CN113411236B - Quality difference router detection method, quality difference router detection device, quality difference router detection equipment and storage medium - Google Patents
Quality difference router detection method, quality difference router detection device, quality difference router detection equipment and storage medium Download PDFInfo
- Publication number
- CN113411236B CN113411236B CN202110702747.1A CN202110702747A CN113411236B CN 113411236 B CN113411236 B CN 113411236B CN 202110702747 A CN202110702747 A CN 202110702747A CN 113411236 B CN113411236 B CN 113411236B
- Authority
- CN
- China
- Prior art keywords
- router
- information
- quality difference
- model
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0805—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability
- H04L43/0817—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability by checking functioning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/08—Testing, supervising or monitoring using real traffic
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Environmental & Geological Engineering (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention belongs to the technical field of communication, and discloses a quality difference router detection method, a quality difference router detection device, quality difference router detection equipment and a storage medium. According to the method, when a quality difference detection instruction is received, the router to be detected is determined according to the quality difference detection instruction; acquiring information of a lower hanging device, information of an upper connection state and information of a receiving and sending message of a router to be detected; constructing characteristic information to be detected according to the information of the lower hanging device, the information of the upper connection state and the information of the message receiving and sending; and based on the characteristic information to be detected, performing quality difference detection on the router to be detected through a preset quality difference detection model to obtain a quality difference detection result. The quality difference detection model is preset to be a pre-trained fusion model, so that the detection accuracy is high, real-time construction is not needed, whether the router is the quality difference router or not can be quickly and accurately detected, the quality difference reason of the router can be determined, and then the quality difference detection result is output.
Description
Technical Field
The present invention relates to the field of communications technologies, and in particular, to a method, an apparatus, a device, and a storage medium for quality difference router detection.
Background
Communication operators are often complained of customers due to poor signals, a large part of which are caused by poor quality routers, such as: routers in many families have not been replaced for many years, the router type is old, the supported wireless rate is low, and the network speed is slow, so that the user complaints are caused; part of the routers have the service life exceeding the service life standard, and the network speed is slow, so that the user complaints are caused; the number of the users connected to part of the routers at the same time is large, and the network speed of a single user is slow, so that the complaint of the users is caused.
At present, the detection modes of the quality difference router are mainly divided into two modes, one mode is to judge whether the current router is the quality difference router or not according to the time of appearing on the market of the router and the network rate, the other mode is to classify the detected data by adopting a data mining method, a quality difference router feature library is constructed by utilizing a large amount of data, and whether the router is the quality difference router or not is judged by utilizing the feature library.
The detection mode of the first quality difference router has obvious defects: if the standard is over-empirical, a router with a large rate meeting the contract signing speed and running stably can be misjudged as a router with poor quality; for different models and different scenes, the network rate condition of the poor quality router is difficult to analyze; the specific cause of the poor quality of the router cannot be determined.
The detection mode of the second quality difference router needs to construct a feature library in advance, needs to acquire a large amount of data, consumes a large amount of server resources to construct the feature library, has extremely high requirements on the completeness of data sources and the sufficiency of server resources of branches of communication operators, is difficult to realize and high in cost, and consumes a large amount of server resources to construct the feature library, so that the whole process consumes a long time, and the requirement of real-time judgment is difficult to meet.
The above is only for the purpose of assisting understanding of the technical aspects of the present invention, and does not represent an admission that the above is prior art.
Disclosure of Invention
The invention mainly aims to provide a quality difference router detection method, a quality difference router detection device, quality difference router detection equipment and a storage medium, and aims to solve the technical problem that the quality difference router cannot be detected quickly and accurately in the prior art.
In order to achieve the above object, the present invention provides a quality difference router detection method, which comprises the following steps:
when a quality difference detection instruction is received, determining a router to be detected according to the quality difference detection instruction;
acquiring information of the lower hanging device, information of the upper connection state and information of a message to be received of the router to be detected;
constructing characteristic information to be detected according to the information of the lower hanging device, the information of the upper connection state and the information of the receiving and sending messages;
and based on the characteristic information to be detected, performing quality difference detection on the router to be detected through a preset quality difference detection model to obtain a quality difference detection result.
Optionally, before the step of determining the router to be detected according to the quality difference detection instruction when the quality difference detection instruction is received, the method further includes:
acquiring data samples stored in a preset sample library, and constructing a data sample set according to the data samples;
training the first type initial model in the initial model set for multiple times according to the data sample set to obtain a plurality of target decision tree models;
training other types of initial models in the initial model set according to the data sample set to obtain a plurality of models to be fused;
and carrying out model fusion on the target decision tree models and the models to be fused so as to obtain a preset quality difference detection model.
Optionally, before the step of obtaining the data sample stored in the preset sample library and constructing the data sample set according to the data sample, the method further includes:
periodically acquiring router operation information of each platform user through an intelligent networking platform;
and constructing a data sample according to the router operation information, and storing the data sample into a preset sample library.
Optionally, before the step of constructing a data sample according to the router operation information and storing the data sample in a preset sample library, the method further includes:
detecting whether the router running information lacks data;
when the router running information lacks data, acquiring historical running information corresponding to the router running information;
performing data completion on the router operation information according to the historical operation information to obtain completed router operation information;
correspondingly, the step of constructing a data sample according to the router operation information and storing the data sample in a preset sample library includes:
and constructing a data sample according to the complemented router operation information, and storing the data sample into a preset sample library.
Optionally, before the step of constructing a data sample according to the router operation information and storing the data sample in a preset sample library, the method further includes:
performing data reduction on the router operation information to obtain reduced router information;
correspondingly, the step of constructing a data sample according to the router operation information and storing the data sample in a preset sample library includes:
and constructing a data sample according to the reduced router information, and storing the data sample into a preset sample library.
Optionally, the step of training the first type initial model in the initial model set multiple times according to the data sample set to obtain multiple target decision tree models includes:
constructing a plurality of data sample subsets from the set of data samples;
and training the first type initial model in the initial model set for multiple times respectively according to the data sample subsets based on the random seeds with the first preset number and the feature subsets with the second preset number to obtain a plurality of target decision tree models.
Optionally, before the step of performing model fusion on the plurality of target decision tree models and the plurality of models to be fused to obtain the preset quality difference detection model, the method further includes:
determining model fusion weights corresponding to each target decision tree model and each model to be fused according to the maximum model evaluation value;
correspondingly, the step of performing model fusion on the target decision tree models and the models to be fused to obtain a preset quality difference detection model includes:
and performing model fusion on the target decision tree models and the models to be fused based on the model fusion weight to obtain a preset quality difference detection model.
In addition, in order to achieve the above object, the present invention further provides a quality-difference router detection apparatus, including the following modules:
the instruction response module is used for determining the router to be detected according to the quality difference detection instruction when the quality difference detection instruction is received;
the information acquisition module is used for acquiring information of the off-hook equipment, the uplink state information and the message receiving and sending information of the router to be detected;
the information construction module is used for constructing the characteristic information to be detected according to the information of the lower hanging device, the information of the upper connection state and the information of the receiving and sending messages;
and the quality difference detection module is used for carrying out quality difference detection on the router to be detected through a preset quality difference detection model based on the characteristic information to be detected so as to obtain a quality difference detection result.
Further, to achieve the above object, the present invention also proposes a quality-difference router detection apparatus including: a processor, a memory and a quality router detection program stored on the memory and executable on the processor, the quality router detection program when executed by the processor implementing the steps of the quality router detection method as described above.
Furthermore, to achieve the above object, the present invention also provides a computer readable storage medium, which stores thereon a quality difference router detection program, which when executed implements the steps of the quality difference router detection method as described above.
According to the router to be detected, when a quality difference detection instruction is received, the router to be detected is determined according to the quality difference detection instruction; acquiring information of a lower hanging device, information of an upper connection state and information of a receiving and sending message of a router to be detected; constructing characteristic information to be detected according to the information of the lower hanging device, the information of the upper connection state and the information of the message receiving and sending; and based on the characteristic information to be detected, performing quality difference detection on the router to be detected through a preset quality difference detection model to obtain a quality difference detection result. The quality difference detection model is preset to be a pre-trained fusion model, the detection accuracy is high, real-time construction is not needed, whether the router is the quality difference router or not can be quickly and accurately detected, the quality difference reason of the router can be determined, then the quality difference detection result is output, and a user or a communication operator staff can clearly determine whether the router to be detected is the quality difference router or not according to the quality difference detection result and clearly determine the quality difference reason when the router to be detected is the quality difference router so as to facilitate subsequent processing.
Drawings
Fig. 1 is a schematic structural diagram of an electronic device in a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating a method for detecting a poor quality router according to a first embodiment of the present invention;
FIG. 3 is a flowchart illustrating a method for detecting a poor quality router according to a second embodiment of the present invention;
FIG. 4 is a schematic diagram illustrating a model parameter tuning process according to an embodiment of the poor quality router detection method of the present invention;
FIG. 5 is a schematic diagram of model fusion according to an embodiment of the method for detecting a difference-only router of the present invention;
fig. 6 is a block diagram of the first embodiment of the quality-difference router detection apparatus according to the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a quality-difference router detection device in a hardware operating environment according to an embodiment of the present invention.
As shown in fig. 1, the electronic device may include: a processor 1001, such as a Central Processing Unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to implement connection communication among these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a Wireless interface (e.g., a Wireless-Fidelity (WI-FI) interface). The Memory 1005 may be a Random Access Memory (RAM) Memory, or may be a Non-Volatile Memory (NVM), such as a disk Memory. The memory 1005 may alternatively be a storage device separate from the processor 1001.
Those skilled in the art will appreciate that the configuration shown in fig. 1 does not constitute a limitation of the electronic device and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components.
As shown in fig. 1, a memory 1005, which is a storage medium, may include therein an operating system, a network communication module, a user interface module, and a quality-difference router detection program.
In the electronic apparatus shown in fig. 1, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the electronic device according to the present invention may be provided in the quality-difference router detection device, and the electronic device calls the quality-difference router detection program stored in the memory 1005 through the processor 1001 and executes the quality-difference router detection method provided by the embodiment of the present invention.
An embodiment of the present invention provides a method for detecting a quality difference router, and referring to fig. 2, fig. 2 is a schematic flow diagram of a first embodiment of the method for detecting a quality difference router according to the present invention.
In this embodiment, the quality difference router detection method includes the following steps:
step S10: and when a quality difference detection instruction is received, determining the router to be detected according to the quality difference detection instruction.
It should be noted that the execution subject of this embodiment may be the quality difference router detection device, and the quality difference router detection device may be an electronic device such as a computer, a server, or other devices with the same or similar functions.
It should be noted that the quality difference detection instruction may be an instruction sent by other devices to the quality difference router detection device, the quality difference router detection instruction may have a router identifier, it is determined according to the quality difference detection instruction that the router to be detected may be the router identifier extracted from the quality difference detection instruction, and the router to be detected is searched according to the router identifier.
Step S20: and acquiring information of the lower hanging device, the upper connection state information and the message receiving and sending information of the router to be detected.
It should be noted that the drop device may be a device that accesses the router to be detected, for example: electronic equipment such as cell-phone, computer, panel computer. The drop device information may include: the information of the MAC address, the connected time length, the access frequency band value (2.4G or 5G), the signal strength, the current receiving rate and the current sending rate of negotiation, the real-time receiving rate and the sending rate of uplink and the like of the lower hanging device can reflect the service condition of the router to be detected, the frequency band preference and the WI-FI quality information of the lower hanging device through the information of the lower hanging device. The uplink state information may include: the management MAC address used by the uplink, the frequency band (2.4G or 5G) used by the WLAN uplink, the uplink signal strength during the uplink, the uplink negotiation receiving rate and sending rate, the uplink real-time receiving rate and sending rate and other information can reflect the running state of the gateway equipment of the router uplink through the uplink state information, and can be used for assisting in judging the reason of poor quality of the router. The messaging information may include: the router can comprehensively reflect the historical operation condition of the router by receiving and sending message information, can be used as supplement of real-time data such as hanging-down equipment information, and is greatly helpful for analyzing the existing operation condition and the historical operation condition.
Step S30: and constructing the characteristic information to be detected according to the information of the lower hanging device, the information of the upper connection state and the information of the receiving and sending messages.
It can be understood that the feature information to be detected is data used for inputting a preset quality difference detection model for analysis, and the preset quality difference detection model may have format requirements on the input data, so that the feature information to be detected can be constructed according to the information of the lower-hanging device, the information of the upper-link state and the information of the receiving and sending messages, so as to perform quality difference detection through the preset quality difference detection model.
In actual use, the step of constructing the feature information to be detected according to the information of the lower-mounted device, the information of the upper-connected state and the information of the transmitting and receiving messages may be to perform feature extraction on the information of the lower-mounted device, the information of the upper-connected state and the information of the transmitting and receiving messages, and the step of constructing the feature information to be detected according to the extracted feature information.
Step S40: and based on the characteristic information to be detected, performing quality difference detection on the router to be detected through a preset quality difference detection model to obtain a quality difference detection result.
It should be noted that the preset quality difference detection model may be obtained by fusing a plurality of basic models, wherein the plurality of basic models may be obtained by training a plurality of different algorithm models, and the preset quality difference detection model may be preset by a manager. The preset quality difference detection model can determine whether the router to be detected is the quality difference router or not according to the input characteristic information to be detected, and outputs a corresponding quality difference detection result. The quality difference detection result may include a quality difference router determination result and a router quality difference reason, where the quality difference router determination result may be represented by 0 or 1, where 0 represents that it is determined as a non-quality difference router, and 1 represents that it is determined as a quality difference router, and the router quality difference reason may include a WiFi coverage problem (WiFi signal of an off-hook device is weak), a WiFi interference problem (2.4G/5G co-channel interference is too large), a WiFi configuration problem (gateway/router parameter configuration is unreasonable), a WiFi usage problem (the off-hook device is too large in number and bandwidth usage is saturated), and the like.
It can be understood that the quality difference router detection device may further display the quality difference detection result after determining the quality difference detection result, so that a user or a service person of a communication operator determines whether the router to be detected is the quality difference router, and determines a reason for the quality difference when the router to be detected is the quality difference router, so as to perform subsequent processing.
In the embodiment, when the quality difference detection instruction is received, the router to be detected is determined according to the quality difference detection instruction; acquiring information of a lower hanging device, information of an upper connection state and information of a receiving and sending message of a router to be detected; constructing characteristic information to be detected according to the information of the lower hanging device, the information of the upper connection state and the information of the message receiving and sending; and based on the characteristic information to be detected, performing quality difference detection on the router to be detected through a preset quality difference detection model to obtain a quality difference detection result. The quality difference detection model is preset to be a pre-trained fusion model, the detection accuracy is high, real-time construction is not needed, whether the router is the quality difference router or not can be quickly and accurately detected, the quality difference reason of the router can be determined, then the quality difference detection result is output, and a user or a communication operator staff can clearly determine whether the router to be detected is the quality difference router or not according to the quality difference detection result and clearly determine the quality difference reason when the router to be detected is the quality difference router so as to facilitate subsequent processing.
Referring to fig. 3, fig. 3 is a flowchart illustrating a method for detecting a quality difference router according to a second embodiment of the present invention.
Based on the first embodiment, before the step S10, the method for detecting a poor quality router of this embodiment further includes:
step S01: the method comprises the steps of obtaining data samples stored in a preset sample library, and constructing a data sample set according to the data samples.
It should be noted that the preset sample library may be a preset database for storing data samples, in which a large number of data samples constructed according to the router operation information collected in advance may be stored. The data sample may include router operation information and an information label marked manually, and the information label may include whether the router is a poor quality router, a reason for the poor quality of the router, and the like. The obtaining of the data samples stored in the preset sample library and the constructing of the data sample set according to the data samples may be obtaining data samples of a preset number of samples from the preset sample library and constructing the data sample set according to a combination of the obtained data samples. Wherein, predetermine the sample quantity and can be set up by managers according to actual need, for example: the preset number of samples is set to 10000.
In practical use, in consideration of timeliness of data, data samples newly added to the preset sample library can be adopted as much as possible when the data samples are obtained from the preset sample library.
Further, in order to reduce the labor cost, before step S01, the method may further include:
periodically acquiring router operation information of each platform user through an intelligent networking platform; and constructing a data sample according to the router operation information, and storing the data sample into a preset sample library.
It should be noted that the router operation information may include information such as drop equipment information, uplink state information, and message receiving and sending information of the router, the intelligent networking platform may be a network management platform set by a communication operator, the routers of each platform user are all connected to the intelligent networking platform, the intelligent networking platform may maintain heartbeat connection with the router through an agreed protocol, and may establish connection with the routers of the platform users through a polling task and a reverse trigger, and periodically collect the router operation information of each platform router.
It can be understood that after the router operation information is acquired, the administrator may not have time to further process the acquired router operation information, so that the router operation information may also be stored in a preset database, and the router operation information may be subsequently read from the database for further processing.
It should be noted that, the constructing of the data sample according to the router operation information may be performed by manually analyzing the router operation information, determining whether the router is a poor quality router, determining a cause of the poor quality, determining an information tag, performing feature extraction on the router operation information to obtain feature information, and constructing the data sample according to the feature information and the information tag.
Further, because there may be various factors affecting during data acquisition, part of data may be missing in the acquired router operation information, if a data sample is directly constructed by the data without being processed, the accuracy and generalization capability of a model trained according to the data sample may be reduced, and if the router operation information missing data is directly deleted, data insufficiency may be caused, in order to overcome the above-mentioned defect, before the step of constructing a data sample according to the router operation information and storing the data sample in a preset sample library, the method may further include:
detecting whether the router operation information is lack of data; when the router running information lacks data, acquiring historical running information corresponding to the router running information; performing data completion on the router operation information according to the historical operation information to obtain completed router operation information;
correspondingly, the step of constructing a data sample according to the router operation information and storing the data sample in a preset sample library may include:
and constructing a data sample according to the complemented router operation information, and storing the data sample into a preset sample library.
It should be noted that detecting whether the router operation information is data may be to compare each data field in the router operation information with a preset data field to be acquired, to determine whether a field is missing, if a field is missing, determine missing data, if no field is missing, obtain a value corresponding to each field, determine whether a value corresponding to each field is present, and if a value corresponding to a field is absent or a value corresponding to a field is null (null), determine missing data.
When data is to be supplemented, there are generally two methods, the first is to predict a value of a missing part in the data by using an algorithm and fill the missing part data with the predicted value, for example: predicting data through algorithms such as a decision tree algorithm or a naive Bayes algorithm, and supplementing missing values according to predicted values; the second is to replace the missing value with the average value, for example: and acquiring data acquired for the first time and the data acquired for the last time, calculating the average value of the data for multiple times, and supplementing according to the average value.
In actual use, historical operation information corresponding to the router operation information can be acquired, data of a missing part of the currently acquired router operation information is presumed according to the historical operation information, and the missing data is supplemented according to the presumed data.
It should be noted that, because the data related to this embodiment is mainly router operating information, and data such as the speed of the router generally does not change much abnormally in a short time, that is, the variation is small, and the missing data of the drop-on device generally has a low ratio, the missing value may also be replaced by an average value, and at this time, the router operating information acquired in the first few cycles and the router operating information acquired in the last few cycles of the router operating information of the missing data may be acquired, the average value of the missing data portion is calculated, and the missing data is supplemented according to the average value.
Further, in order to improve data processing efficiency, before the step of constructing the data sample according to the router operation information and storing the data sample in a preset sample library, the method may further include
Performing data reduction on the router operation information to obtain reduced router information;
correspondingly, the step of constructing a data sample according to the router operation information and storing the data sample in a preset sample library includes:
and constructing a data sample according to the reduced router information, and storing the data sample into a preset sample library.
It should be noted that, in a scenario where the data volume is extremely large (more than a million), a long time is required for data analysis and mining, and server performance and resources are consumed, so that the acquired data can be compressed, the data is simplified, and the data integrity is not affected by a data reduction technology, which can help to improve the data quality and improve the data processing efficiency.
Data reduction in this embodiment may include at least one of the following four ways:
(1) and (3) reducing the dimension of the data, and deleting a part of data which is low in correlation or irrelevant to the quality difference analysis of the router in the operation information of the router, such as: and deleting the MAC address of the down-hanging device in the down-hanging device information.
(2) And data compression, which can reduce the data size of the collected router operation information by using an encoding mechanism.
(3) The data is reduced and smaller data is used instead of some larger data.
(4) And (3) generating a concept hierarchy, wherein higher-level attributes or data with smaller concept merging strength are used, such as: the receiving and sending rates of a plurality of drop devices are integrated into one data, instead of one receiving and sending rate for each drop device.
Step S02: and training the first type initial model in the initial model set for multiple times according to the data sample set to obtain a plurality of target decision tree models.
It should be noted that a plurality of initial models of different types may be stored in the initial model set, and the initial models of different types may be models constructed according to different algorithms. The first type of initial model may be a model constructed according to a Gradient Boosting Decision Tree (GBDT) algorithm, where the Gradient Boosting Decision Tree algorithm is an iterative Decision Tree algorithm whose core is that a plurality of weak learners are repeatedly trained and iterated, and finally the weak learners are combined into a strong learner, and the performance of the strong learner is greatly improved compared with that of a single weak learner.
Further, in order to improve the effect of model fusion, step S02 of the present embodiment may include:
constructing a plurality of data sample subsets from the set of data samples; and training the first type initial model in the initial model set for multiple times respectively according to the data sample subsets based on the random seeds with the first preset number and the feature subsets with the second preset number to obtain a plurality of target decision tree models.
It should be noted that, the constructing of the plurality of data sample subsets according to the data sample set may be selecting data samples with a preset number of subset samples from the data sample set to construct the data sample subsets, and the data sample subsets may be repeatedly executed for multiple times, where in order to ensure the difference of each target decision tree model, improve the effect of model fusion, and possibly ensure the difference of data samples in each data sample subset.
It should be noted that the random seeds may affect the samples examined by the single tree in the decision tree and the features selected when splitting the nodes, and the random seeds are fixed in the training process, so that the same result can be obtained under the condition that the random seeds are not changed, and the feature subsets may be manually set for grouping the features, so as to indicate which part of the features in the data samples are used for model training. The first preset number and the second preset number may be set by a manager according to actual needs, for example: the first preset number is set to 30 and the second preset number is set to 10.
In practical use, in order to ensure that each target decision tree model obtained by training has certain difference as much as possible so as to obtain the best training result, each random seed may be different from each other, and each feature subset is also different from each other.
In practical use, the first type initial model may be a model constructed according to the GBDT algorithm, the parameters of which are shown in table 1 below:
TABLE 1GBDT parameters
In the actual training process, the parameter tuning process for the GBDT model may be as shown in fig. 4.
In practical use, a data sample subset can be selected from a plurality of data sample subsets, then the first type initial model is trained according to a first preset number of random seeds and a second preset number of feature subsets based on the selected data sample subset, a plurality of base learners are obtained, and then the base learners are subjected to equal-ratio fusion (due to the fact that adopted algorithms are consistent and the accuracy is basically similar, the base learners can be subjected to equal-ratio fusion according to a voting method without setting weights), so that a target decision tree model is obtained. And repeating the steps to obtain a plurality of target decision tree models according to the plurality of data sample subsets.
For example: assuming that the number of the data sample subsets is A, B, C, the number of the random seeds is four, namely, S1, S2, S3 and S4, and the number of the feature subsets is 3, namely, T1, T2 and T3, the data sample subset a can be selected first, the first type initial model is trained according to a and S1, parameters are adjusted continuously, the base learner a-S1 is obtained, the base learners a-S1 are executed for multiple times, then 7 base learners a-S1, a-S2, a-S3, a-S4, a-T1, a-T2 and a-T3 are obtained, and then the 7 base learners are fused in equal ratios, so that the target decision tree model MA corresponding to the data sample subset can be obtained, the above steps are repeated, and then the MB and the MC are obtained, so that a plurality of target decision tree models are obtained.
Step S03: and training other types of initial models in the initial model set according to the data sample set to obtain a plurality of models to be fused.
It should be noted that there may be a plurality of other types of initial models, and each of the initial models may be an initial model constructed by using a different algorithm, for example: other types of initial models may have 3, each of which is an initial model constructed by an AdaBoost algorithm, a vector machine algorithm, or a Random Forest algorithm (RF).
Step S04: and carrying out model fusion on the target decision tree models and the models to be fused so as to obtain a preset quality difference detection model.
It should be noted that, when performing model fusion on a plurality of target decision tree models and a plurality of models to be fused, a Voting method (Voting) may be used for the fusion, where the Voting method may be a Hard Voting method (Hard Voting) or a Soft Voting method (Soft Voting), and this embodiment does not limit this.
It can be understood that the preset quality difference detection model obtained by model fusion of the plurality of target decision tree models and the plurality of models to be fused has better comprehensive performance compared with a single target decision tree model or a single model to be fused.
Further, since each model to be fused is obtained by training an initial model constructed based on different algorithms, and the initial model is different from the algorithm used by the target decision tree model, the accuracy may be different, if the model is still fused by using an equal ratio at this time, the effect of the fused model may not be optimal, and in order to ensure the using effect of the fused model, before the step of performing model fusion on the plurality of target decision tree models and the plurality of models to be fused to obtain the preset quality difference detection model, the method may further include:
determining model fusion weights corresponding to each target decision tree model and each model to be fused according to the maximum model evaluation value;
correspondingly, the step of performing model fusion on the target decision tree models and the models to be fused to obtain a preset quality difference detection model includes:
and performing model fusion on the target decision tree models and the models to be fused based on the model fusion weight to obtain a preset quality difference detection model.
It should be noted that the model evaluation Score can be an F1 Score (F1 Score), and the F1 Score can take into account both the accuracy and the recall of the model. Taking the maximum model evaluation score as a basis, determining the model fusion weights corresponding to each target decision tree model and each model to be fused may be that the weights of the models are all set to 1, then fusion is performed, a test set is used for testing and calculating F1 scores, then the weights of the models are adjusted according to a preset step length, the test is continuously performed and the F1 scores are calculated, and the weight value of each model when the F1 score is maximum is taken as the model fusion weight of each model.
It can be understood that, based on the model fusion weight, the multiple target decision tree models and the multiple models to be fused are subjected to model fusion, and the obtained preset quality difference detection model can ensure that the F1 score is maximum, that is, the comprehensive performance of the model is optimal, and the use effect is best.
For convenience of understanding, the description is made with reference to fig. 5, but the present solution is not limited thereto, and fig. 5 is a schematic diagram of model fusion, in which 30 different random seeds and 10 different feature subsets are set, an initial model constructed according to the GBDT algorithm is trained according to the data sample subsets to obtain a first objective decision tree model, which is used as a Voting Classifier (GBDT1), then a Voting Classifier (GBDT2) and a Voting Classifier (GBDT3) are obtained by training the initial model constructed according to the GBDT algorithm according to the data sample subsets, then a model to be fused is obtained by training the initial model constructed according to the AdaBoost algorithm according to the data sample set to obtain a model to be fused, which is used as a Voting Classifier (Ada), then the initial model constructed according to the random algorithm is trained according to the data sample set to obtain a model to be fused, which is used as a Voting Classifier (RF), and then model fusion weights are determined, and fusing the models according to the model fusion weight to obtain a fused voting classifier, and taking the fused voting classifier as a FINAL RESULT (FINAL RESULT) to obtain a preset quality difference detection model.
In the embodiment, a data sample stored in a preset sample library is obtained, and a data sample set is constructed according to the data sample; training the first type initial model in the initial model set for multiple times according to the data sample set to obtain a plurality of target decision tree models; training other types of initial models in the initial model set according to the data sample set to obtain a plurality of models to be fused; and carrying out model fusion on the target decision tree models and the models to be fused so as to obtain a preset quality difference detection model. Because the model constructed by a single algorithm is not adopted, but the models constructed by a plurality of different algorithms are fused after the training is finished, the comprehensive performance of the preset quality difference detection model is better.
Furthermore, an embodiment of the present invention further provides a storage medium, where the storage medium stores a quality difference router detection program, and the quality difference router detection program, when executed by a processor, implements the steps of the quality difference router detection method described above.
Referring to fig. 6, fig. 6 is a block diagram illustrating a first embodiment of the poor quality router detection apparatus according to the present invention.
As shown in fig. 6, the quality-difference router detection apparatus provided in the embodiment of the present invention includes:
the instruction response module 601 is configured to determine, when receiving the quality difference detection instruction, a router to be detected according to the quality difference detection instruction;
an information obtaining module 602, configured to collect information of the under-hung device, information of an uplink state, and information of a message to be received of the router to be detected;
an information construction module 603, configured to construct feature information to be detected according to the information of the drop device, the uplink state information, and the message receiving and sending information;
and a quality difference detection module 604, configured to perform quality difference detection on the router to be detected through a preset quality difference detection model based on the characteristic information to be detected, so as to obtain a quality difference detection result.
In the embodiment, when the quality difference detection instruction is received, the router to be detected is determined according to the quality difference detection instruction; acquiring information of a lower hanging device, information of an upper connection state and information of a receiving and sending message of a router to be detected; constructing characteristic information to be detected according to the information of the lower hanging device, the information of the upper connection state and the information of the message receiving and sending; and based on the characteristic information to be detected, performing quality difference detection on the router to be detected through a preset quality difference detection model to obtain a quality difference detection result. The quality difference detection model is preset to be a pre-trained fusion model, the detection accuracy is high, real-time construction is not needed, whether the router is the quality difference router or not can be quickly and accurately detected, the quality difference reason of the router can be determined, then the quality difference detection result is output, and a user or a communication operator staff can clearly determine whether the router to be detected is the quality difference router or not according to the quality difference detection result and clearly determine the quality difference reason when the router to be detected is the quality difference router so as to facilitate subsequent processing.
Further, the instruction response module 601 is further configured to obtain data samples stored in a preset sample library, and construct a data sample set according to the data samples; training the first type initial model in the initial model set for multiple times according to the data sample set to obtain a plurality of target decision tree models; training other types of initial models in the initial model set according to the data sample set to obtain a plurality of models to be fused; and carrying out model fusion on the target decision tree models and the models to be fused so as to obtain a preset quality difference detection model.
Further, the instruction response module 601 is further configured to periodically acquire router operation information of each platform user through the intelligent networking platform; and constructing a data sample according to the router operation information, and storing the data sample into a preset sample library.
Further, the instruction response module 601 is further configured to detect whether the router operation information lacks data; when the router running information lacks data, acquiring historical running information corresponding to the router running information; performing data completion on the router operation information according to the historical operation information to obtain completed router operation information;
the instruction response module 601 is further configured to construct a data sample according to the completed router operation information, and store the data sample in a preset sample library.
Further, the instruction response module 601 is further configured to perform data reduction on the router operation information to obtain reduced router information;
the instruction response module 601 is further configured to construct a data sample according to the reduced router information, and store the data sample in a preset sample library.
Further, the instruction response module 601 is further configured to construct a plurality of data sample subsets according to the data sample set; and training the first type initial model in the initial model set for multiple times respectively according to the data sample subsets based on the random seeds with the first preset number and the feature subsets with the second preset number to obtain a plurality of target decision tree models.
Further, the instruction response module 601 is further configured to determine model fusion weights corresponding to each target decision tree model and each model to be fused based on the maximized model evaluation score;
the instruction response module 601 is further configured to perform model fusion on the multiple target decision tree models and the multiple models to be fused based on the model fusion weight, so as to obtain a preset quality difference detection model.
It should be understood that the above is only an example, and the technical solution of the present invention is not limited in any way, and in a specific application, a person skilled in the art may set the technical solution as needed, and the present invention is not limited thereto.
It should be noted that the above-described work flows are only exemplary, and do not limit the scope of the present invention, and in practical applications, a person skilled in the art may select some or all of them to achieve the purpose of the solution of the embodiment according to actual needs, and the present invention is not limited herein.
In addition, the technical details that are not described in detail in this embodiment may refer to the method for detecting a quality difference router provided in any embodiment of the present invention, and are not described herein again.
Further, it is to be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention or portions thereof that contribute to the prior art may be embodied in the form of a software product, where the computer software product is stored in a storage medium (e.g. Read Only Memory (ROM)/RAM, magnetic disk, optical disk), and includes several instructions for enabling a terminal device (e.g. a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (9)
1. A quality difference router detection method is characterized by comprising the following steps:
when a quality difference detection instruction is received, determining a router to be detected according to the quality difference detection instruction;
acquiring information of the lower hanging device, information of the upper connection state and information of a message to be received of the router to be detected;
constructing characteristic information to be detected according to the information of the lower hanging device, the information of the upper connection state and the information of the receiving and sending messages;
based on the characteristic information to be detected, performing quality difference detection on the router to be detected through a preset quality difference detection model to obtain a quality difference detection result;
before the step of determining the router to be detected according to the quality difference detection instruction when the quality difference detection instruction is received, the method further includes:
acquiring data samples stored in a preset sample library, and constructing a data sample set according to the data samples;
training the first type initial model in the initial model set for multiple times according to the data sample set to obtain a plurality of target decision tree models;
training other types of initial models in the initial model set according to the data sample set to obtain a plurality of models to be fused;
and carrying out model fusion on the target decision tree models and the models to be fused so as to obtain a preset quality difference detection model.
2. The method for detecting a poor quality router according to claim 1, wherein before the step of obtaining the data samples stored in the preset sample library and constructing the data sample set according to the data samples, the method further comprises:
periodically acquiring router operation information of each platform user through an intelligent networking platform;
and constructing a data sample according to the router operation information, and storing the data sample into a preset sample library.
3. The method for detecting a poor quality router according to claim 2, wherein before the step of constructing the data sample according to the router operation information and storing the data sample in a preset sample library, the method further comprises:
detecting whether the router operation information is lack of data;
when the router running information lacks data, acquiring historical running information corresponding to the router running information;
performing data completion on the router operation information according to the historical operation information to obtain completed router operation information;
correspondingly, the step of constructing a data sample according to the router operation information and storing the data sample in a preset sample library includes:
and constructing a data sample according to the complemented router operation information, and storing the data sample into a preset sample library.
4. The method for detecting a poor quality router according to claim 2, wherein before the step of constructing the data sample according to the router operation information and storing the data sample in a preset sample library, the method further comprises:
performing data reduction on the router operation information to obtain reduced router information;
correspondingly, the step of constructing a data sample according to the router operation information and storing the data sample in a preset sample library includes:
and constructing a data sample according to the reduced router information, and storing the data sample into a preset sample library.
5. The method of detecting a poor quality router of claim 1 wherein the step of training a first type of initial model in an initial set of models a plurality of times based on the set of data samples to obtain a plurality of objective decision tree models comprises:
constructing a plurality of data sample subsets from the set of data samples;
and training the first type initial model in the initial model set for multiple times respectively according to the data sample subsets based on the random seeds with the first preset number and the feature subsets with the second preset number to obtain a plurality of target decision tree models.
6. The method according to claim 1, wherein before the step of performing model fusion on the target decision tree models and the models to be fused to obtain the preset quality difference detection model, the method further comprises:
determining model fusion weights corresponding to each target decision tree model and each model to be fused according to the maximum model evaluation value;
correspondingly, the step of performing model fusion on the target decision tree models and the models to be fused to obtain a preset quality difference detection model includes:
and performing model fusion on the target decision tree models and the models to be fused based on the model fusion weight to obtain a preset quality difference detection model.
7. A quality difference router detection apparatus, comprising:
the instruction response module is used for determining the router to be detected according to the quality difference detection instruction when the quality difference detection instruction is received;
the information acquisition module is used for acquiring information of the lower hanging device, information of the upper connection state and information of a receiving and sending message of the router to be detected;
the information construction module is used for constructing the characteristic information to be detected according to the information of the lower hanging device, the information of the upper connection state and the information of the receiving and sending messages;
the quality difference detection module is used for carrying out quality difference detection on the router to be detected through a preset quality difference detection model based on the characteristic information to be detected so as to obtain a quality difference detection result;
the instruction response module is further used for obtaining data samples stored in a preset sample library and constructing a data sample set according to the data samples; training the first type initial model in the initial model set for multiple times according to the data sample set to obtain a plurality of target decision tree models; training other types of initial models in the initial model set according to the data sample set to obtain a plurality of models to be fused; and carrying out model fusion on the target decision tree models and the models to be fused so as to obtain a preset quality difference detection model.
8. A quality-difference router detection apparatus, characterized in that the quality-difference router detection apparatus comprises: processor, memory and a quality router detection program stored on the memory and executable on the processor, the quality router detection program when executed by the processor implementing the steps of the quality router detection method according to any of claims 1-6.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a quality-difference router detection program, which when executed performs the steps of the quality-difference router detection method according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110702747.1A CN113411236B (en) | 2021-06-23 | 2021-06-23 | Quality difference router detection method, quality difference router detection device, quality difference router detection equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110702747.1A CN113411236B (en) | 2021-06-23 | 2021-06-23 | Quality difference router detection method, quality difference router detection device, quality difference router detection equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113411236A CN113411236A (en) | 2021-09-17 |
| CN113411236B true CN113411236B (en) | 2022-06-14 |
Family
ID=77682941
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110702747.1A Active CN113411236B (en) | 2021-06-23 | 2021-06-23 | Quality difference router detection method, quality difference router detection device, quality difference router detection equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113411236B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115065606B (en) * | 2022-05-31 | 2023-10-27 | 中移（杭州）信息技术有限公司 | Home-wide mass spectrometry method, device, equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3223486A2 (en) * | 2016-03-25 | 2017-09-27 | Cisco Technology, Inc. | Distributed anomaly detection management |
| WO2021020198A1 (en) * | 2019-07-26 | 2021-02-04 | 富士フイルム株式会社 | Information processing device, program, learned model, diagnostic assistance device, learning device, and method for generating prediction model |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8139475B2 (en) * | 2004-07-29 | 2012-03-20 | Telecom Italia S.P.A. | Method and system for fault and performance recovery in communication networks, related network and computer program product therefor |
| CN101841475B (en) * | 2010-04-21 | 2011-12-28 | 北京星网锐捷网络技术有限公司 | System for testing performance of GRE (Ground Radar Equipment) router and method thereof |
| CN104639389A (en) * | 2014-12-31 | 2015-05-20 | 北京奇虎科技有限公司 | Router evaluating device and method |
| CN106604312B (en) * | 2016-11-10 | 2020-12-22 | 王蕴卓 | Router test data visualization method and system |
| CN106452975B (en) * | 2016-11-18 | 2019-10-11 | 上海斐讯数据通信技术有限公司 | A kind of method and system of test router |
| CN107094104B (en) * | 2017-04-07 | 2019-10-29 | 杭州东信北邮信息技术有限公司 | A kind of test method and system of WIFI router load-carrying properties |
| CN111988201B (en) * | 2020-09-01 | 2021-10-22 | 杭州东方通信软件技术有限公司 | Automatic identification method and device for quality difference router |
-
2021
- 2021-06-23 CN CN202110702747.1A patent/CN113411236B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3223486A2 (en) * | 2016-03-25 | 2017-09-27 | Cisco Technology, Inc. | Distributed anomaly detection management |
| WO2021020198A1 (en) * | 2019-07-26 | 2021-02-04 | 富士フイルム株式会社 | Information processing device, program, learned model, diagnostic assistance device, learning device, and method for generating prediction model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113411236A (en) | 2021-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111325463A (en) | Data quality detection method, device, equipment and computer readable storage medium | |
| EP3169018B1 (en) | Method and device for establishing performance measurement task and processing performance measurement result | |
| CN107809766B (en) | Method and device for generating machine learning sample for network optimization | |
| WO2019184640A1 (en) | Indicator determination method and related device thereto | |
| CN113869521A (en) | Method, device, computing equipment and storage medium for constructing prediction model | |
| CN112187512B (en) | Port automatic expansion method, device and equipment based on flow monitoring | |
| WO2014040633A1 (en) | Identifying fault category patterns in a communication network | |
| US11558769B2 (en) | Estimating apparatus, system, method, and computer-readable medium, and learning apparatus, method, and computer-readable medium | |
| WO2022061900A1 (en) | Method for determining fault autonomy capability and related device | |
| CN103891383A (en) | Profiling wi-fi channel congestion and interference to optimize channel selection | |
| CN104427547A (en) | Business and network association test method, device and system | |
| EP3101842A1 (en) | System and method for network management automation | |
| CN113411236B (en) | Quality difference router detection method, quality difference router detection device, quality difference router detection equipment and storage medium | |
| CN110856188B (en) | Communication method, apparatus, system, and computer-readable storage medium | |
| CN108111346A (en) | The definite method, apparatus and storage medium of frequent item set in warning association analysis | |
| CN113727092B (en) | Video monitoring quality inspection method and device based on decision tree | |
| CN113660687B (en) | Network difference cell processing method, device, equipment and storage medium | |
| Zhang et al. | Workload patterns for quality-driven dynamic cloud service configuration and auto-scaling | |
| CN112785418B (en) | Credit risk modeling method, apparatus, device and computer readable storage medium | |
| CN109788251A (en) | Method for processing video frequency, device and storage medium | |
| CN114244691A (en) | Video service fault positioning method and device and electronic equipment | |
| CN112437469B (en) | Quality of service guarantee method, apparatus and computer readable storage medium | |
| US20230188440A1 (en) | Automatic classification of correlated anomalies from a network through interpretable clustering | |
| CN116303072A (en) | Full-link performance test method and device | |
| CN110544182B (en) | Power distribution communication network fusion control method and system based on machine learning technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |