Private line networking hidden danger prediction method and device

Abstract

The invention provides a private line networking hidden danger prediction method and device, and relates to the technical field of artificial intelligence. The method comprises the following steps: acquiring communication analysis data for predicting the hidden danger of the private line networking, wherein the communication analysis data comprises fault analysis data, degradation analysis data and hidden danger analysis data; inputting the fault analysis data into a trained private line network fault prediction model to obtain a private line fault probability; inputting the degradation analysis data into a trained special line business performance degradation prediction model to obtain a special line degradation probability; and if the special line fault probability is greater than a fault threshold value and / or the special line degradation probability is greater than a degradation threshold value, inputting the hidden danger analysis data into a trained special line networking hidden danger real-time prediction model to obtain a special line networking hidden danger probability. Therefore, the artificial intelligence model can be used for predicting the fault probability of the private line network, the deterioration probability of the service performance of the private line and the hidden danger probability of the private line networking, thereby achieving the real-time, comprehensive and rapid prediction of the hidden danger of the private line networking, and greatly reducing the labor cost.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to a method and device for predicting hidden dangers in dedicated line networking. Background Art

[0002] Network risk prediction technology is a network security technology that aims to identify, predict and discover potential risks and vulnerabilities in the network. Common methods used in network risk prediction include expert rules, fault trees, simulation, etc.

[0003] Among them, expert rules are used: by defining the network topology type and the possible hidden dangers of each structure, and then using the expert rule system for matching and prediction; fault tree is used: by analyzing the network topology and possible faults of the equipment, and then using the fault tree for prediction; simulation is used: by simulating the working conditions of the network topology and equipment, and then observing and predicting possible hidden dangers.

[0004] However, network risk prediction technologies such as expert rules, fault trees, and simulations mostly focus on predicting and analyzing static risks in network structures, and it is difficult to truly reflect the prediction of dynamic risks in network structures due to the actual network environment and various uncertain factors. It requires a lot of manpower and time for analysis and maintenance, and it is difficult to cover all risk types and situations, which is an urgent problem to be solved. Summary of the invention

[0005] The present invention provides a method and device for predicting hidden dangers in dedicated line networking.

[0006] In a first aspect, the present disclosure provides a method for predicting hidden dangers in dedicated line networking, comprising: obtaining communication analysis data for predicting hidden dangers in dedicated line networking, wherein the communication analysis data comprises: fault analysis data, degradation analysis data and hidden danger analysis data; inputting the fault analysis data into a trained dedicated line network fault prediction model to obtain a dedicated line failure probability; inputting the degradation analysis data into a trained dedicated line service performance degradation prediction model to obtain a dedicated line degradation probability; if the dedicated line failure probability is greater than a failure threshold, and / or the dedicated line degradation probability is higher than a degradation threshold, inputting the hidden danger analysis data into a trained dedicated line networking hidden danger real-time prediction model to obtain a dedicated line networking hidden danger probability.

[0007] In some embodiments, obtaining communication analysis data for predicting dedicated line networking risks includes: obtaining dedicated line real-time communication data; obtaining analysis data for predicting dedicated line networking risks from the dedicated line real-time communication data; and performing one-hot encoding conversion on the analysis data to generate communication analysis data.

[0008] In some embodiments, the above method further includes: obtaining a fault analysis sample data set; training a dedicated line network fault prediction model according to the fault analysis sample data set to generate a trained dedicated line network fault prediction model.

[0009] In some embodiments, obtaining a fault analysis sample data set includes: obtaining first dedicated line data, wherein the first dedicated line data includes dedicated line basic data, dedicated line historical performance data, dedicated line historical fault data, dedicated line alarm data, and dedicated line network element equipment log data; determining fault sample data from the first dedicated line data; performing one-hot encoding conversion on the fault sample data; filling missing value data in the fault sample data after the one-hot encoding conversion with zeros, and deduplicating the duplicated data to obtain a fault analysis sample data set.

[0010] In some embodiments, the above method further includes: obtaining a degradation analysis sample data set; training a dedicated line service performance degradation prediction model according to the degradation analysis sample data set to generate a trained dedicated line service performance degradation prediction model.

[0011] In some embodiments, obtaining a degradation analysis sample data set includes: obtaining second dedicated line data, wherein the second dedicated line data includes transmission dedicated line performance data, Internet dedicated line performance data, and voice dedicated line performance data; determining the degradation sample data from the second dedicated line data; performing one-hot encoding conversion on the degradation sample data; performing zero padding on missing value data in the degraded sample data after the one-hot encoding conversion, and performing deduplication processing to obtain the degradation analysis sample data set.

[0012] In some embodiments, the above method also includes: obtaining a hidden danger analysis sample data set; training a dedicated line networking hidden danger real-time prediction model according to the hidden danger analysis sample data set to generate a trained dedicated line networking hidden danger real-time prediction model.

[0013] In some embodiments, obtaining a hidden danger analysis sample data set includes: obtaining third dedicated line data, wherein the third dedicated line data includes networking model data corresponding to the dedicated line service assurance level, dedicated line network management networking topology data, and routing segment quality inspection data; determining hidden danger sample data from the third dedicated line data; performing one-hot encoding conversion on the hidden danger sample data; filling missing value data in the hidden danger sample data after the one-hot encoding conversion with zeros, and deduplicating the duplicated data to obtain the hidden danger analysis sample data set.

[0014] In a second aspect, the present disclosure provides a dedicated line networking hidden danger prediction device, comprising: a data acquisition unit, used to acquire communication analysis data for predicting dedicated line networking hidden dangers, wherein the communication analysis data comprises: fault analysis data, degradation analysis data and hidden danger analysis data; a first processing unit, used to input the fault analysis data into a trained dedicated line network fault prediction model to obtain a dedicated line failure probability; a second processing unit, used to input the degradation analysis data into a trained dedicated line service performance degradation prediction model to obtain a dedicated line degradation probability; a third processing unit, used to input the hidden danger analysis data into a trained dedicated line networking hidden danger real-time prediction model to obtain a dedicated line networking hidden danger probability if the dedicated line failure probability is greater than a failure threshold, and / or the dedicated line degradation probability is higher than a degradation threshold.

[0015] In a third aspect, the present disclosure provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the dedicated line networking hidden danger prediction method disclosed in an embodiment of the present disclosure.

[0016] In a fourth aspect, the present disclosure provides a computer-readable storage medium, in which computer execution instructions are stored. When the computer execution instructions are executed by a processor, they are used to implement the dedicated line networking risk prediction method disclosed in an embodiment of the present disclosure.

[0017] In a fifth aspect, the present disclosure provides a computer program product, including a computer program, which, when executed by a processor, implements the dedicated line networking risk prediction method disclosed in an embodiment of the present disclosure.

[0018] The technical solution provided by the embodiments of the present disclosure brings at least the following beneficial effects:

[0019] In the disclosed embodiments, an artificial intelligence model can be used to evaluate the status of a dedicated line, predict the probability of dedicated line network failure, the probability of dedicated line service performance degradation, and the probability of dedicated line networking hidden dangers, thereby achieving real-time, comprehensive and rapid prediction of dedicated line networking hidden dangers and significantly reducing labor costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0021] Figure 1 The present invention is a flowchart of a method for predicting hidden dangers in a dedicated line network according to an exemplary embodiment;

[0022] Figure 2 A schematic diagram of a method for training a dedicated line network fault prediction model according to an exemplary embodiment;

[0023] Figure 3 It is a schematic diagram of a training method of a dedicated line service performance degradation prediction model according to an exemplary embodiment;

[0024] Figure 4 It is a schematic diagram of a training method of a real-time prediction model of hidden dangers in dedicated line networking according to an exemplary embodiment;

[0025] Figure 5 The present invention is a flowchart of a method for predicting hidden dangers in government-enterprise dedicated line networking according to an exemplary embodiment;

[0026] Figure 6 It is a structural diagram of a dedicated line networking hidden danger prediction device according to an exemplary embodiment;

[0027] Figure 7 is a structural diagram of a dedicated line networking hidden danger prediction device according to another exemplary embodiment;

[0028] Figure 8 is a structural diagram of a dedicated line networking hidden danger prediction device according to another exemplary embodiment;

[0029] Fig. 9 is a structural diagram of a dedicated line networking hidden danger prediction device according to another exemplary embodiment;

[0030] Fig.10 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure.

[0031] The above drawings show clear embodiments of the present disclosure, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present disclosure in any way, but to illustrate the concepts of the present disclosure to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0032] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0033] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in the embodiments of the present disclosure are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0034] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0035] Among the related technologies, network risk prediction technologies such as expert rules, fault trees, and simulations mostly focus on predicting and analyzing static risks in network structures, which are difficult to truly reflect the prediction of dynamic risks in network structures due to the actual network environment and various uncertain factors. A large amount of manpower and time are required for analysis and maintenance, and it is difficult to cover all risk types and situations.

[0036] Based on this, the embodiment of the present disclosure provides a method and device for predicting hidden dangers in dedicated line networking, wherein the method includes: obtaining communication analysis data for predicting hidden dangers in dedicated line networking, wherein the communication analysis data includes: fault analysis data, degradation analysis data and hidden danger analysis data; inputting the fault analysis data into a trained dedicated line network fault prediction model to obtain the probability of dedicated line faults; inputting the degradation analysis data into a trained dedicated line service performance degradation prediction model to obtain the probability of dedicated line degradation; if the probability of dedicated line faults is greater than the fault threshold, and / or the probability of dedicated line degradation is higher than the degradation threshold, inputting the hidden danger analysis data into a trained real-time prediction model for hidden dangers in dedicated line networking to obtain the probability of hidden dangers in dedicated line networking. Thus, the use of an artificial intelligence model can evaluate the status of a dedicated line, predict the probability of dedicated line network faults, the probability of dedicated line service performance degradation and the probability of hidden dangers in dedicated line networking, and realize real-time, comprehensive and rapid prediction of hidden dangers in dedicated line networking, greatly reducing labor costs.

[0037] Figure 1 The present invention is a flowchart of a method for predicting hidden dangers in dedicated line networking according to an exemplary embodiment.

[0038] like Figure 1 As shown, the dedicated line networking risk prediction method provided by the embodiment of the present disclosure includes but is not limited to the following steps:

[0039] S1, obtaining communication analysis data for predicting hidden dangers of dedicated line networking.

[0040] Among them, communication analysis data includes: fault analysis data, degradation analysis data and hidden danger analysis data.

[0041] In the disclosed embodiment, the dedicated line networking hidden danger prediction method is executed by a dedicated line networking hidden danger prediction device, wherein the dedicated line networking hidden danger prediction device can be an electronic device, such as a mobile phone, a computer, a tablet, etc.

[0042] In the disclosed embodiment, the electronic device executes the dedicated line networking hidden danger prediction method by installing a specific application. For example, the electronic device is a mobile phone, and the specific application is installed in the mobile phone to execute the dedicated line networking hidden danger prediction method.

[0043] Among them, fault analysis data is used to predict the probability of private line network failure, degradation analysis data is used to predict the probability of private line service performance degradation, and hidden danger analysis data is used to predict the probability of private line network hidden dangers.

[0044] In view of the ever-increasing carrying requirements of government and enterprise dedicated lines, it is crucial to build an end-to-end high-quality dedicated support network and provide valuable user service levels and perception. In the disclosed embodiment, communication analysis data for predicting the hidden dangers of government and enterprise dedicated line networking is obtained. Among them, the fault analysis data is used to predict the probability of network failure of government and enterprise dedicated lines, the degradation analysis data is used to predict the probability of performance degradation of government and enterprise dedicated line services, and the hidden danger analysis data is used to predict the probability of hidden dangers of government and enterprise dedicated line networks.

[0045] In some embodiments, S1, obtaining communication analysis data for predicting dedicated line networking risks, includes: obtaining dedicated line real-time communication data; obtaining analysis data for predicting dedicated line networking risks from dedicated line real-time communication data; performing one-hot encoding conversion on the analysis data to generate communication analysis data.

[0046] In an embodiment of the present disclosure, real-time communication data of a dedicated line is obtained, including at least one of dedicated line basic data, dedicated line performance data, dedicated line fault data, dedicated line alarm data, and dedicated line network element device log data; analysis data for predicting hidden dangers of dedicated line networking is obtained from at least one of the dedicated line basic data, dedicated line performance data, dedicated line fault data, dedicated line alarm data, and dedicated line network element device log data; the analysis data is converted into a one-hot encoding to generate fault analysis data.

[0047] The basic data of the dedicated line includes at least one of the product number, product type, service guarantee level, activation time, product status, and customer code.

[0048] Among them, product types include: APN IoT dedicated line, SMS dedicated line, Internet dedicated line, data dedicated line, voice dedicated line, cloud network dedicated line. Business guarantee levels include A, AA, AAA, AAA+, and ordinary. Product status includes archiving and resource configuration.

[0049] In the disclosed embodiment, analysis data for predicting dedicated line networking risks is obtained from dedicated line basic data, and at least one of product type, service assurance level, and product status can be obtained. At least one of product type, service assurance level, and product status is converted into a one-hot encoding to generate fault analysis data.

[0050] It can be understood that One-Hot Encoding (One-Hot Encoding) is a feature encoding method that is mainly used to convert discrete features into continuous features so that machine learning algorithms can better process them.

[0051] Among them, the dedicated line performance data includes at least one of the following: downstream average bandwidth utilization, downstream peak bandwidth utilization, upstream average bandwidth utilization, upstream peak bandwidth utilization, A-end average delay, A-end packet loss rate, Z-end average delay, and Z-end packet loss rate.

[0052] In the disclosed embodiment, analysis data for predicting hidden dangers in dedicated line networking is obtained from dedicated line performance data, and at least one of the downstream average bandwidth utilization, downstream peak bandwidth utilization, upstream average bandwidth utilization, upstream peak bandwidth utilization, A-end average delay, A-end packet loss rate, Z-end average delay, and Z-end packet loss rate can be obtained and converted into a one-hot encoding to generate fault analysis data.

[0053] The dedicated line fault data includes at least one of the fault work order number, customer name, product number, equipment name, dispatch time, filing time, fault type and network element name.

[0054] Among them, the fault types include network reasons, client reasons, customer reasons, other reasons, and no fault.

[0055] In the disclosed embodiment, analysis data for predicting dedicated line networking risks is obtained from dedicated line performance data. At least one of the fault work order number, customer name, product number, equipment name, dispatch time, archiving time, fault type and network element name can be obtained and converted into a one-hot encoding to generate fault analysis data.

[0056] The dedicated line alarm data includes at least one of an alarm fingerprint, an alarm occurrence time, an alarm title, a network element name, an alarm location object, an active alarm count, and a product number.

[0057] In the disclosed embodiment, analysis data for predicting hidden dangers in dedicated line networking is obtained from dedicated line performance data. At least one of the alarm fingerprint, alarm occurrence time, alarm title, network element name, alarm location object, active alarm count, and product number can be obtained and converted into a one-hot encoding to generate fault analysis data.

[0058] The dedicated line network element device log data includes at least one of a network element device identifier, a device name, a CPU utilization rate, and a board utilization rate.

[0059] In the disclosed embodiment, analysis data for predicting hidden dangers of dedicated line networking is obtained from dedicated line performance data. At least one of the network element device identification, device name, CPU utilization and board utilization can be obtained for one-hot encoding conversion to generate fault analysis data.

[0060] In the disclosed embodiments, real-time communication data of dedicated lines are obtained, including transmission dedicated line performance data, Internet dedicated line performance data and voice dedicated line performance data. Analysis data for predicting hidden dangers of dedicated line networking are obtained from the transmission dedicated line performance data, Internet dedicated line performance data and voice dedicated line performance data. The analysis data is converted into one-hot encoding to generate degradation analysis data.

[0061] Among them, the transmission line performance data includes at least one of the product number, customer code, customer name, data statistics start time, data statistics end time, sending bandwidth utilization, receiving bandwidth utilization, sending rate, receiving rate, A-end packet loss rate, A-end delay, A-end jitter delay, Z-end packet loss rate, Z-end delay, Z-end jitter delay, A-end sending optical power, A-end receiving optical power, A-end CRC error packet number, Z-end sending optical power, Z-end receiving optical power, and Z-end CRC error packet number.

[0062] In the disclosed embodiment, analysis data for predicting hidden dangers of dedicated line networking is obtained from dedicated line performance data, and at least one of the following can be obtained: product number, customer code, customer name, data statistics start time, data statistics end time, sending bandwidth utilization, receiving bandwidth utilization, sending rate, receiving rate, A-end packet loss rate, A-end delay, A-end jitter delay, Z-end packet loss rate, Z-end delay, Z-end jitter delay, A-end sending optical power, A-end receiving optical power, A-end CRC error packet number, Z-end sending optical power, Z-end receiving optical power, and Z-end CRC error packet number, and then converted into a one-hot encoding to generate degradation analysis data.

[0063] Among them, the Internet dedicated line performance data includes product number, customer code, customer name, data statistics start time, data statistics end time, downstream average bandwidth utilization, downstream peak bandwidth utilization, upstream average bandwidth utilization, upstream peak bandwidth utilization, ONUID, ONU optical power, packet loss rate, maximum delay, minimum delay, average delay, maximum jitter, minimum jitter, average jitter, upstream traffic and downstream traffic. At least one of the following.

[0064] In the disclosed embodiment, analysis data for predicting hidden dangers of dedicated line networking is obtained from dedicated line performance data, and product number, customer code, customer name, data statistics start time, data statistics end time, downstream average bandwidth utilization, downstream peak bandwidth utilization, upstream average bandwidth utilization, upstream peak bandwidth utilization, ONUID, ONU optical power, packet loss rate, maximum delay, minimum delay, average delay, maximum jitter, minimum jitter, average jitter, at least one of upstream traffic and downstream traffic can be obtained for unique hot encoding conversion to generate degradation analysis data.

[0065] Among them, the voice dedicated line performance data includes at least one of the product number, customer code, customer name, data statistics start time, data statistics end time, number of calls, number of call connections, call establishment time, call connection rate, average call establishment time, initial call establishment delay, terminating call establishment delay and terminating call connection delay.

[0066] In the disclosed embodiment, analysis data for predicting hidden dangers in dedicated line networking is obtained from dedicated line performance data, and at least one of the product number, customer code, customer name, data statistics start time, data statistics end time, number of calls, number of call connections, call establishment duration, call connection rate, average call establishment duration, initial call establishment delay, final call establishment delay and final call connection delay can be obtained and converted into a one-hot encoding to generate degradation analysis data.

[0067] In the disclosed embodiment, real-time communication data of the dedicated line is obtained, including networking model data corresponding to the dedicated line service assurance level, dedicated line network management networking topology data, and routing segment quality inspection data; analysis data for predicting dedicated line networking risks is obtained from the networking model data corresponding to the dedicated line service assurance level, dedicated line network management networking topology data, and routing segment quality inspection data; the analysis data is converted into one-hot encoding to generate risk analysis data.

[0068] Among them, the networking model data corresponding to the dedicated line service assurance level includes at least one of the service assurance level, product type and networking model requirements.

[0069] Among them, product types include: APN IoT dedicated line, SMS dedicated line, Internet dedicated line, data dedicated line, voice dedicated line, cloud network dedicated line. Service guarantee levels include A, AA, AAA, AAA+, and ordinary.

[0070] In an embodiment of the present disclosure, analysis data for predicting dedicated line networking risks is obtained from dedicated line performance data. Networking model data including at least one of service assurance level, product type and networking model requirements can be obtained and converted into one-hot encoding to generate risk analysis data.

[0071] The dedicated line network management networking topology data includes at least one of nodes node identification (ID), nodes node type, lines node ID and lines node type.

[0072] Among them, the node types include OBD, OLT, ONU, PTN, etc. The line node types include optical cable, pigtail, etc.

[0073] In an embodiment of the present disclosure, analysis data for predicting dedicated line networking risks is obtained from dedicated line performance data. At least one of nodes node ID, nodes node type, Lines node ID and Lines node type can be obtained and converted into a one-hot encoding to generate risk analysis data.

[0074] The routing segment quality inspection data includes at least one of the product number, product type, service guarantee level, number of power supplies, number of boards, and number of routing breakpoints.

[0075] In the disclosed embodiment, analysis data for predicting hidden dangers in dedicated line networking is obtained from dedicated line performance data. At least one of the product number, product type, service assurance level, number of power supplies, number of boards, and number of routing breakpoints can be obtained and converted into a one-hot encoding to generate hidden danger analysis data.

[0076] S2: Input the fault analysis data into the trained private line network fault prediction model to obtain the private line fault probability.

[0077] In the disclosed embodiment, after the fault analysis data is acquired, the fault analysis data is input into a trained dedicated line network fault prediction model to obtain the dedicated line fault probability.

[0078] Among them, the trained dedicated line network fault prediction model can be obtained by training using the xgboost model.

[0079] In some embodiments, the above method further includes: obtaining a fault analysis sample data set; training a dedicated line network fault prediction model according to the fault analysis sample data set to generate a trained dedicated line network fault prediction model.

[0080] In some embodiments, obtaining a fault analysis sample data set includes: obtaining first dedicated line data, wherein the first dedicated line data includes dedicated line basic data, dedicated line historical performance data, dedicated line historical fault data, dedicated line alarm data, and dedicated line network element equipment log data; determining fault sample data from the first dedicated line data; performing one-hot encoding conversion on the fault sample data; filling missing value data in the fault sample data after the one-hot encoding conversion with zeros, and deduplicating the duplicated data to obtain a fault analysis sample data set.

[0081] In the disclosed embodiment, firstly, the first dedicated line data is obtained, including: dedicated line basic data, dedicated line historical performance data, dedicated line historical fault data, dedicated line alarm data and dedicated line network element equipment log data. As shown in Table 1 below, the fault sample data including fault time, fault type, alarm time, service impact, delay, packet loss, etc. are selected, and the fields used for model training are used as input features (that is, fault sample data) to perform preliminary data cleaning.

[0082] Table 1

[0083]

[0084]

[0085] It should be noted that in the column of features converted to training data, fields indicated as not being input into the training model are not used for model training, fields indicated as being directly input into the model can be directly used as input data for model training, and fields indicated as being converted to One-hot encoding are used as input data for model training after One-hot encoding conversion.

[0086] Secondly, the input data (i.e., fault sample data) is preprocessed, and the nominal data is converted to One-hot encoding. One-hot encoding uses an N-bit state register to encode N states. Each state has its own independent register bit, and at any time, only one of them is valid. After the fault sample data is converted to One-hot encoding, it can not only preserve the feature information of the data but also achieve the function of expanding the features; for missing value data, 0 is filled; all data are deduplicated to ensure the uniqueness of the training data, and the fault analysis sample data set is obtained.

[0087] Finally, the fault analysis sample data set is divided into training set, validation set and test set in the ratio of 0.7, 0.2 and 0.1. Whether it is a fault, whether it is degraded, and whether it affects the service are used as labels to train the xgboost model. The trained dedicated line network fault prediction model is obtained to predict the probability of dedicated line network failure.

[0088] For example, Figure 2 As shown in the figure, taking the government and enterprise dedicated line as an example, we first collect samples to obtain the dedicated line basic data, dedicated line historical performance data, dedicated line historical fault data, dedicated line alarm data and dedicated line network element equipment log data, perform data cleaning and data preprocessing, and then train the xgboost model to obtain the trained dedicated line network fault prediction model for predicting the failure probability of the government and enterprise dedicated line network.

[0089] S3, inputting the degradation analysis data into the trained dedicated line service performance degradation prediction model to obtain the dedicated line degradation probability.

[0090] In the disclosed embodiment, after the degradation analysis data is acquired, the degradation analysis data is input into a trained dedicated line service performance degradation prediction model to obtain the dedicated line failure probability.

[0091] The trained dedicated line service performance degradation prediction model can be obtained by training using the xgboost model.

[0092] In some embodiments, the above method further includes: obtaining a degradation analysis sample data set; training a dedicated line service performance degradation prediction model according to the degradation analysis sample data set to generate a trained dedicated line service performance degradation prediction model.

[0093] In some embodiments, obtaining a degradation analysis sample data set includes: obtaining second dedicated line data, wherein the second dedicated line data includes transmission dedicated line performance data, Internet dedicated line performance data, and voice dedicated line performance data; determining the degradation sample data from the second dedicated line data; performing one-hot encoding conversion on the degradation sample data; performing zero padding on missing value data in the degraded sample data after the one-hot encoding conversion, and performing deduplication processing to obtain the degradation analysis sample data set.

[0094] In the embodiment of the present disclosure, firstly, the second dedicated line data is obtained, including transmission dedicated line performance data, Internet dedicated line performance data and voice dedicated line performance data, and then it is determined whether the data is degraded data according to relevant rules. The degradation standard is as follows: as long as one of the following indicators is abnormal, it is degraded;

[0095] ① The HTTP service request success rate is less than 90%, which is an abnormality; ② The TCP request success rate is less than 90%, which is an abnormality; ③ The PING delay of key customers is >50, which is an abnormality; ④ The PING packet loss rate is >3%, which is an abnormality; ⑤ The peak rate of the upstream instantaneous rate is lower than 70% of the average of the past peak rate, which is an abnormality; ⑥ The peak rate of the downstream instantaneous rate is lower than 70% of the average of the past peak rate, which is an abnormality.

[0096] As shown in Table 2 below, the fields that can be used for model training are selected as input features (i.e., degraded sample data) for preliminary data cleaning.

[0097] Table 2

[0098]

[0099]

[0100]

[0101] It should be noted that in the column of features converted to training data, fields indicated as not being input into the training model are not used for model training, fields indicated as being directly input into the model can be directly used as input data for model training, and fields indicated as being converted to One-hot encoding are used as input data for model training after One-hot encoding conversion.

[0102] Secondly, the input data (i.e., degraded sample data) is preprocessed, and the nominal data is converted to One-hot encoding. One-hot encoding uses an N-bit state register to encode N states. Each state has its own independent register bit, and at any time, only one of them is valid. After the data is converted to One-hot encoding, it can not only save the feature information of the data but also achieve the function of expanding the features; for missing value data, 0 is filled; all data are deduplicated to ensure the uniqueness of the training data, and the degradation analysis sample data set is obtained.

[0103] Finally, the degradation analysis sample data set is divided into training set, validation set and test set in the ratio of 0.7, 0.2 and 0.1. Whether degradation occurs is used as a label to train the xgboost model. The trained dedicated line service performance degradation prediction model is obtained to predict the probability of dedicated line service performance degradation.

[0104] For example, Figure 3 As shown in the figure, taking the government and enterprise dedicated line as an example, we first collect samples to obtain the transmission dedicated line performance data, Internet dedicated line performance data and voice dedicated line performance data, perform data cleaning and data preprocessing, and then train the xgboost model to obtain the trained dedicated line service performance degradation prediction model for predicting the probability of government and enterprise dedicated line service performance degradation.

[0105] S4: If the dedicated line failure probability is greater than the failure threshold, and / or the dedicated line degradation probability is higher than the degradation threshold, the hidden danger analysis data is input into the trained dedicated line networking hidden danger real-time prediction model to obtain the dedicated line networking hidden danger probability.

[0106] In the disclosed embodiment, if the dedicated line failure probability is greater than the failure threshold, and / or the dedicated line degradation probability is higher than the degradation threshold, when the hidden danger analysis data is obtained, the hidden danger analysis data is input into the trained real-time prediction model of the hidden dangers of the dedicated line networking to obtain the hidden danger probability of the dedicated line networking.

[0107] Among them, the trained dedicated line networking hidden danger real-time prediction model can be trained using the xgboost model.

[0108] In some embodiments, the above method also includes: obtaining a hidden danger analysis sample data set; training a dedicated line networking hidden danger real-time prediction model according to the hidden danger analysis sample data set to generate a trained dedicated line networking hidden danger real-time prediction model.

[0109] In some embodiments, obtaining a hidden danger analysis sample data set includes: obtaining third dedicated line data, wherein the third dedicated line data includes networking model data corresponding to the dedicated line service assurance level, dedicated line network management networking topology data, and routing segment quality inspection data; determining hidden danger sample data from the third dedicated line data; performing one-hot encoding conversion on the hidden danger sample data; filling missing value data in the hidden danger sample data after the one-hot encoding conversion with zeros, and deduplicating the duplicated data to obtain the hidden danger analysis sample data set.

[0110] In the disclosed embodiment, firstly, the third dedicated line data is obtained, including the networking model data corresponding to the dedicated line service assurance level, the dedicated line network management networking topology data and the routing segment quality inspection data. As shown in Table 3 below, the fields that can be used for model training are selected as input features (that is, hidden danger sample data) to perform preliminary data cleaning.

[0111] Table 3

[0112]

[0113]

[0114] It should be noted that in the column of features converted to training data, fields indicated as not being input into the training model are not used for model training, fields indicated as being directly input into the model can be directly used as input data for model training, and fields indicated as being converted to One-hot encoding are used as input data for model training after One-hot encoding conversion.

[0115] Secondly, the input data (that is, the hidden danger sample data) is preprocessed, and the nominal data is converted to One-hot encoding. One-hot encoding uses an N-bit state register to encode N states. Each state has its own independent register bit, and at any time, only one of them is valid. After the data is converted to One-hot encoding, it can not only save the feature information of the data but also achieve the function of expanding the features; for missing value data, 0 is filled; all data are deduplicated to ensure the uniqueness of the training data, and the hidden danger sample data set is obtained.

[0116] Finally, the hidden danger sample data set is divided into training set, validation set and test set in the ratio of 0.7, 0.2 and 0.1. Whether there is a hidden danger is used as a label to train the xgboost model, and the trained dedicated line network hidden danger real-time prediction model is obtained to predict the probability of dedicated line network hidden dangers.

[0117] For example, Figure 4 As shown in the figure, taking the government and enterprise dedicated line as an example, we first collect samples to obtain the networking model data, dedicated line network management networking topology data and routing segment quality inspection data corresponding to the dedicated line service assurance level, perform data cleaning and data preprocessing, and then train the xgboost model to obtain the trained dedicated line networking hidden danger real-time prediction model for predicting the probability of government and enterprise dedicated line network hidden dangers.

[0118] For example, Figure 5 As shown, the dedicated line networking hidden danger prediction method provided by the embodiment of the present disclosure is introduced by taking the government-enterprise dedicated line as an example. First, real-time data collection of the dedicated line is performed, and the collected data is shown in Tables 1, 2 and 3 above. Then, the dedicated line network fault prediction is performed according to the data in Table 1, and the dedicated line fault prediction probability is obtained according to the data in Table 1 by calling the API to request the government-enterprise dedicated line network fault prediction model, and the data is returned through the API; and the dedicated line service performance degradation prediction is performed according to the data in Table 2, and the dedicated line degradation prediction probability is obtained according to the data in Table 2 by calling the API to request the government-enterprise dedicated line service performance degradation prediction model, and the data is returned through the API. Among them, when the dedicated line failure prediction probability is higher than the threshold and the dedicated line degradation prediction probability is higher than the threshold, the dedicated line networking hidden danger probability prediction is performed, and according to the data in Table 3, the government-enterprise dedicated line networking hidden danger real-time prediction model is called by the API to obtain the dedicated line networking hidden danger prediction probability according to the data in Table 3, and the data is returned through the API to obtain the dedicated line networking hidden danger probability.

[0119] The presently disclosed embodiments provide a real-time prediction capability for government and enterprise dedicated line networking hazards based on historical performance indicators of dedicated lines. The capability consists of the following parts: government and enterprise dedicated line network fault prediction model training, government and enterprise dedicated line service performance degradation prediction model training, government and enterprise dedicated line networking hazard real-time prediction model training, and government and enterprise dedicated line networking hazard real-time prediction.

[0120] 1. Training of the government-enterprise private line network fault prediction model. In addition to using private line alarm data, this process also integrates private line fault data, private line performance data, and private line network element equipment log data. By preprocessing these four aspects of data and inputting them into the intelligent model, the government-enterprise private line network fault prediction model is trained through the improved xgboost method to comprehensively evaluate the private line status and predict the probability of government-enterprise private line network failure. This makes up for the blind spots in the prediction of the impact of private line performance, private line failure, network element equipment status, etc. on the private line network failure, which is an original technology of the proposal.

[0121] 2. Training of the performance degradation prediction model for government and enterprise dedicated line services. This process uses the transmission dedicated line performance data, Internet dedicated line performance data, and voice dedicated line performance data. After data cleaning and preprocessing, the data is classified based on expert experience rules to determine whether the data is degraded. The performance degradation prediction model for government and enterprise dedicated line services is trained based on the improved xgboost algorithm to predict the probability of performance degradation of government and enterprise dedicated line services, extract relevant data features of multiple dedicated lines, and obtain multiple information of dedicated line networking to improve the accuracy of performance degradation prediction.

[0122] 3. Training of real-time prediction model for hidden dangers in government-enterprise dedicated line networks. This process uses static elements of dedicated line networks such as dedicated line networking model data, dedicated line network management networking topology data, and routing segment quality inspection data, and uses government-enterprise dedicated line network fault prediction, government-enterprise dedicated line service performance degradation prediction and other means to predict dynamic elements such as potential risks and trend changes caused by dedicated line faults and dedicated line service performance degradation. Based on the improved xgboost algorithm, the real-time prediction model for hidden dangers in government-enterprise dedicated line networks is trained to predict the probability of hidden dangers in government-enterprise dedicated line networks. It not only pays attention to static elements of dedicated line networks, but also integrates dynamic elements such as potential risks and trend changes of faults and performance degradation, and predicts the probability of hidden dangers in dedicated line networks with the help of intelligent models. This is the unique technical innovation of the proposal.

[0123] 4. Real-time prediction of network risks of government-enterprise dedicated lines. This process uses the network failure prediction model of government-enterprise dedicated lines and the service performance degradation prediction model of government-enterprise dedicated lines to predict failures and degradations of government-enterprise dedicated lines. If the failure prediction probability or degradation prediction probability is higher than the preset threshold, the real-time prediction model of network risks of government-enterprise dedicated lines is used to predict the probability of network risks of government-enterprise dedicated lines in real time. By combining three xgboost models, real-time and rapid prediction of multi-dimensional field data is achieved, which greatly reduces labor costs while using a large number of relevant data features to achieve cost reduction and efficiency improvement.

[0124] Taking the government-enterprise dedicated line as the research object, the network fault prediction model and the service performance degradation prediction model of the government-enterprise dedicated line are trained by using the xgboost algorithm to predict the fault probability and degradation probability of the government-enterprise dedicated line. When the fault prediction probability or degradation prediction probability is higher than the preset threshold, the real-time prediction model of the network hidden danger of the government-enterprise dedicated line is used to predict the network hidden danger probability of the government-enterprise dedicated line networking data in real time. This method can realize network fault prediction of customer business equipment and routing from the business layer and access layer, and combine the dedicated line network networking resource data to perform automatic inspection of networking hidden dangers and business performance degradation prediction, forming an automatic inspection capability and repair mechanism based on network data capabilities, ensuring the high quality of the business network and timely handling of hidden dangers.

[0125] During the implementation process, for the training of the government and enterprise dedicated line network fault prediction model, it is preferred to process the dedicated line basic data, dedicated line performance data, dedicated line fault data, dedicated line alarm data and dedicated line network element equipment log data through basic data cleaning, missing value filling and preprocessing, and then use the xgboost algorithm to train the government and enterprise dedicated line network fault prediction model.

[0126] For the training of the performance degradation prediction model for government and enterprise dedicated line services, it is preferred to process the transmission dedicated line performance data, Internet dedicated line performance data and voice dedicated line performance data through data cleaning and preprocessing, and then use the xgboost algorithm to train the performance degradation prediction model for government and enterprise dedicated line services.

[0127] For the training of the real-time prediction model for the hidden dangers of the government-enterprise dedicated line network, it is preferred to perform data cleaning and preprocessing, use the failure probability of the government-enterprise dedicated line network fault prediction model, the probability of performance degradation of the government-enterprise dedicated line service, and the networking model data corresponding to the dedicated line service assurance level, the dedicated line network management networking topology data, and the routing segment quality inspection data for processing, and then use the xgboost algorithm to train the real-time prediction model for the hidden dangers of the government-enterprise dedicated line network.

[0128] For the real-time prediction of network hazards of government and enterprise dedicated lines, it is preferred to use the government and enterprise dedicated line network fault prediction model and the government and enterprise dedicated line service performance degradation prediction model to perform fault prediction and degradation prediction. When the failure prediction probability or degradation prediction probability is higher than the preset threshold, the government and enterprise dedicated line network hidden danger real-time prediction model is used to perform real-time network hidden danger probability prediction on the government and enterprise dedicated line networking data.

[0129] The beneficial technical effect of the embodiments disclosed herein is that by combining dedicated line network networking resource data, network fault prediction is performed on customer business equipment and routing from the business layer and access layer, forming an automatic inspection capability and repair mechanism based on network data capabilities, ensuring the high quality of the business network and timely handling of hidden dangers.

[0130] Figure 6 The present invention is a structural diagram of a dedicated line networking hidden danger prediction device according to an exemplary embodiment.

[0131] like Figure 6 As shown, the dedicated line networking hidden danger prediction device 1 includes: a data acquisition unit 11, a first processing unit 12, a second processing unit 13 and a third processing unit 14.

[0132] The data acquisition unit 11 is used to acquire communication analysis data for predicting hidden dangers of dedicated line networking, wherein the communication analysis data includes: fault analysis data, degradation analysis data and hidden danger analysis data.

[0133] The first processing unit 12 is used to input the fault analysis data into the trained dedicated line network fault prediction model to obtain the dedicated line fault probability.

[0134] The second processing unit 13 is used to input the degradation analysis data into the trained dedicated line service performance degradation prediction model to obtain the dedicated line degradation probability.

[0135] The third processing unit 14 is used to input the hidden danger analysis data into the trained real-time prediction model of hidden dangers in dedicated line networking to obtain the hidden danger probability of dedicated line networking if the failure probability of the dedicated line is greater than the failure threshold and / or the degradation probability of the dedicated line is higher than the degradation threshold.

[0136] In some embodiments, the data acquisition unit 11 is specifically used to acquire dedicated line real-time communication data; acquire analysis data for predicting dedicated line networking risks from the dedicated line real-time communication data; and perform one-hot encoding conversion on the analysis data to generate communication analysis data.

[0137] like Figure 7 As shown, in some embodiments, the dedicated line networking hidden danger prediction device 1 further includes: a first data set acquisition unit 15 and a first model training unit 16.

[0138] The first data set acquisition unit 15 is used to acquire a fault analysis sample data set.

[0139] The first model training unit 16 is used to train the dedicated line network fault prediction model according to the fault analysis sample data set to generate a trained dedicated line network fault prediction model.

[0140] In some embodiments, the first data set acquisition unit 15 is specifically used to acquire first dedicated line data, wherein the first dedicated line data includes dedicated line basic data, dedicated line historical performance data, dedicated line historical fault data, dedicated line alarm data and dedicated line network element equipment log data; determine fault sample data from the first dedicated line data; perform one-hot encoding conversion on the fault sample data; fill missing value data in the fault sample data after the one-hot encoding conversion with zeros, and deduplicate data to obtain a fault analysis sample data set.

[0141] like Figure 8 As shown, in some embodiments, the dedicated line networking hidden danger prediction device 1 further includes: a second data set acquisition unit 17 and a second model training unit 18.

[0142] The second data set acquisition unit 17 is used to acquire a degradation analysis sample data set.

[0143] The second model training unit 18 is used to train the dedicated line service performance degradation prediction model according to the degradation analysis sample data set to generate a trained dedicated line service performance degradation prediction model.

[0144] In some embodiments, the second data set acquisition unit 17 is specifically used to acquire second dedicated line data, wherein the second dedicated line data includes transmission dedicated line performance data, Internet dedicated line performance data and voice dedicated line performance data; determine the degraded sample data from the second dedicated line data; perform one-hot encoding conversion on the degraded sample data; fill the missing value data in the degraded sample data after the one-hot encoding conversion with zeros, and deduplicate the duplicated data to obtain a degradation analysis sample data set.

[0145] like Fig. 9 As shown, in some embodiments, the dedicated line networking hidden danger prediction device 1 further includes: a third data set acquisition unit 19 and a third model training unit 20.

[0146] A third data set acquisition unit 19 is used to acquire a hidden danger analysis sample data set;

[0147] The third model training unit 20 is used to train the dedicated line networking hidden danger real-time prediction model according to the hidden danger analysis sample data set to generate a trained dedicated line networking hidden danger real-time prediction model.

[0148] In some embodiments, the third data set acquisition unit 19 is specifically used to acquire third dedicated line data, wherein the third dedicated line data includes networking model data corresponding to the dedicated line service assurance level, dedicated line network management networking topology data and routing segment quality inspection data; determine hidden danger sample data from the third dedicated line data; perform one-hot encoding conversion on the hidden danger sample data; fill missing value data in the hidden danger sample data after the one-hot encoding conversion with zeros, and deduplicate data to obtain a hidden danger analysis sample data set.

[0149] It should be noted that, in the embodiment of the present disclosure, the relevant description of the dedicated line networking hidden danger prediction device 1 can refer to the relevant description of the dedicated line networking hidden danger prediction method mentioned above, and will not be repeated here.

[0150] In the disclosed embodiment, the beneficial effects achieved by the dedicated line networking hidden danger prediction device are the same as the beneficial effects achieved by the above method. Please refer to the relevant description in the above method, which will not be repeated here.

[0151] According to an embodiment of the present disclosure, an electronic device is also provided, comprising: a processor; and a memory for storing instructions executable by the processor, wherein the processor is configured to: implement the dedicated line networking risk prediction method disclosed in the embodiment of the present disclosure.

[0152] In order to implement the above embodiment, the embodiment of the present disclosure also proposes a storage medium.

[0153] Among them, when the instructions in the storage medium are executed by the processor, the processor is enabled to execute the dedicated line networking hidden danger prediction method disclosed in the embodiment of the present disclosure.

[0154] In order to implement the above embodiments, the embodiments of the present disclosure also provide a computer program product.

[0155] Among them, when the computer program product is executed by the processor of an electronic device, it enables the electronic device to execute the dedicated line networking risk prediction method disclosed in the embodiment of the present disclosure.

[0156] Fig.10 The figure is a structural block diagram of an electronic device according to an exemplary embodiment. Figure 7 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0157] like Fig.10 As shown, the electronic device 1000 includes a processor 111, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 112 or a program loaded from a memory 116 to a random access memory (RAM) 113. In the RAM 113, various programs and data required for the operation of the electronic device 1000 are also stored. The processor 111, the ROM 112, and the RAM 113 are connected to each other via a bus 114. An input / output (I / O) interface 115 is also connected to the bus 114.

[0158] The following components are connected to the I / O interface 115: a memory 116 including a hard disk, etc.; and a communication part 117 including a network interface card such as a local area network (LAN) card, a modem, etc., and the communication part 117 performs communication processing via a network such as the Internet; a drive 118 is also connected to the I / O interface 115 as needed.

[0159] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 117. When the computer program is executed by the processor 111, the above-mentioned functions defined in the method of the present disclosure are executed.

[0160] In an exemplary embodiment, a storage medium including instructions is also provided, such as a memory including instructions, and the instructions can be executed by the processor 111 of the electronic device 1000 to complete the above method. Optionally, the storage medium can be a non-transitory computer-readable storage medium, for example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0161] In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0162] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0163] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for predicting hidden dangers in dedicated line networking, characterized in that: include: Acquire communication analysis data for predicting hidden dangers of dedicated line networking, wherein the communication analysis data includes: fault analysis data, degradation analysis data and hidden danger analysis data; Inputting the fault analysis data into a trained private line network fault prediction model to obtain a private line fault probability; Inputting the degradation analysis data into a trained dedicated line service performance degradation prediction model to obtain a dedicated line degradation probability; If the dedicated line failure probability is greater than the failure threshold, and / or the dedicated line degradation probability is higher than the degradation threshold, the hidden danger analysis data is input into the trained dedicated line networking hidden danger real-time prediction model to obtain the dedicated line networking hidden danger probability.

2. The method according to claim 1, characterized in that The obtaining of communication analysis data for predicting hidden dangers of dedicated line networking includes: Obtain real-time communication data of dedicated lines; Acquire analysis data for predicting hidden dangers of dedicated line networking from the dedicated line real-time communication data; The analysis data is converted into one-hot encoding to generate the communication analysis data.

3. The method according to claim 1 or 2, characterized in that: The method further comprises: Obtain a sample data set for fault analysis; The dedicated line network fault prediction model is trained according to the fault analysis sample data set to generate the trained dedicated line network fault prediction model.

4. The method according to claim 3, characterized in that The step of obtaining a fault analysis sample data set includes: Acquire first dedicated line data, wherein the first dedicated line data includes dedicated line basic data, dedicated line historical performance data, dedicated line historical fault data, dedicated line alarm data, and dedicated line network element device log data; Determine fault sample data from the first dedicated line data; Performing one-hot encoding conversion on the fault sample data; The missing value data in the fault sample data after the one-hot encoding conversion is filled with zeros, and duplicate data is removed to obtain the fault analysis sample data set.

5. The method according to claim 1 or 2, characterized in that: The method further comprises: Obtain a degradation analysis sample data set; The dedicated line service performance degradation prediction model is trained according to the degradation analysis sample data set to generate the trained dedicated line service performance degradation prediction model.

6. The method according to claim 5, characterized in that The step of obtaining a degradation analysis sample data set includes: Acquire second dedicated line data, wherein the second dedicated line data includes transmission dedicated line performance data, Internet dedicated line performance data, and voice dedicated line performance data; Determining degraded sample data from the second dedicated line data; Performing one-hot encoding conversion on the degraded sample data; The missing value data in the degraded sample data after the one-hot encoding conversion is filled with zeros, and the duplicate data is deduplicated to obtain the degradation analysis sample data set.

7. The method according to claim 1 or 2, characterized in that: The method further comprises: Obtain a sample data set for hidden danger analysis; The dedicated line networking hidden danger real-time prediction model is trained according to the hidden danger analysis sample data set to generate the trained dedicated line networking hidden danger real-time prediction model.

8. The method according to claim 7, characterized in that The step of obtaining a hidden danger analysis sample data set includes: Acquire third dedicated line data, wherein the third dedicated line data includes networking model data corresponding to the dedicated line service assurance level, dedicated line network management networking topology data, and routing segment quality inspection data; Determine hidden danger sample data from the third dedicated line data; Performing one-hot encoding conversion on the hidden danger sample data; The missing value data in the hidden danger sample data after the one-hot encoding conversion is filled with zeros, and the duplicate data is deduplicated to obtain the hidden danger analysis sample data set.

9. A dedicated line network hidden danger prediction device, characterized in that it includes: A data acquisition unit, used to acquire communication analysis data for predicting hidden dangers of dedicated line networking, wherein the communication analysis data includes: fault analysis data, degradation analysis data and hidden danger analysis data; A first processing unit, configured to input the fault analysis data into a trained private line network fault prediction model to obtain a private line fault probability; A second processing unit is used to input the degradation analysis data into a trained dedicated line service performance degradation prediction model to obtain a dedicated line degradation probability; The third processing unit is used to input the hidden danger analysis data into the trained real-time prediction model of the hidden danger of the dedicated line networking to obtain the hidden danger probability of the dedicated line networking if the failure probability of the dedicated line is greater than the failure threshold, and / or the degradation probability of the dedicated line is higher than the degradation threshold.

10. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 8 when executing the computer program.

Images