be used, an error will be generated. When certain errors occur within the outbound call system, an error will also be reported. When certain log output error keywords are detected, an alarm message will also be generated.

[0004] However, conventional technologies have the problem of being unable to accurately locate the fault type. Summary of the Invention

[0005] Based on this, it is necessary to provide an outbound call fault analysis method, device, computer equipment and storage medium that can accurately locate the fault type in order to solve the above technical problems.

[0006] In a first aspect, the present application provides a method for analyzing outbound call failures. The method comprises:

[0007] Inputting the time period to be predicted into a preset neural network model, predicting the outbound call failure rate within the time period and the number of outbound call failures corresponding to various types of failures caused by the outbound call system itself within the time period;

[0008] When the outbound call failure rate is greater than a preset first threshold, obtaining a weighted number corresponding to each fault type according to the number of outbound call failures and the preset weight value corresponding to each fault type;

[0009] A target fault type of the outbound call failure occurring within the time period is determined according to the number of outbound call failures and the weighted numbers.

[0010] In one embodiment, determining the target fault type of the outbound call failure occurring within the time period according to the number of outbound call failures and the weighted values includes:

[0011] Determining candidate fault types of outbound call failures occurring within the time period based on the number of outbound call failures;

[0012] Determining the weighted number corresponding to the candidate fault type according to each of the weighted numbers;

[0013] The target fault type is determined from the candidate fault types according to the weighted numbers corresponding to the candidate fault types.

[0014] In one embodiment, determining the target fault type from the candidate fault types according to the weighted numbers corresponding to the candidate fault types includes:

[0015] The candidate fault type whose weighted number is greater than a preset second threshold is determined as the target fault type.

[0016] In one embodiment, the method further comprises:

[0017] When the outbound call failure rate is greater than the preset first threshold, an alarm prompt message is output.

[0018] In one embodiment, the training process of the neural network model includes:

[0019] Obtain the total number of outbound call failures within the sample period and the standard number of outbound call failures corresponding to various types of failures caused by outbound call system failures.

[0020] Determining a standard outbound call failure rate corresponding to the sample time period based on each of the standard outbound call times and the total number of outbound call failures;

[0021] Inputting the sample time period into the initial neural network model, predicting the sample outbound call failure rate within the sample time period and the number of sample outbound call failures corresponding to multiple fault types occurring within the sample time period;

[0022] The initial neural network model is trained according to the standard outbound call failure rate, the number of standard outbound call failures, the sample outbound call failure rate and the number of sample outbound call failures to obtain the neural network model.

[0023] In a second aspect, the present application further provides an outbound call fault analysis device. The device comprises:

[0024] A prediction module, configured to input a time period to be predicted into a preset neural network model, and predict the outbound call failure rate within the time period and the number of outbound call failures corresponding to various types of failures caused by the outbound call system itself within the time period;

[0025] a calculation module, configured to obtain a weighted number corresponding to each fault type according to the number of outbound call failures and the preset weight values corresponding to each fault type when the outbound call failure rate is greater than a preset first threshold;

[0026] The determination module is used to determine the target fault type of the outbound call failure occurring in the time period according to the number of outbound call failures and the weighted numbers.

[0027] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are performed:

[0028] Inputting the time period to be predicted into a preset neural network model, predicting the outbound call failure rate within the time period and the number of outbound call failures corresponding to various types of failures caused by the outbound call system itself within the time period;

[0029] When the outbound call failure rate is greater than a preset first threshold, obtaining a weighted number corresponding to each fault type according to the number of outbound call failures and the preset weight value corresponding to each fault type;

[0030] A target fault type of the outbound call failure occurring within the time period is determined according to the number of outbound call failures and the weighted numbers.

[0031] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps:

[0032] Inputting the time period to be predicted into a preset neural network model, predicting the outbound call failure rate within the time period and the number of outbound call failures corresponding to various types of failures caused by the outbound call system itself within the time period;

[0033] When the outbound call failure rate is greater than a preset first threshold, obtaining a weighted number corresponding to each fault type according to the number of outbound call failures and the preset weight value corresponding to each fault type;

[0034] A target fault type of the outbound call failure occurring within the time period is determined according to the number of outbound call failures and the weighted numbers.

[0035] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the following steps:

[0036] Inputting the time period to be predicted into a preset neural network model, predicting the outbound call failure rate within the time period and the number of outbound call failures corresponding to various types of failures caused by the outbound call system itself within the time period;

[0037] When the outbound call failure rate is greater than a preset first threshold, obtaining a weighted number corresponding to each fault type according to the number of outbound call failures and the preset weight value corresponding to each fault type;

[0038] A target fault type of the outbound call failure occurring within the time period is determined according to the number of outbound call failures and the weighted numbers.

[0039] The above-mentioned outbound call failure analysis method, device, computer equipment and storage medium can accurately predict the outbound call failure rate within the time period and the number of outbound call failures corresponding to various fault types caused by the outbound call system's own faults within the time period by inputting the time period to be predicted into a preset neural network model. Therefore, when the outbound call failure rate is greater than the preset first threshold, the weighted number corresponding to each fault type can be obtained according to the number of outbound call failures corresponding to each fault type and the preset weight value corresponding to each fault type. Then, the target fault type of the outbound call failure occurring within the time period can be determined according to the number of outbound call failures corresponding to each fault type and the weighted number corresponding to each fault type. Compared with the method of monitoring the availability of each module constituting the outbound call system in traditional technology, the above process can determine whether the outbound call failure is caused by the concurrency limit exceeding of the outbound call system, and can accurately locate the target fault type of the outbound call failure caused by the outbound call system's own faults within the time period to be predicted. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a diagram showing an application environment of an outbound call fault analysis method in one embodiment;

[0041] Figure 2 1 is a flow chart of an outbound call fault analysis method according to an embodiment;

[0042] Figure 3 A flowchart of an outbound call fault analysis method according to another embodiment;

[0043] Figure 4 A flowchart of an outbound call fault analysis method according to another embodiment;

[0044] Figure 5 A flowchart of an outbound call fault analysis method according to another embodiment;

[0045] Figure 6 This is a structural block diagram of an outbound call fault analysis device in one embodiment;

[0046] Figure 7 It is a structural block diagram of an outbound call fault analysis device in another embodiment;

[0047] Figure 8 It is a structural block diagram of an outbound call fault analysis device in another embodiment;

[0048] Figure 9 It is a structural block diagram of an outbound call fault analysis device in another embodiment;

[0049] Figure 10 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0051] It should be noted that the outbound call fault analysis method, device, computer equipment and storage medium of the present application can be applied in the field of artificial intelligence technology, and can also be used in other technical fields besides the field of artificial intelligence technology. The present application does not limit the application field of the outbound call fault analysis method, device, computer equipment and storage medium.

[0052] The outbound call fault analysis method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, computer device 104 may be integrated with an outbound calling system, which can be used to simultaneously make calls to multiple terminals 102. Terminals 102 may include, but are not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices may include smart speakers, smart TVs, smart air conditioners, and smart car devices. Portable wearable devices may include smart watches, smart bracelets, and head-mounted devices. Computer device 104 may be implemented as a standalone computer device or a computer device cluster consisting of multiple computer devices.

[0053] Traditionally, outbound call system monitoring typically involves monitoring the availability of each module that makes up the system. For example, if a component becomes unavailable, an error message will be generated. Event monitoring occurs when certain errors occur within the system, such as generating an alarm when certain log output contains incorrect keywords. Furthermore, monitoring of important transactions is also performed, such as the generation of intelligent outbound call recordings. These three distinct monitoring dimensions ensure the normal operation of the intelligent outbound call system. However, traditional outbound call system monitoring cannot accurately pinpoint the type of outbound call failure that has occurred.

[0054] In one embodiment, Figure 2 As shown, a method for analyzing outbound call failures is provided. Figure 1 The computer device in the example is used to illustrate the process, including the following steps:

[0055] S201: Input the time period to be predicted into a preset neural network model to predict the outbound call failure rate within the time period and the number of outbound call failures corresponding to various fault types caused by the outbound call system itself within the time period.

[0056] The neural network model may be a long short-term neural network (LSTM). It is understood that LSTM has a long-term memory function, which can memorize information for a long time, including values of indefinite time lengths, thereby reducing the learning difficulty of the neural network. Optionally, in this embodiment, the first layer of the neural network model is an LSTM layer, and the second layer is a fully connected layer.

[0057] It is understandable that the external system may make outbound calls to multiple terminals simultaneously. Therefore, the outbound call issued by the outbound call system may fail due to the excessive concurrency of the outbound call system, or due to a fault in the outbound call system itself. Generally, if the outbound call fails due to the excessive concurrency of the outbound call system, the call task will be reissued after a certain period of time. For example, in this embodiment, the outbound call failure caused by the outbound call system's own fault can be any of the following: outbound call failure caused by a call routing failure of the outbound call system, outbound call failure caused by the unavailability of the outbound call system's server, and outbound call failure caused by the outbound call system's active release due to exceeding the service control ringing timeout.

[0058] The outbound call failure rate in the predicted time period in this embodiment refers to the ratio of the number of outbound call failures corresponding to various fault types caused by the outbound call system's own faults in the predicted time period to the total number of calls issued by the outbound call system.

[0059] For example, in this embodiment, the time period to be predicted can be 8:00-9:00 on a certain day, or 13:00-15:00 on a certain day. Furthermore, as an example, the number of calls issued by the outbound call system from 10:00-12:00 every day may be relatively large. During this time period, there may be cases where outbound call failures occur due to a large number of concurrent calls. Therefore, in some scenarios, for 10:00-12:00 every day, the server can divide the time period to be predicted into 10:00-10:30, 10:30-11:00, 11:00-11:30, and 11:30-12:00.

[0060] S202 , when the outbound call failure rate is greater than a preset first threshold, obtaining a weighted number corresponding to each fault type according to the number of outbound call failures and the preset weight value corresponding to each fault type.

[0061] The fault type in this embodiment can be determined based on the fault type corresponding to outbound call failures caused by faults in the outbound call system itself during a historical period. The weight value corresponding to each fault type can be used to represent the importance of each fault type.

[0062] Optionally, the weight values corresponding to each fault type can be stored in a weight log in advance, and when the weight values corresponding to the preset fault types need to be calculated, the preset weight values corresponding to each fault type can be directly obtained from the weight log. It is understandable that when multiple fault types caused by faults in the outbound call system itself are updated, the updated fault types can be synchronized to the above-mentioned weight log, even if the weight log stores the updated fault types and the weight values corresponding to the updated fault types. For example, the weight values corresponding to each fault type can be shown in Table 1.

[0063] Table 1

[0064]

[0065] Specifically, the preset first threshold value can be independently determined based on an empirical value. In this embodiment, if the above-mentioned outbound call failure rate exceeds the preset first threshold value, it means that the failure rate caused by the outbound call system itself is too high, which may affect the normal operation of the outbound call system. Therefore, it is necessary to determine the type of outbound call failure that occurred within the time period in order to solve it in a timely manner. Optionally, in this embodiment, the number of outbound call failures corresponding to each fault type and the weight value corresponding to each fault type can be multiplied to obtain the weighted number corresponding to each fault type. For example, assuming that the number of outbound call failures corresponding to a certain fault type is 5 and the weight value corresponding to the fault type is 0.7, the weighted number corresponding to the fault type is 3.5.

[0066] S203: Determine a target fault type of outbound call failures occurring within a time period according to the number of outbound call failures and the weighted values.

[0067] Optionally, in this embodiment, the target fault type for outbound call failures occurring during the aforementioned time period can be determined based on the fault type with the greatest number of outbound call failures and the weight value corresponding to the fault type with the greatest number of outbound call failures. For example, if the fault type with the greatest number of outbound call failures has the largest weight value, then that fault type can be determined as the target fault type. Furthermore, the target fault type for outbound call failures occurring during the time period determined can be stored in a log, facilitating user confirmation of the fault type for outbound call failures occurring during the time period. Exemplarily, in this embodiment, the determined target fault type can be a single fault type or multiple fault types.

[0068] In the above-mentioned outbound call failure analysis method, by inputting the time period to be predicted into the preset neural network model, the outbound call failure rate within the time period and the number of outbound call failures corresponding to various fault types caused by the outbound call system's own faults within the time period can be accurately predicted. Therefore, when the outbound call failure rate is greater than the preset first threshold, the weighted number corresponding to each fault type can be obtained according to the number of outbound call failures corresponding to each fault type and the preset weight value corresponding to each fault type. Then, according to the number of outbound call failures corresponding to each fault type and the weighted number corresponding to each fault type, the target fault type of the outbound call failure occurring within the time period can be determined. Compared with the method of monitoring the availability of each module constituting the outbound call system in traditional technology, the above-mentioned process can determine whether the outbound call failure is caused by the concurrency limit exceeding of the outbound call system, and can accurately locate the target fault type of the outbound call failure caused by the outbound call system's own fault within the time period to be predicted.

[0069] In the above scenario where the target fault type of outbound call failures occurring within the predicted time period is determined based on the number of outbound call failures corresponding to each fault type and the weighted number corresponding to each fault type, candidate fault types may be determined first, and then the target fault type may be determined from the candidate fault types. In one embodiment, Figure 3 As shown, the above S203 includes:

[0070] S301: Determine candidate fault types of outbound call failures occurring within a time period according to the number of outbound call failures.

[0071] Optionally, in this embodiment, the outbound call failure counts corresponding to various fault types caused by faults in the outbound call system during the aforementioned time period may be sorted in reverse order, and the first few fault types in reverse order may be selected as candidate fault types for outbound call failures occurring during the aforementioned time period. For example, the first five fault types sorted in reverse order by call failure count may be determined as candidate fault types. Furthermore, the determined candidate fault types may be stored in the aforementioned log.

[0072] S302: Determine the weighted number corresponding to the candidate fault type according to each weighted number.

[0073] Optionally, in this embodiment, the weighted numbers corresponding to the candidate fault types can be found from the weighted numbers corresponding to the fault types. For example, the fault types may include type A, type B, type C, and type D, and the weighted numbers corresponding to the fault types include the weighted number of type A, the weighted number of type B, the weighted number of type C, and the weighted number of type D. If the candidate fault types may be type A and type C, the weighted number of the candidate fault type can be determined from the weighted number of type A, the weighted number of type B, the weighted number of type C, and the weighted number of type D.

[0074] S303 : Determine a target fault type from the candidate fault types according to the weights corresponding to the candidate fault types.

[0075] Optionally, in this embodiment, the candidate fault types whose weighted numbers corresponding to the candidate fault types are greater than a preset second threshold value can be determined as the above-mentioned target fault types. Optionally, the value of the above-mentioned second threshold value can be determined based on the weight value corresponding to each fault type and the number of historical outbound call failures corresponding to each fault type. For example, if the weight value corresponding to a certain fault type is 0.1 and the number of historical outbound call failures corresponding to the fault type is 5, the above-mentioned second threshold value can be taken as 0.5. Optionally, in this embodiment, the weighted numbers corresponding to the candidate fault types can also be arranged in order from large to small, and the candidate fault types in the first three arrangements can be taken as the above-mentioned target fault types, or the candidate fault types in the first two arrangements can also be taken as the above-mentioned target fault types.

[0076] In this embodiment, based on the number of outbound calls corresponding to each fault type, candidate fault types for outbound call failures occurring within the predicted time period can be screened out from a variety of fault types, so that the weighted numbers corresponding to the candidate fault types can be quickly determined based on the weighted numbers of each fault type. Since the efficiency of determining the weighted numbers corresponding to the candidate fault types is improved, the target fault type can be quickly determined from the candidate fault types based on the weighted numbers corresponding to the candidate fault types, thereby improving the efficiency of determining the target fault type.

[0077] In the scenario of determining candidate fault types of outbound call failures occurring within a predicted time period based on the number of outbound call failures corresponding to each fault type, in one embodiment, S301 includes:

[0078] The number of outbound call failures greater than a preset threshold is matched to a fault type and determined as a candidate fault type.

[0079] Specifically, in this embodiment, the number of outbound call failures corresponding to each fault type can be compared with a preset number threshold, and the fault type corresponding to the number of outbound call failures greater than the preset number threshold is determined as the above-mentioned candidate fault type. Optionally, the determined candidate fault type can be one fault type or multiple fault types. For example, taking the preset number threshold as 5, if the number of outbound call failures corresponding to fault type A is 8, the number of outbound call failures corresponding to fault type B is 15, and the number of outbound call failures corresponding to fault type C is 2, then fault type A and fault type B can be determined as the above-mentioned candidate fault types.

[0080] In this embodiment, by comparing the number of outbound call failures corresponding to each fault type with a preset number threshold, the fault type corresponding to the number of outbound call failures greater than the preset number threshold can be quickly determined, so that the fault type corresponding to the number of outbound call failures greater than the number threshold can be determined as a candidate fault type, that is, this method improves the efficiency of determining the candidate fault type.

[0081] In some scenarios, if a high outbound call failure rate is detected, an alarm message may be output to inform the user that the outbound call failure rate is high within the predicted time period. In one embodiment, the method further includes: outputting an alarm message when the outbound call failure rate is greater than a preset first threshold.

[0082] Specifically, in this embodiment, when the outbound call failure rate of the outbound call system is greater than the above-mentioned preset first threshold, an alarm prompt message can be output. Optionally, the alarm prompt message can be a flashing alarm light, a voice prompt message, or a text prompt message. Exemplarily, the above-mentioned preset first threshold can be 50%. If the outbound call failure rate obtained above is greater than 50%, an alarm prompt message can be output. Furthermore, if the above-mentioned outbound call failure rate is the outbound call failure rate in the time period of 8:00-9:00 on a certain day, the output alarm prompt message can be a text prompt message of "The outbound call failure rate is high between 8:00-9:00".

[0083] In this embodiment, when the outbound call failure rate is greater than the preset first threshold, the user can be informed in time by outputting an alarm prompt message, so that the user can check the outbound call system in time and solve the fault of the outbound call system as soon as possible, thereby ensuring the normal use of the outbound call system.

[0084] In the above scenario where the time period to be predicted is input into a preset neural network model, the neural network model is a pre-trained neural network model. Figure 4 As shown in Figure 2, the training process of the neural network model includes:

[0085] S401, obtaining the total number of outbound call failures within a sample time period and the standard number of outbound call failures corresponding to various types of failures caused by the outbound call system itself.

[0086] Optionally, in this embodiment, the outbound call system can be monitored during a sample period to obtain the total number of outbound call failures and the standard number of outbound call failures corresponding to various fault types caused by faults in the outbound call system itself. For example, between 08:10 and 08:15, the total number of outbound call failures obtained for the outbound call system during the sample period is 4000, and the standard number of outbound call failures corresponding to various fault types caused by faults in the outbound call system itself is 200. The outbound call system can be monitored and the information obtained can be shown in Table 2.

[0087] Table 2

[0088]

[0089] S402: Determine a standard outbound call failure rate corresponding to a sample time period based on the number of standard outbound call failures and the total number of outbound call failures.

[0090] Optionally, in this embodiment, the sum of the standard outbound call failure times corresponding to various fault types can be taken, and the ratio of this sum to the total number of outbound call failures can be determined as the standard outbound call failure rate corresponding to the above-mentioned sample time period. Furthermore, as an optional implementation method, the ratio of the above-mentioned sum to the total number of outbound call failures can be rounded, and the rounded ratio can be determined as the standard outbound call failure rate corresponding to the above-mentioned sample time period. For example, if the sample time period is 08:10 to 08:15, the total number of outbound call failures N3 from 08:10 to 08:15 is 4000, and the standard number of outbound call failures E2 corresponding to various fault types caused by faults in the outbound call system itself is 200, then the determined standard outbound call failure rate is 5%.

[0091] S403: Input the sample time period into the initial neural network model to predict the sample outbound call failure rate within the sample time period and the number of sample outbound call failures corresponding to various fault types occurring within the sample time period.

[0092] Specifically, before predicting the sample outbound call failure rate and the sample outbound call failure counts corresponding to various types of failures that occurred within the sample time period, it is necessary to first normalize the standard outbound call failure counts. Optionally, the Z-score normalization method can be used to pre-process the data, scaling the data so that it falls into a specific range. For example, for the standard outbound call failure count sequence x1, x2, ..., x n The changes that can be made can be: here Where s is the standard deviation of the original data, is the mean of the original data. Then the new sequence y1,y2,...,y n has a mean of 0 and a variance of 1 and is dimensionless.

[0093] Optionally, after normalizing the data of the standard outbound call failure number, the normalized standard outbound call failure number can also be set as a data type that can be recognized by LSTM, and then the processed sample time period is used as the input of the neural network. The output of the neural network is the outbound call failure rate within the sample time period and the outbound call failure number corresponding to various fault types caused by the outbound call system's own faults within the sample time period.

[0094] In addition, before training the initial neural network model, the obtained sample set can be divided into a training set and a test set according to a preset ratio. For example, the sample set can be divided into a training set and a test set according to a ratio of 7:3.

[0095] S404: Training the initial neural network model based on the standard outbound call failure rate, the number of standard outbound call failures, the sample outbound call failure rate, and the number of sample outbound call failures to obtain a neural network model.

[0096] It is understandable that training the initial neural network model requires establishing the loss function and optimizer of the neural network. Here, the loss function is selected as the cross entropy loss function, and the optimizer is selected as the adaptive gradient descent algorithm.

[0097] At the end of the neural network training, observe the fitting of the neural network. First, observe the changes in the loss function during training. If loss oscillations occur, it means that the model is not very stable. There may be problems with the selection of input data or unreasonable design of the loss function. During the training process, verify the accuracy of the outbound call failure rate within the time period output by the neural network and the number of outbound call failures corresponding to various fault types caused by the outbound call system's own faults within the time period in the training set and the validation set. If it is found that the accuracy of the training set is relatively high, while the accuracy of the validation set is not high enough, it means that the model is overfitting, which will cause the prediction results to be very poor, and the model needs to be optimized. Optionally, in this embodiment, the value of the first loss function of the initial neural network model can be obtained based on the standard outbound call failure rate and the sample outbound call failure rate output by the initial neural network model. The value of the second loss function of the initial neural network model can be obtained based on the standard outbound call failure number corresponding to each fault type and the sample outbound call failure number of each fault type output by the initial neural network model. The parameters of the initial neural network model are adjusted using the value of the first loss function and the value of the second loss function until the value of the first loss function and the value of the second loss function reach a stable or minimum value, and the initial neural network model at this time is determined as the above-mentioned neural network model. For example, Figure 5 As shown, in this embodiment, relevant monitoring data of the outbound call system can be collected first, and after processing the monitoring data, the processed monitoring data is used to train the LSTM monitoring model, and then the trained monitoring model is used to predict the outbound call failure rate in the predicted time period and the number of outbound call failures corresponding to various fault types caused by the outbound call system itself during the predicted time period.

[0098] In this embodiment, by obtaining the total number of outbound call failures within the sample time period and the standard outbound call failure numbers corresponding to various fault types caused by the outbound call system's own faults, the standard outbound call numbers and the total number of outbound call failures can be quickly determined, and then the standard outbound call failure rate corresponding to the sample time period can be quickly determined. In this way, the sample time period can be input into the initial neural network model to predict the sample outbound call failure rate within the sample time period and the sample outbound call failure numbers corresponding to various fault types occurring within the sample time period. The initial neural network model can then be accurately trained based on the standard outbound call failure rate, the standard outbound call failure numbers, the sample outbound call failure rate, and the sample outbound call failure numbers, thereby improving the accuracy of the obtained neural network model.

[0099] In a specific embodiment, the outbound call fault analysis method provided by the present application is described with a complete embodiment, and the method includes the following steps:

[0100] S1, obtaining the total number of outbound call failures within the sample time period and the standard number of outbound call failures corresponding to various types of failures caused by the outbound call system itself;

[0101] S2, determining the standard outbound call failure rate corresponding to the sample time period based on the number of standard outbound calls and the total number of outbound call failures;

[0102] S3, inputting the sample time period into the initial neural network model, predicting the sample outbound call failure rate within the sample time period and the number of sample outbound call failures corresponding to various fault types occurring within the sample time period;

[0103] S4, training the initial neural network model based on the standard outbound call failure rate, the number of standard outbound call failures, the sample outbound call failure rate, and the number of sample outbound call failures to obtain a neural network model;

[0104] S5, inputting the time period to be predicted into a preset neural network model to predict the outbound call failure rate within the time period and the number of outbound call failures corresponding to various types of failures caused by the outbound call system itself within the time period;

[0105] S6: When the outbound call failure rate is greater than a preset first threshold, an alarm prompt message is output.

[0106] S7, when the outbound call failure rate is greater than a preset first threshold, obtaining a weighted number corresponding to each fault type according to the number of outbound call failures and the preset weight value corresponding to each fault type;

[0107] S8, determining candidate fault types of outbound call failures occurring within a time period based on the number of outbound call failures, and determining the fault type corresponding to the number of outbound call failures greater than a preset number threshold as the candidate fault type;

[0108] S9, determining the weighted number corresponding to the candidate fault type according to the weighted numbers of each weight;

[0109] S10 , determining a target fault type from the candidate fault types according to the weights corresponding to the candidate fault types, and determining a candidate fault type whose weight is greater than a preset second threshold as the target fault type.

[0110] For the working principle of the outbound call fault analysis method provided in this embodiment, please refer to the detailed description in the above embodiment, which will not be repeated here.

[0111] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0112] Based on the same inventive concept, embodiments of the present application also provide an outbound call fault analysis device for implementing the above-mentioned outbound call fault analysis method. The implementation solution provided by this device is similar to the implementation solution described in the above-mentioned method. Therefore, the specific limitations of one or more embodiments of the outbound call fault analysis device provided below can be found in the above-mentioned limitations of the outbound call fault analysis method and will not be repeated here.

[0113] In one embodiment, Figure 6 As shown, an outbound call fault analysis device is provided, comprising: a first prediction module 11, a calculation module 12 and a first determination module 13, wherein:

[0114] The first prediction module 11 is used to input the time period to be predicted into a preset neural network model, and predict the outbound call failure rate within the time period and the number of outbound call failures corresponding to various fault types caused by the outbound call system itself within the time period.

[0115] The calculation module 12 is configured to obtain a weighted number corresponding to each fault type according to the number of outbound call failures and the preset weight values corresponding to each fault type when the outbound call failure rate is greater than a preset first threshold.

[0116] The first determining module 13 is configured to determine a target fault type of outbound call failures occurring within a time period according to the number of outbound call failures and the weighted values.

[0117] The outbound call fault analysis device provided in this embodiment can execute the above method embodiment. Its implementation principle and technical effects are similar and will not be described in detail here.

[0118] In one embodiment, Figure 7 As shown, the first determining module 13 includes: a first determining unit 131, a second determining unit 132 and a third determining unit 133, wherein:

[0119] The first determining unit 131 is configured to determine candidate fault types of outbound call failures occurring within a time period according to the number of outbound call failures.

[0120] The second determining unit 132 is configured to determine a weighted number corresponding to a candidate fault type according to each weighted number.

[0121] The third determining unit 133 is configured to determine a target fault type from the candidate fault types according to weights corresponding to the candidate fault types.

[0122] The outbound call fault analysis device provided in this embodiment can execute the above method embodiment. Its implementation principle and technical effects are similar and will not be described in detail here.

[0123] In one embodiment, the third determining unit 133 is configured to determine a candidate fault type whose weight is greater than a preset second threshold as a target fault type.

[0124] The outbound call fault analysis device provided in this embodiment can execute the above method embodiment. Its implementation principle and technical effects are similar and will not be described in detail here.

[0125] In one embodiment, the second determining unit 131 is configured to determine a fault type corresponding to a number of outbound call failures greater than a preset number threshold as a candidate fault type.

[0126] The outbound call fault analysis device provided in this embodiment can execute the above method embodiment. Its implementation principle and technical effects are similar and will not be described in detail here.

[0127] In one embodiment, Figure 8 As shown, the above device further includes: an alarm module 14, wherein:

[0128] The alarm module 14 is configured to output an alarm prompt message when the outbound call failure rate is greater than a preset first threshold.

[0129] The outbound call fault analysis device provided in this embodiment can execute the above method embodiment. Its implementation principle and technical effects are similar and will not be described in detail here.

[0130] In one embodiment, Figure 9 As shown, the above device further includes: an acquisition module 15, a second determination module 16, a second prediction module 17 and a training module 18, wherein:

[0131] The acquisition module 15 is configured to acquire the total number of outbound call failures within a sample time period and the standard number of outbound call failures corresponding to various types of failures caused by the outbound call system itself.

[0132] The second determining module 16 is configured to determine a standard outbound call failure rate corresponding to a sample time period according to each standard outbound call number and the total number of outbound call failures.

[0133] The second prediction module 17 is used to input the sample time period into the initial neural network model to predict the sample outbound call failure rate within the sample time period and the number of sample outbound call failures corresponding to various fault types occurring within the sample time period.

[0134] The training module 18 is used to train the initial neural network model according to the standard outbound call failure rate, the number of standard outbound call failures, the sample outbound call failure rate and the number of sample outbound call failures to obtain a neural network model.

[0135] The outbound call fault analysis device provided in this embodiment can execute the above method embodiment. Its implementation principle and technical effects are similar and will not be described in detail here.

[0136] Each module in the outbound call fault analysis device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a computer device memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0137] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 10 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, an outbound call fault analysis method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse.

[0138] Those skilled in the art will understand that Figure 10The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0139] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0140] The time period to be predicted is input into a preset neural network model to predict the outbound call failure rate within the time period and the number of outbound call failures corresponding to various types of failures caused by the outbound call system itself within the time period;

[0141] When the outbound call failure rate is greater than a preset first threshold, a weighted number corresponding to each fault type is obtained according to the number of outbound call failures and the preset weight value corresponding to each fault type;

[0142] According to the number of outbound call failures and the weighted values of each weight, the target fault type of the outbound call failures occurring within the time period is determined.

[0143] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0144] Determine candidate fault types for outbound call failures occurring within a time period based on the number of outbound call failures.

[0145] Determine the weighted number corresponding to the candidate fault type according to each weighted number;

[0146] According to the weighted numbers corresponding to the candidate fault types, the target fault type is determined from the candidate fault types.

[0147] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0148] The candidate fault type whose weighted number is greater than the preset second threshold is determined as the target fault type.

[0149] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0150] The number of outbound call failures greater than a preset threshold is matched to a fault type and determined as a candidate fault type.

[0151] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0152] When the outbound call failure rate is greater than a preset first threshold, an alarm prompt message is output.

[0153] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0154] Obtain the total number of outbound call failures within the sample period and the standard number of outbound call failures corresponding to various types of failures caused by outbound call system failures.

[0155] Determine the standard outbound call failure rate corresponding to the sample time period based on the number of standard outbound calls and the total number of outbound call failures;

[0156] Input the sample time period into the initial neural network model to predict the sample outbound call failure rate within the sample time period and the number of sample outbound call failures corresponding to various fault types occurring within the sample time period;

[0157] The initial neural network model is trained according to the standard outbound call failure rate, the number of standard outbound call failures, the sample outbound call failure rate and the number of sample outbound call failures to obtain a neural network model.

[0158] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0159] The time period to be predicted is input into a preset neural network model to predict the outbound call failure rate within the time period and the number of outbound call failures corresponding to various types of failures caused by the outbound call system itself within the time period;

[0160] When the outbound call failure rate is greater than a preset first threshold, a weighted number corresponding to each fault type is obtained according to the number of outbound call failures and the preset weight value corresponding to each fault type;

[0161] According to the number of outbound call failures and the weighted values of each weight, the target fault type of the outbound call failures occurring within the time period is determined.

[0162] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0163] Determine candidate fault types for outbound call failures occurring within a time period based on the number of outbound call failures.

[0164] Determine the weighted number corresponding to the candidate fault type according to each weighted number;

[0165] According to the weighted numbers corresponding to the candidate fault types, the target fault type is determined from the candidate fault types.

[0166] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0167] The candidate fault type whose weighted number is greater than the preset second threshold is determined as the target fault type.

[0168] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0169] The number of outbound call failures greater than a preset threshold is matched to a fault type and determined as a candidate fault type.

[0170] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0171] When the outbound call failure rate is greater than a preset first threshold, an alarm prompt message is output.

[0172] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0173] Obtain the total number of outbound call failures within the sample period and the standard number of outbound call failures corresponding to various types of failures caused by outbound call system failures.

[0174] Determine the standard outbound call failure rate corresponding to the sample time period based on the number of standard outbound calls and the total number of outbound call failures;

[0175] Input the sample time period into the initial neural network model to predict the sample outbound call failure rate within the sample time period and the number of sample outbound call failures corresponding to various fault types occurring within the sample time period;

[0176] The initial neural network model is trained according to the standard outbound call failure rate, the number of standard outbound call failures, the sample outbound call failure rate and the number of sample outbound call failures to obtain a neural network model.

[0177] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0178] The time period to be predicted is input into a preset neural network model to predict the outbound call failure rate within the time period and the number of outbound call failures corresponding to various types of failures caused by the outbound call system itself within the time period;

[0179] When the outbound call failure rate is greater than a preset first threshold, a weighted number corresponding to each fault type is obtained according to the number of outbound call failures and the preset weight value corresponding to each fault type;

[0180] According to the number of outbound call failures and the weighted values of each weight, the target fault type of the outbound call failures occurring within the time period is determined.

[0181] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0182] Determine candidate fault types for outbound call failures occurring within a time period based on the number of outbound call failures.

[0183] Determine the weighted number corresponding to the candidate fault type according to each weighted number;

[0184] According to the weighted numbers corresponding to the candidate fault types, the target fault type is determined from the candidate fault types.

[0185] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0186] The candidate fault type whose weighted number is greater than the preset second threshold is determined as the target fault type.

[0187] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0188] The number of outbound call failures greater than a preset threshold is matched to a fault type and determined as a candidate fault type.

[0189] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0190] When the outbound call failure rate is greater than a preset first threshold, an alarm prompt message is output.

[0191] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0192] Obtain the total number of outbound call failures within the sample period and the standard number of outbound call failures corresponding to various types of failures caused by outbound call system failures.

[0193] Determine the standard outbound call failure rate corresponding to the sample time period based on the number of standard outbound calls and the total number of outbound call failures;

[0194] Input the sample time period into the initial neural network model to predict the sample outbound call failure rate within the sample time period and the number of sample outbound call failures corresponding to various fault types occurring within the sample time period;

[0195] The initial neural network model is trained according to the standard outbound call failure rate, the number of standard outbound call failures, the sample outbound call failure rate and the number of sample outbound call failures to obtain a neural network model.

[0196] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0197] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application can include a variety of non-volatile and volatile memories. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include a variety of relational databases and non-relational databases. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, data processing logic devices based on quantum computing, and the like.

[0198] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0199] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A method for analyzing outbound call failures, characterized in that: The method comprises: The time period to be predicted is input into a preset neural network model to predict the outbound call failure rate within the time period and the number of outbound call failures corresponding to various types of failures caused by the outbound call system itself within the time period; the outbound call failure rate within the predicted time period refers to the ratio of the number of outbound call failures corresponding to various types of failures caused by the outbound call system itself within the time period to the total number of calls issued by the outbound call system; When the outbound call failure rate is greater than a preset first threshold, obtaining a weighted number corresponding to each fault type according to the number of outbound call failures and the preset weight value corresponding to each fault type; Determining candidate fault types of outbound call failures occurring within the time period based on the number of outbound call failures; Determining the weighted number corresponding to the candidate fault type according to each of the weighted numbers; A target fault type is determined from the candidate fault types according to the weighted numbers corresponding to the candidate fault types.

2. The method according to claim 1, characterized in that The step of determining the target fault type from the candidate fault types according to weights corresponding to the candidate fault types includes: The candidate fault type whose weighted number is greater than a preset second threshold is determined as the target fault type.

3. The method according to claim 1, characterized in that The determining, based on the number of outbound call failures, candidate fault types of outbound call failures occurring within the time period includes: The fault type corresponding to the number of outbound call failures that is greater than a preset number threshold is determined as the candidate fault type.

4. The method according to any one of claims 1 to 3, characterized in that The method further comprises: When the outbound call failure rate is greater than the preset first threshold, an alarm prompt message is output.

5. The method according to claim 1, wherein The training process of the neural network model includes: Obtain the total number of outbound call failures within the sample period and the standard number of outbound call failures corresponding to various types of failures caused by outbound call system failures. Determining a standard outbound call failure rate corresponding to the sample time period according to each of the standard outbound call failure numbers and the total outbound call failure number; Inputting the sample time period into the initial neural network model, predicting the sample outbound call failure rate within the sample time period and the number of sample outbound call failures corresponding to multiple fault types occurring within the sample time period; The initial neural network model is trained according to the standard outbound call failure rate, the number of standard outbound call failures, the sample outbound call failure rate and the number of sample outbound call failures to obtain the neural network model.

6. An outbound call fault analysis device, characterized in that: The device comprises: A prediction module is configured to input a time period to be predicted into a preset neural network model, and predict an outbound call failure rate within the time period and the number of outbound call failures corresponding to various types of failures caused by the outbound call system itself within the time period; the outbound call failure rate within the predicted time period refers to the ratio of the number of outbound call failures corresponding to various types of failures caused by the outbound call system itself within the time period to the total number of calls issued by the outbound call system; a calculation module, configured to obtain a weighted number corresponding to each fault type according to the number of outbound call failures and the preset weight values corresponding to each fault type when the outbound call failure rate is greater than a preset first threshold; A determination module is used to determine the candidate fault types of outbound call failures that occurred within the time period based on the number of outbound call failures; determine the weighted numbers corresponding to the candidate fault types based on the weighted numbers; and determine the target fault type from the candidate fault types based on the weighted numbers corresponding to the candidate fault types.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images