CN109040155A - Asset identification method and computer equipment - Google Patents
Asset identification method and computer equipment Download PDFInfo
- Publication number
- CN109040155A CN109040155A CN201710428726.9A CN201710428726A CN109040155A CN 109040155 A CN109040155 A CN 109040155A CN 201710428726 A CN201710428726 A CN 201710428726A CN 109040155 A CN109040155 A CN 109040155A
- Authority
- CN
- China
- Prior art keywords
- assets
- data
- space
- transmission data
- identified
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/51—Discovery or management thereof, e.g. service location protocol [SLP] or web services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Signal Processing (AREA)
- Game Theory and Decision Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the present invention provides a kind of asset identification method and computer equipment.The described method includes: obtaining the first transmission data of the assets to be identified in preset time period, the first transmission data include the amount of assets carried out data transmission with assets to be identified and data packet number；First transmission data are mapped into the first space, determine the first transmission data in the position in the first space；The system type of the corresponding assets to be identified of the first transmission data is determined in the corresponding relationship of the position in the first space in the position in the first space and predetermined system type and first sample transmission data according to the first transmission data.The method transmits data by the first of assets to be identified, the first transmission data are obtained in the position in the first space, and data are transmitted in the corresponding relationship of the position in the first space according to system type and first sample, determine the system type of assets to be identified, thus it can reach system type belonging to the identification assets of automatic intelligent, to improve working efficiency.
Description
Technical field
The present embodiments relate to a kind of information technology field, especially a kind of asset identification method and computer equipment.
Background technique
Business refers to enterprise and organizes a series of summation of processes such as production and operating activities, issued transaction.
With the introducing of information technology, business is close with IT (Information Technology, information technology)
Coupling is together.
From the perspective of IT, business includes IT support system (abbreviation business support system), the business datum, industry of business
The participant for the process and business of being engaged in.Wherein, business support system is the foundation stone of business, includes the various of bearer service operation
Software and hardware IT resource, such as the network equipment, safety equipment, host, database, middleware etc..
The O&M department liable hardware assets of business and the management of software asset, by the hardware of equipment, software and the two
In conjunction with being referred to as assets.These IT resources combine, and one group of generation particular customer value of shared is appointed
Business, is formed business support system.
Business support system topological diagram, be exactly on traditional asset management, using business as tie build assets it
Between correlation view.Based on service topology, user open-and-shut can know system belonging to each assets clearly, and understand
The current operation conditions of the business of Capital operation and safe condition.In service topology, using visual icon representation assets
Virtual condition, can visually see very much the state of assets, be normal or unavailable, or there is alarm.If industry
Business is broken down, actually or can rapidly check the host database that goes wrong goes wrong, or exchange
Machine goes wrong, and quickly and easily carries out traffic failure diagnosis along service topology.
The molding for the first time of business support system topological diagram is substantially to plan when constructing Integrated Solution at system Construction initial stage,
Again by manually drawing.But after system is formally online, it may be become due to performance issue, operation expanding etc.
It more adjusts, thus business support system topological diagram is also required to accordingly be updated.
Based on the update mode of business support system topological diagram is mainly updated with manual maintenance in the prior art, according to process
The more new data of circulation, or the more new data got based on automatic collection pass through artificial side for assets to be identified
Formula determines system belonging to assets to be identified, matches topological structure one by one, to achieve the purpose that update system topological figure.
It is understood that system topological figure update mode has following defects that asset replacement data dependence manually repeats
Identification, inefficiency.
Currently, there are no corresponding methods to solve the problems, such as manual identified inefficiency for the prior art.
Summary of the invention
In view of the drawbacks of the prior art, the embodiment of the present invention provides a kind of asset identification method and computer equipment.
On the one hand, the embodiment of the present invention provides a kind of asset identification method, comprising: obtains to be identified in preset time period
First transmission data of assets, the first transmission data include the amount of assets carried out data transmission with the assets to be identified
And data packet number；The first transmission data are mapped into the first space, determine the first transmission data in the first space
Position；It is passed according to the first transmission data in the position in the first space and predetermined system type and first sample
Transmission of data determines the system class of the corresponding assets to be identified of the first transmission data in the corresponding relationship of the position in the first space
Type.
On the other hand, the embodiment of the present invention also provides a kind of computer equipment, including memory, processor, bus and
Store the computer program that can be run on a memory and on a processor, which is characterized in that the processor executes the journey
Following methods are realized when sequence:
Obtain preset time period in assets to be identified first transmission data, it is described first transmission data include with it is described
The amount of assets and data packet number that assets to be identified carry out data transmission；The first transmission data are mapped into the first sky
Between, determine the first transmission data in the position in the first space；Data are transmitted in the position in the first space according to described first
And predetermined system type and first sample transmission data determine described the in the corresponding relationship of the position in the first space
The system type of the corresponding assets to be identified of one transmission data.
First space has predetermined logistic regression curve, the logistic regression curve and preset system class
Type is corresponding, and is to transmit what data determined according to the first sample of the system；Correspondingly, the determination the first transmission number
According to the system type of corresponding assets to be identified, specifically: according to the first transmission data in the first space and the logic
The relative position of regression curve, determines whether the corresponding assets to be identified of the first transmission data belong to the system.
After the system type of the corresponding assets to be identified of data is transmitted in the determination described first, the method is also wrapped
It includes:
The second transmission data of the assets to be identified in preset time period are obtained, second data include described to be identified
The application user type of assets carry out data transmission in assets and the system number-of-packet and the assets in the system；It will
The second transmission data map to second space, determine the second transmission data in the position of second space；According to described
Second transmission data are in the position of second space and predetermined Asset Type and the second sample delivery data in the second sky
Between position corresponding relationship, determine the Asset Type of the corresponding assets to be identified of the second transmission data.
The second space has at least one predetermined mass center, and the mass center is corresponding with preset Asset Type,
It and is determined according to the second sample delivery data of the assets；
Correspondingly, determine the Asset Type of the corresponding assets to be identified of the second transmission data, specifically: will with it is described
Assets class of the second transmission data in second space apart from the nearest corresponding Asset Type of mass center, as the assets to be identified
Type.
After the Asset Type of the corresponding assets to be identified of determination the second sample delivery data, the method is also
Include:
The third for obtaining the assets to be identified in preset time period transmits data, third transmission data include it is described to
Identify the data packet number of the transmission between assets and the assets of Asset Type of the same race, the cluster include it is multiple with described wait know
The identical assets of Asset Type of other assets；Third transmission data are mapped into third space, determine the third transmission
Position of the data in third space；Position and predetermined collection realm of the data in third space are transmitted according to the third
Type and third sample delivery data determine that the third transmission data are corresponding wait know in the corresponding relationship of the position in third space
The group type of other assets.
The third space has predetermined polynomial fitting curve, the polynomial fitting curve and preset collection realm
Type is corresponding, and is determined according to the third sample delivery data of the cluster；Correspondingly, described that the third is transmitted into data
Third space is mapped to, determines that the third transmission data in the position in third space, specially obtain the third transmission number
According to the matched curve in third space；
It is described that data are transmitted in the position in third space and predetermined group type and third according to the third
Sample delivery data determine the corresponding assets to be identified of the third transmission data in the corresponding relationship of the position in third space
Group type, specifically:
Determine the fitting coefficient of matched curve Yu the predetermined matched curve of the third transmission data；It is by fitting
The corresponding group type of matched curve when number is maximum, the group type as assets to be identified.
As shown from the above technical solution, asset identification method provided in an embodiment of the present invention, Operation Server and business clothes
Business device, the method transmit data by the first of assets to be identified, obtain the first transmission data in the position in the first space, and
Corresponding relationship according to system type and first sample transmission data in the position in the first space, the system for determining assets to be identified
Thus type can reach system type belonging to the identification assets of automatic intelligent, to improve working efficiency.
Detailed description of the invention
Fig. 1 is a kind of flow diagram of asset identification method provided in an embodiment of the present invention；
Fig. 2 is a kind of flow diagram for asset identification method that further embodiment of this invention provides；
Fig. 3 is the image for the sigmoid function that further embodiment of this invention provides；
Fig. 4 is that the logistic regression that a kind of machine training for asset identification method that further embodiment of this invention provides obtains is bent
Line schematic diagram；
Fig. 5 is that a kind of system ownership for asset identification method that further embodiment of this invention provides judges automatically schematic diagram；
Fig. 6 is a kind of flow diagram for asset identification method that further embodiment of this invention provides；
Fig. 7 is a kind of flow diagram for asset identification method that further embodiment of this invention provides；
Fig. 8-11 is respectively a kind of k-means algorithm signal for asset identification method that further embodiment of this invention provides
Figure；
Figure 12 is the server money that a kind of machine training for asset identification method that further embodiment of this invention provides obtains
It produces class and belongs to model schematic；
A kind of server assets class ownership model judgement for asset identification method that Figure 13 further embodiment of this invention provides
Schematic diagram；
Figure 14 is a kind of flow diagram for asset identification method that further embodiment of this invention provides；
Figure 15 is a kind of flow diagram for asset identification method that further embodiment of this invention provides；
Figure 16 is a kind of polynomial fitting curve of a cluster of asset identification method that further embodiment of this invention provides
Schematic diagram；
Multiple cluster polynomial fittings are bent in a kind of system for asset identification method that Figure 17 further embodiment of this invention provides
Line schematic diagram；
A kind of cluster ownership model for asset identification method that Figure 18 further embodiment of this invention provides judges schematic diagram；
Figure 19 is a kind of asset identification systems schematic diagram that further embodiment of this invention provides；
Figure 20 is a kind of structural schematic diagram for computer equipment that further embodiment of this invention provides.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical solution in the embodiment of the present invention is explicitly described, it is clear that described embodiment is the present invention
Embodiment a part of the embodiment, instead of all the embodiments.
In the present embodiment, by the combination of the hardware of equipment, software and the two, assets are referred to as.
For purposes of illustration only, being illustrated so that assets are servers as an example.
It, may be to each money since performance issue, operation expanding etc. need during the operation maintenance of business support system
Production changes adjustment, so as to cause there are unknown assets can be provided based on machine learning for assets to be identified in process
Identification is produced, with system belonging to determination assets to be identified.
Business support system topological diagram includes multiple application systems, such as CM system (Customer Relationship
Management, client management system), BOSS system (Business&Operation Support System, service operation
Support system), customer service system etc..
Fig. 1 shows a kind of flow diagram of asset identification method provided in an embodiment of the present invention.
Referring to Fig.1, method provided in an embodiment of the present invention specifically includes the following steps:
Step 11, obtain preset time period in assets to be identified first transmission data, it is described first transmission data packet
Include the amount of assets carried out data transmission with the assets to be identified and data packet number.
Optionally, for assets to be identified, based on the mark of assets to be identified, undertaking to perform work within a time limit and according to specifications to have by network obtains preset time
First transmission data of the assets to be identified in section.
Wherein, the assets to be identified can be server to be identified, and the amount of assets is and server to be identified
The quantity of other servers carried out data transmission, the data packet number is between server to be identified and other servers
The quantity of the data packet of transmission.
Network job contract tool can obtain the data packet of the assets to be identified and other assets transmission, analyze data packet
It is available that there are the other assets information of incidence relation with the assets to be identified.
Specifically, refer to there are incidence relation and transmitted with the assets to be identified.Other assets information includes other
The quantity of assets, that is, how many assets is transmitted with the assets to be identified and the attribute of other assets.Other moneys
The attribute of production is the parameters such as system belonging to each assets, Asset Type, cluster belonging to assets.In addition, dividing data packet
The data packet number of the assets transmission to be identified also can be obtained in analysis.
For example, it based on unknown server ip, is grabbed by tcpdump order and passes through the service in designated time period
The flow packet of device network interface card.
Wherein, Tcpdump (dump the traffic on a network), according to the definition of user on network
The packet analysis tool intercepted and captured of data packet.It supports to be directed to network layer, agreement, host, network or the filtering of port, and provides
The logical statements such as and, or, not.
Specifically, flow packet includes system time, comes source host, port > destination host, port, packet parameter etc., solution
Flow packet is precipitated and neutralizes other IP of the IP there are incidence relation, and counts the number that the IP is associated with flow packet between IP with other
Amount.
The first transmission data are mapped to the first space by step 12, determine the first transmission data in the first sky
Between position.
In the present embodiment, the mode that a variety of data visualizations can be used obtains the first transmission data in the first space
Position.
Optionally, first space can be two-dimensional coordinate system, and the first transmission data are mapped to a two dimension and are sat
Mark system mode can there are many.The ordinate of the transmission data can be determined, and according to the data according to the amount of assets
Packet quantity determines the abscissa of the transmission data, obtains determining the first transmission data in the position in the first space.
Step 13, according to the first transmission data in the position in the first space and predetermined system type and the
One sample delivery data determine the corresponding assets to be identified of the first transmission data in the corresponding relationship of the position in the first space
System type.
Before this step, predetermined system type and first sample transmission data are obtained in the position in the first space
Corresponding relationship, the corresponding relationship is obtained by machine learning algorithm.
Optionally, it is based on known server ip, is grabbed by tcpdump order and passes through the server in designated time period
The flow packet of network interface card, parsing outflow packet neutralizes other IP of the IP there are incidence relation, and counts the number of flow packet between IP
Amount.
The IP and asset database that parse are compared, the asset database, which refers to, is stored with all O&M departments
The database of responsible device hardware and software configuration information, each server assets have at least one IP record, according to
Lookup obtains unknown assets, and with statistical method, summarizes the relational matrix between unknown assets and known assets, as follows
Table 1.
In table, horizontal item is unknown assets, and list is known assets, and list item is the quantity of flow packet.
For example, the quantity for mutually sending out flow packet in acquisition time section between two assets of w1 and a1 is 5.
By the asset history data of known system, data are transmitted as first sample, are instructed by machine learning algorithm
Practice, obtains the system ownership model of known assets, i.e. each system type and first sample transmission data is in the first space
The corresponding relationship of position, by the first transmission data in the position in the first space, with first sample transmission data in the first sky
Between position matched, if matching, can determine that the corresponding assets to be identified of the first transmission data are attributed to the system, if
It mismatches, then is matched with another first sample transmission data in the position in the first space, and so on, thus, it is possible to by being somebody's turn to do
Model judges automatically whether new unknown assets belong to the system.
Asset identification method provided in this embodiment, at least has following technical effect that
By obtaining the first transmission data in the position in the first space according to the first of assets to be identified the transmission data
It sets, and is determined according to predetermined system type and first sample transmission data in the corresponding relationship of the position in the first space
Described first transmits the system type of the corresponding assets to be identified of data, thus can reach belonging to the identification assets of automatic intelligent
System type, to improve working efficiency.
Fig. 2 shows a kind of flow diagrams for asset identification method that further embodiment of this invention provides.
Referring to Fig. 2, on the basis of the above embodiments, asset identification method provided in this embodiment.First space
With predetermined logistic regression curve, the logistic regression curve is corresponding with preset system type, and is according to
What the first sample transmission data of system determined.
Wherein, for each application system in business support system, each system is returned in the corresponding logic in the first space
Return curve, the logistic regression curve can be used for determining whether assets to be identified belong to the corresponding system of the logistic regression curve
System.
The method step 13 determines that the mode of the system type of the corresponding assets to be identified of the first transmission data can
There are many, the present embodiment illustrates one of.
Step 13 ', according to it is described first transmission data and the logistic regression curve relative position, determine described first
Whether the corresponding assets to be identified of transmission data belong to the system.
The first transmission data are being determined behind the position in the first space, by returning in the first space with the logic
The relative position for returning curve can classify the first transmission data according to the sort feature of the logistic regression curve,
Determine whether assets to be identified belong to the system.
In step 13 ' before, the asset history data based on known system, by the machine learning algorithm of logistic regression into
Row training obtains the system ownership model of known assets.
Specifically, logistic regression (Logistic Regression, LR) is also known as logistic regression analysis, is classification and pre-
One of method of determining and calculating predict to future outcomes by the performance of historical data.
Specifically, the first sample for obtaining the known assets in preset time period transmits data, the first sample transmission
Data include the amount of assets and data packet number with the known assets transmission；Using logistic regression algorithm to first sample
This transmission data are trained, and determine the logistic regression curve corresponding with preset system type.
Logistic regression algorithm predicts the value of a discrete type dependent variable using known independent variable.Its algorithm is as follows: often
The target of the regression algorithm of rule is to fit a polynomial function f (x), so that predicted value and the error of true value are minimum.Tool
Body formula is as follows:
F (x)=c0+c1xi+…+cn-1xn
In formula, n is Characteristic Number, and c is the fitting coefficient of each feature.
Assuming that data set has n independent features, x1To xnFor n feature of sample, wherein [x1…xn] be input to
Amount, so the process of training is exactly to be determined at [c0,c2…cn-1] value so that output of the expression formula for multiple input vectors
It is worth accuracy highest.
In order to make f (x) that can there is good logic judgment property, it is desirable to directly sample of the expression with feature x
It is assigned to the probability of certain class, for example can indicate that x is divided into positive class when f (x)>0.5, the expression of f (x)<0.5 is divided into anti-class,
And f (x) is always between [0,1].Introduce sigmoid function.This function is defined as follows:
Fig. 3 shows the image of the sigmoid function of further embodiment of this invention offer.
Referring to Fig. 3, sigmoid function have current embodiment require that characteristic, domain exists in all real numbers, codomain
It between [0,1], and is 0.5 in 0 point value.
F (x) is changed into the means of sigmoid function are as follows:
Enabling p (x)=1 is the probability that the sample with feature x is assigned to classification 1, then p (x)/[1-p (x)] is defined as allowing
Step is than (odds ratio).
Introduce logarithm:
P (x) solution is come out by above formula and obtains following formula:
After the sigmoid function needed, next only needs as usual linear regression, fit the formula
Middle n parameter c.This transformation is referred to as logit transformation, also referred to as logical transition.
Data, which are transmitted, according to the first sample of the system determines the logistic regression curve, specifically:
By logistic regression algorithm, to known system assets, i.e. first sample transmission data carry out machine training, realize such as
Under:
Independent variable is two accountings, and the molecule of first accounting is associated section in known system assets and appointing system
Points, denominator are appointing system total node number；The molecule of first accounting is other passes in known system assets and appointing system
The flow packet number of interlink point, denominator are appointing system total flow packet number.
Dependent variable is whether known system assets belong to certain known system.
It based on the above variable, is trained by logistic regression algorithm, can obtain appointing system according to p (x) training, i.e., in advance
If the logistic regression curve of system.
The logic that the machine training that Fig. 4 shows a kind of asset identification method of further embodiment of this invention offer obtains is returned
Return curve synoptic diagram.
Referring to Fig. 4, in the first space, ordinate is first accounting, and abscissa is second accounting, figure
Middle curve is logistic regression curve.
Wherein, the rounded node in curve lower left is the IP for being not belonging to the system, the node that curve upper right side is square
Classify to can realize according to the sort feature of the logistic regression curve to node for the IP for belonging to the system, determines
Whether node belongs to the system.
For the unknown asset node compared after flow packet capturing, the association quantity of unknown node and other nodes is counted
With flow packet quantity, it can judge automatically whether belong to the system with system ownership model.
The system ownership that Fig. 5 shows a kind of asset identification method of further embodiment of this invention offer judges automatically signal
Figure.
Referring to Fig. 5, if judgement knows the first transmission data in the circle of the lower left of the position in curve in the first space
Circle label, expression are not belonging to the corresponding system of the logistic regression curve.If judgement knows that the first transmission data exist
The position in the first space indicates to belong to the corresponding system of the logistic regression curve in the circles mark in the upper right side of curve
System.
Asset identification method provided in this embodiment, at least has following technical effect that
By the way that logistic regression curve is arranged in the first space, according to the opposite position of the first transmission data and logistic regression curve
It sets, determines whether the corresponding assets to be identified of the first transmission data belong to the system, thus can reach the knowledge of automatic intelligent
System type belonging to other assets, to fast and accurately carry out asset identification.
Fig. 6 shows a kind of flow diagram of asset identification method of further embodiment of this invention offer.
Referring to Fig. 6, on the basis of the above embodiments, after the step 13, can further implement using the present invention
The method that example provides identifies the Asset Type in system.
Optionally, there are many Asset Types in a system, for example, APP server, WEB server, cache server,
Interface server, file server, database server etc..
Method provided in an embodiment of the present invention specifically includes the following steps:
Step 21, obtain preset time period in assets to be identified second transmission data, second data include institute
State the application of number-of-packet and the assets in the system that the assets in assets to be identified and the system carry out data transmission
User type.
Optionally, for assets to be identified, based on the mark of assets to be identified, undertaking to perform work within a time limit and according to specifications to have by network obtains preset time
Second transmission data of the assets to be identified in section.
Wherein, the assets to be identified can be server to be identified, the number-of-packet be server to be identified with
The quantity for the data packet transmitted between server in the system, the service referred in the system using user type
The user type of application software on device.
For example, the server in the system is APP server, can also be there are many class in addition to the user of APP type
The user of type is on APP server.
Network job contract tool can obtain the data packet of the assets to be identified and other assets transmission, analyze data packet
The number-of-packet of assets transmission in the assets to be identified and the system can be obtained, apply user in assets in the system
Type.
The second transmission data are mapped to second space by step 22, determine the second transmission data in the second sky
Between position.
In the present embodiment, the mode that a variety of data visualizations can be used obtains the second transmission data in second space
Position.Optionally, the second space can be two-dimensional coordinate system, and the second transmission data are mapped to a two-dimensional coordinate
The mode of system can there are many.
Specifically, the abscissa of the second transmission data, and root can be determined according to the application user type of the assets
According to the number-of-packet that assets in system are transmitted, total coordinate of the transmission data is determined, obtain determining the second transmission data
In the position of second space.
Step 23, according to the second transmission data in the position of second space and predetermined Asset Type and the
Two sample delivery data determine the corresponding assets to be identified of the second transmission data in the corresponding relationship of the position of second space
Asset Type.
Before this step, predetermined Asset Type and the second sample delivery data are obtained in the position of second space
Corresponding relationship, the corresponding relationship is obtained by machine learning algorithm.
Optionally, the system based on assets to be identified is grabbed by tcpdump order and passes through the service in designated time period
The flow packet of device network interface card, parsing outflow packet neutralizes the IP, and there are other IP in the system of incidence relation, and count and flow between IP
Measure the quantity of packet, the application software user in the assets.
The IP and asset database that parse are compared, the asset database, which refers to, is stored with all O&M departments
The database of responsible device hardware and software configuration information, each server assets have at least one IP record, according to
Look-up table 1 obtains the relationship of unknown assets and known system.
For example, if determining that system belonging to obtained assets to be identified is the known system in table 1 by step 13
1, the relevant information of assets in known system 1 can be obtained.
The asset history data of known system are instructed as the second sample delivery data by machine learning algorithm
Practice, obtains the assets ownership model of known assets, i.e. each Asset Type and the second sample delivery data is in second space
The corresponding relationship of position, by the second transmission data in the position of second space, with the second sample delivery data in the second sky
Between position matched.
For example, the second sample delivery data are that the assets of APP server are corresponding with Asset Type in the position of second space,
If the second transmission data can determine described the in the location matches of the position of second space and the assets of APP server
The two corresponding assets to be identified of transmission data are APP server, if mismatch, then with correspond to WEB server in second space
The position of second sample delivery data is matched, and so on, thus, it is possible to new unknown assets are judged automatically by the model
Which kind of server assets belonged to.
Asset identification method provided in this embodiment, at least has following technical effect that
By obtaining the second transmission data in the position of second space according to the second of assets to be identified the transmission data
It sets, and is determined according to predetermined Asset Type and the second sample delivery data in the corresponding relationship of the position of second space
Described second transmits the Asset Type of the corresponding assets to be identified of data, thus can reach belonging to the identification assets of automatic intelligent
Asset Type, to improve working efficiency.
Fig. 7 shows a kind of flow diagram of asset identification method of further embodiment of this invention offer.
Referring to Fig. 7, on the basis of the above embodiments, asset identification method provided in this embodiment.The second space
With at least one predetermined mass center, the mass center is corresponding with preset Asset Type, and be according to the assets
What two sample delivery data determined.
The method step 23 determines that the mode of the Asset Type of the corresponding assets to be identified of the second transmission data can
There are many, the present embodiment illustrates one of.
Step 23 ', by with the second transmission data apart from the corresponding Asset Type of nearest mass center, as described wait know
The Asset Type of other assets.
It is determining that described second transmits data behind the position of second space, is passing through the second transmission described in second space
Data distance, according to the sort feature of the mass center, can classify the second transmission data at a distance from the mass center,
Determine whether assets to be identified belong to the Asset Type.
In step 23 ' before, after the system ownership for determining unknown assets, asset history data and K- based on known system
Means algorithm carries out machine training and obtains model, judges which kind of Asset Type is unknown assets belong to.
Specifically, k-means algorithm is a kind of very common clustering algorithm, its basic thought is: finding k by iteration
A kind of splitting scheme of a cluster, so that the mean value clustered with this k is come global error resulting when representing corresponding Different categories of samples
It is minimum.
Specifically, the second sample delivery data of the known assets in preset time period are obtained, second data include
With the number-of-packet of assets known in system transmission and the application user type of the known assets；It is poly- using k-means
Class algorithm is trained the second sample delivery data, determines the mass center corresponding with preset Asset Type.
The basis of k-means algorithm is minimal error sum-of-squares criterion.Its cost function is:
In formula, μ c (i) indicates the mean value of the characteristic value of the mean value, that is, algorithm of ith cluster itself.
In this step, cost function is the smaller the better, and intuitively, all kinds of interior samples are more similar, with such
Square-error between value is smaller, sums to the obtained square-error of all classes, that is, can verify that whether is each cluster when being divided into k class
It is optimal.
The cost function of above formula can not be minimized with the method for parsing, can only there is the method for iteration.K-means algorithm be by
Sample clustering is at k cluster (cluster), and wherein k is given, solution procedure are as follows: constantly restrains each sample by calculating
A classification c (i) is all belonged to, algorithm description is as follows:
1, k cluster center of mass point is randomly selected；
2, following procedure is repeated until convergence.
For each sample i, its class that should belong to is calculated:
In formula, c (i) indicates the classification of i, μjIndicate the mass center of j.
For each class j, such mass center is recalculated:
In formula, μjIndicate the mass center of j.
The k-means algorithm that Fig. 8-11 respectively illustrates a kind of asset identification method of further embodiment of this invention offer shows
It is intended to.
Referring to Fig. 8-11, show respectively in the identical situation of sample, 4 kinds of mass centers being calculated.
The mass center is determined according to the second sample delivery data of the system, specifically: it is right by k-means algorithm
Known system assets, i.e. the second sample delivery data carry out machine training, are accomplished by
Figure 12 shows the service that a kind of machine training of asset identification method of further embodiment of this invention offer obtains
Device assets class belongs to model schematic.
Referring to Fig.1 2,6 cluster center of mass point are taken, 6 class server assets are respectively represented, count the application in well known server
User type, and the flow packet quantity between other nodes confirm that the class of known assets belongs to by calculating, known to drafting
Server distribution figure, then its mass center is recalculated to every a kind of assets, it repeats the above process, until all well known server assets
All it is referred in figure.
Figure 13 shows a kind of server assets class ownership mould of asset identification method of further embodiment of this invention offer
Type judges schematic diagram.
Referring to Fig.1 3, for newly-increased unknown server assets, count thereon using user type and other nodes
Flow packet quantity, is shown on distribution map in the same way, by the judgement of the distance between different center of mass point, comes automatic
Confirm which kind of server assets is unknown assets belong to.
For example, if judgement knows the second transmission data in the position of second space and the mass center of APP server
Recently, it indicates to belong to Asset Type of the same race with APP server.
Asset identification method provided in this embodiment, at least has following technical effect that
By the way that mass center is arranged in second space, according to the second transmission data at a distance from mass center, the second transmission data are determined
Which kind of Asset Type is corresponding assets to be identified belong to, and thus can reach assets class belonging to the identification assets of automatic intelligent
Type, to fast and accurately carry out asset identification.
Figure 14 shows a kind of flow diagram of asset identification method of further embodiment of this invention offer.
Referring to Fig.1 4, on the basis of the above embodiments, after the step 23, can further it implement using the present invention
The method that example provides identifies the group type in system.
Optionally, in the same system, the assets of Asset Type of the same race may make up multiple clusters, and each cluster includes multiple
The identical assets of Asset Type.
Method provided in an embodiment of the present invention specifically includes the following steps:
Step 31, the third for obtaining the assets to be identified in preset time period transmit data, and the third transmits data packet
The number-of-packet and the assets in the system for including transmission between the assets to be identified and the assets of Asset Type of the same race
Mark.
Optionally, for assets to be identified, based on the mark of assets to be identified, undertaking to perform work within a time limit and according to specifications to have by network obtains preset time
The third of assets to be identified in section transmits data.
Wherein, the assets to be identified can be server to be identified, and the data packet number is server to be identified
With the quantity for belonging to the data packet transmitted between same type of server in the system with server to be identified, the system
The mark of assets in system can be the number for belonging to same type of server with server to be identified.
For example, server to be identified is APP server, known APP server total amount is 7,7 in the system
APP server has corresponding server identification such as number 1-7, obtains APP server distribution to be identified and 7 APP servers
The quantity of the data packet carried out data transmission.
Network job contract tool can obtain the data packet of the assets to be identified and other assets transmission, analyze data packet
The asset identification of the assets to be identified and assets transmission, and the number-of-packet of transmission can be obtained.
Third transmission data are mapped to third space by step 32, determine the third transmission data in third sky
Between position.
The mode that a variety of data visualizations can be used obtains the third transmission data in the position in third space.It is optional
Ground, the third space can be two-dimensional coordinate system, and the mode that third transmission data are mapped to a two-dimensional coordinate system can
There are many.Specifically, the abscissa of the third transmission data can be determined according to the asset identification, and transmitted according to assets
Number-of-packet, determine total coordinate of the transmission data, obtain determining the third transmission data in the position in third space.
Step 33 transmits data according to the third in the position in third space and predetermined group type and the
Three sample delivery data determine the corresponding assets to be identified of the third transmission data in the corresponding relationship of the position in third space
Group type.
Before this step, predetermined group type and third sample delivery data are obtained in the position in third space
Corresponding relationship, the corresponding relationship is obtained by machine learning algorithm.
Optionally, the system based on assets to be identified is grabbed by tcpdump order and passes through the service in designated time period
The flow packet of device network interface card, parsing outflow packet neutralizes the IP, and there are other IP in the system of incidence relation, and count and flow between IP
Measure the quantity of packet.
The IP and asset database that parse are compared again, the asset database, which refers to, is stored with all O&M portions
The database of door responsible device hardware and software configuration information, each server assets have at least one IP record, root
The group type of the assets of known system is obtained according to look-up table.
The asset history data of known system are instructed as third sample delivery data by machine learning algorithm
Practice, obtains the cluster ownership model of known assets, i.e. cluster belonging to each assets and third sample delivery data is in third
The corresponding relationship of the position in space exists by third transmission data in the position in third space with third sample delivery data
The position in third space is matched.
For example, third sample delivery data are corresponding with the assets of the first group type in the position in third space, if described
Third transmits data in the location matches of the position in third space and the assets of the first cluster, then can determine the third transmission number
Belong to the first cluster according to corresponding assets to be identified, if mismatch, then with the assets of the second group type in third space
Position is matched, and so on, thus, it is possible to judge automatically which type of cluster is new unknown assets belong to by the model.
Asset identification method provided in this embodiment, at least has following technical effect that
By transmitting data according to the third of assets to be identified, the third transmission data are obtained in the position in third space
It sets, and is determined according to predetermined group type and third sample delivery data in the corresponding relationship of the position in third space
The third transmits the group type of the corresponding assets to be identified of data, thus can reach belonging to the identification assets of automatic intelligent
Group type, to improve working efficiency.
Figure 15 shows a kind of flow diagram of asset identification method of further embodiment of this invention offer.
Referring to Fig.1 5, on the basis of the above embodiments, asset identification method provided in this embodiment.The third space
With predetermined polynomial fitting curve, the polynomial fitting curve is corresponding with preset group type, and is according to
What the third sample delivery data of cluster determined.It wherein, include a variety of assets classes for a certain system in business support system
Type, the same type of multiple clusters of the composition of assets correspond to a polynomial fitting curve in third space for each cluster, described
Polynomial fitting curve can be used for determining whether assets to be identified belong to the corresponding cluster of the polynomial fitting curve.
The method step 32, determine position of the third transmission data in third space mode can there are many, this
Embodiment illustrates one of.
The described method includes:
Step 32 ', obtain third transmission data in the matched curve in third space.
For server assets, itself and associated server IP and flow packet quantity, digital simulation curve f are countedw(x)。
Specifically, according to the amount of assets with the assets transmission to be identified, the horizontal seat of the third transmission data is determined
Mark, and according to the data packet number, it determines the ordinate of third transmission data, can be obtained according to fitting algorithm in the prior art
Matched curve to the determination third transmission data in third space.
For example, using polynomial fitting algorithm:
The set of a certain type function y=f (x) and m data point (x, y) is given, minimization absolute deviation | yi-f
(xi) | quadratic sum, that is, determine function y=f (x) in parameter, carry out minimization:
In formula, for the parameter in function as independent variable, the quadratic sum of absolute deviation utilizes the function of many variables as objective function
Extreme value theory can solve.
Such as: set the form of expected model are as follows:
Y=f (x)=a0+a1x+…+anxn
In formula, n is fixed.
Minimization (by taking n=2 as an example) is required by least-squares estimation:
S seeks local derviation to parameter, enables it be equal to zero and obtains equation:
The matrix of normal equation group solution indicates:
Then normal equation group are as follows:
(AAT) a=ATy
If AAT is reversible, the solution of normal equation are as follows:
A=(AAT)-1ATy
Obtain y value, matched curve of the as described third transmission data in third space.
The method step 33 determines that the mode of the group type of the corresponding assets to be identified of the third transmission data can
There are many, the present embodiment illustrates one of.
The described method includes:
Step 331 determines that the third transmits the fitting coefficient of matched curve Yu the predetermined matched curve of data.
Step 332, by fitting coefficient maximum when the corresponding group type of matched curve, the cluster as assets to be identified
Type.
Calculate the fitting coefficient of matched curve and the predetermined matched curve of third transmission data in third space.Its
In, fitting coefficient is bigger, illustrates and which curve is more close to which cluster belonged to.
The fitting of matched curve Yu the predetermined matched curve of data is transmitted by the third described in third space
The third can be transmitted data and classified, determine which cluster is assets to be identified belong to by the calculating of coefficient.
Before step 331, after the Asset Type ownership for determining unknown assets, for Asset Type of the same race, there are multiple collection
Group, also needs which cluster determination belongs to, can be based on the asset history data of known system and the machine learning algorithm of polynomial fitting
It carries out machine training and obtains model, judge which cluster is unknown assets belong to.
Specifically, the third sample delivery data of the known assets in preset time period, the third sample delivery are obtained
Data include the amount of assets of known assets and the outer assets transmission of cluster, and the number-of-packet of transmission；It is worthwhile using multivariate quasi
Method is trained the third sample delivery data, determines the polynomial fitting curve corresponding with preset group type.
By the machine learning algorithm of polynomial fitting, trained in real time to the flow packet capturing information progress of known system cluster
Model out, realization judge automatically the cluster of newly-increased unknown node.
The outer associated server IP of statistical cluster interacts flow packet quantity with cluster server respectively, then takes mean value, then
The matched curve f (x) of known cluster is calculated by polynomial fitting algorithm.
Figure 16 shows a kind of polynomial fitting of a cluster of asset identification method of further embodiment of this invention offer
Curve synoptic diagram.
Referring to Fig.1 6, for multiple clusters in system, each cluster is analyzed and processed.Obtain each cluster and collection
The quantity of the outer assets transmitted of group, as the abscissa in third space, for the flow packet total amount of each cluster transmission, as vertical
Each known cluster can be obtained in the polynomial fitting curve in third space in coordinate.
Figure 17 shows multiple clusters in a kind of system of asset identification method of further embodiment of this invention offer are polynary
Matched curve schematic diagram.
Referring to Fig.1 7, with same method, it can obtain the polynomial fitting curve f of all clusters in the system1(x)、f2
(x)……fn(x)。
The cluster ownership model judgement that Figure 18 shows a kind of asset identification method of further embodiment of this invention offer is shown
It is intended to.
Referring to Fig.1 8, for newly-increased unknown server assets, the amount of assets of its transmission is counted, with flow packet quantity,
It is shown in third in the same way spatially.
Specifically, itself and associated server IP and flow packet quantity are counted, regression curve f is calculatedw(x), and with it is known
The regression curve f of cluster1(x) it compares.
By following formula digital simulation coefficients, fitting coefficient is bigger, illustrates and which which curve is more close to belonging to
Cluster.
In formula, R is fitting coefficient, and n is cluster number, and fw (xi) is that every curve xi corresponds to y value.
Asset identification method provided in this embodiment, at least has following technical effect that
By the matched curve of each cluster in third space setting system, data and matched curve are transmitted according to third
Fitting coefficient determines the cluster of the corresponding assets ownership to be identified of third transmission data, thus can reach the knowledge of automatic intelligent
Group type belonging to other assets, to fast and accurately carry out asset identification.
Figure 19 shows a kind of asset identification systems schematic diagram of further embodiment of this invention offer.
Refering to fig. 19, a kind of asset identification systems that another embodiment provides are invented, the system comprises flow packet capturing moulds
Block, system ownership judgment module, assets ownership judgment module, cluster belong to judgment module.
Wherein, flow packet capturing module is used to be based on known server ip, grabs designated time period by tcpdump order
The interior flow packet by the server network interface card, parsing outflow packet neutralizes other IP of the IP there are incidence relation, and counts IP
Between flow packet quantity.The IP and asset database that parse are compared again, find out unknown assets, and with statistical
Method summarizes the relational matrix between unknown assets and known assets, such as table 1.
System belongs to judgment module and is used for the asset history data based on known system, passes through the machine learning of logistic regression
Algorithm is trained, and obtains the system ownership model of known assets, and can be judged automatically new unknown assets by the model and be
It is no to belong to the system.
Specifically, system ownership judgment module can run the step of realizing the method such as Fig. 1 and Fig. 2.
By logistic regression algorithm, machine training is carried out to known system assets, the logistic regression that can be trained such as Fig. 4 is bent
Line.
For the unknown asset node compared after flow packet capturing, the association quantity of unknown node and other nodes is counted
With flow packet quantity, the system ownership model more than can judge automatically whether belong to the system, and system as shown in Figure 5 is returned
Category judges automatically.
After Asset Type judgment module is used to determine the system ownership of unknown assets, the module is according to known each in system
Class assets information and K-Means algorithm carry out machine training and obtain model, judge which kind of Asset Type is unknown assets belong to.
Specifically, Asset Type judgment module can run the step of realizing the method such as Fig. 6 and Fig. 7.
By k-means algorithm, machine training is carried out to known system assets, is implemented as follows:
6 cluster center of mass point are taken, 6 class server assets are respectively represented, count the application user type in well known server,
And the flow packet quantity between other nodes, confirm that the class of known assets belongs to by calculating, draws well known server point
Butut, then its mass center is recalculated to every a kind of assets, it repeats the above process, until all well known server assets are all referred to
In figure, the server assets class of Figure 12 such as can be trained and belong to model.
For newly-increased unknown server assets, the flow packet number using user type and other nodes thereon is counted
Amount, is shown on distribution map in the same way, unknown to automatically confirm that by the judgement of the distance between different center of mass point
Assets belong to which kind of server assets, server assets class ownership model judgement as shown in figure 13.
Cluster belongs to judgment module for the machine learning algorithm by polynomial fitting, grabs to the flow of known system cluster
Package informatin carries out training in real time and obtains model, and realization judges automatically the cluster of newly-increased unknown node.
Specifically, cluster ownership judgment module can run the step of realizing the method such as Figure 14 and Figure 15.
By polynomial fitting algorithm, the outer associated server IP of statistical cluster interacts flow packet number with cluster server respectively
Then amount takes mean value, then the matched curve f (x) of known cluster is calculated by polynomial fitting algorithm, can train such as Figure 16
A cluster polynomial fitting curve.
With same method, the matched curve of all clusters in the system can be obtained, as shown in figure 17.
For newly-increased unknown server assets, itself and associated server IP and flow packet quantity, digital simulation are counted
Curve fw (x), and comparing with the matched curve f1 (x) of known cluster, digital simulation coefficient, fitting coefficient is bigger, illustrate and
Which curve is more close to belonging to which cluster, cluster ownership model judgement as shown in figure 18.
Asset identification systems provided in an embodiment of the present invention automatically grab assets information, and are compared with asset database
It is right, unknown assets information is verified out, then by the algorithm of machine learning, by unknown assets by system ownership, assets ownership, cluster
Ownership etc. carries out Classification and Identification, updates into system topological figure.It is particularly used in the method for realizing above method embodiment, this reality
Example is applied to repeat no more.
Asset identification systems provided in this embodiment, at least have following technical effect that
Determine that unknown assets belong to which system, which class assets in system, in assets automatically by machine learning algorithm
Which cluster, reduce the process of manual confirmation, improve the accuracy of more new data.
Figure 20 shows a kind of structural schematic diagram of computer equipment of further embodiment of this invention offer.
Refering to Figure 20, computer equipment provided in an embodiment of the present invention, the computer equipment includes memory
(memory) 201 it, processor (processor) 202, bus 203 and is stored on memory 201 and can transport on a processor
Capable computer program.Wherein, the processor 201, memory 202 complete mutual communication by the bus 203.
Optionally, the computer equipment may also include communication interface (Communications Interface) 204, institute
Communication interface 204 is stated for the information transmission between the equipment and other communication equipments.
The processor 201 is used to call the program instruction in the memory 202, realizes such as when executing described program
The method of Fig. 1-2, and also realize following method:
First space has predetermined logistic regression curve, logistic regression curve and preset system type pair
It answers, and is to transmit what data determined according to the first sample of the system, specifically: obtain the known assets in preset time period
First sample transmit data, first sample transmission data include the amount of assets and data with the known assets transmission
Packet quantity；First sample transmission data are trained using logistic regression algorithm, determining and preset system type pair
The logistic regression curve answered.
In another embodiment, the method such as Fig. 6-7 is realized when the processor executes described program, and is also realized
Following method:
The second space has at least one predetermined mass center, and mass center is corresponding with preset Asset Type, and is
It is determined according to the second sample delivery data of assets, specifically: obtain the second sample of the known assets in preset time period
Data are transmitted, second data include answering with the number-of-packet of assets known in system transmission and the known assets
Use user type；The second sample delivery data are trained using k-means clustering algorithm, determining and preset assets
The corresponding mass center of type.
In another embodiment, the method such as Figure 14-15 is realized when the processor executes described program, and also real
Existing following method:
The third space has predetermined polynomial fitting curve, the polynomial fitting curve and preset collection realm
Type is corresponding, and is determined according to the third sample delivery data of the cluster, specifically: it obtains known in preset time period
The third sample delivery data of assets, the third sample delivery data include the assets of known assets and the outer assets transmission of cluster
Quantity, and the number-of-packet of transmission；The third sample delivery data are trained using polynomial fitting algorithm, determine with
The corresponding polynomial fitting curve of preset group type.
Computer equipment provided in this embodiment can be used for executing the corresponding program of method of above method embodiment, this
Implementation repeats no more.
Computer equipment provided in this embodiment, at least has following technical effect that
It is realized when executing described program by the processor according to the first of assets to be identified the transmission data, obtains first
Data are transmitted in the position in the first space, and data are transmitted in pair of the position in the first space according to system type and first sample
It should be related to, determine the system type of assets to be identified, thus can reach system type belonging to the identification assets of automatic intelligent,
To improve working efficiency.
Finally, it should be noted that above embodiments are only to illustrate the technical solution of the embodiment of the present invention, rather than it is limited
System；Although the embodiment of the present invention is described in detail with reference to the foregoing embodiments, those skilled in the art should
Understand: it is still possible to modify the technical solutions described in the foregoing embodiments, or to part of technical characteristic
It is equivalently replaced；And these are modified or replaceed, each reality of the embodiment of the present invention that it does not separate the essence of the corresponding technical solution
Apply the spirit and scope of a technical solution.
Claims (10)
1. a kind of asset identification method characterized by comprising
The first transmission data of the assets to be identified in preset time period are obtained, the first transmission data include with described wait know
The amount of assets and data packet number that other assets carry out data transmission；
The first transmission data are mapped into the first space, determine the first transmission data in the position in the first space；
It is transmitted according to the first transmission data in the position in the first space and predetermined system type and first sample
Data determine the system class of the corresponding assets to be identified of the first transmission data in the corresponding relationship of the position in the first space
Type.
2. according to the method described in claim 1, it is characterized by: first space has predetermined logistic regression bent
Line, the logistic regression curve is corresponding with preset system type, and is true according to the first sample of system transmission data
Fixed；
Correspondingly, it is described according to the first transmission data in the position in the first space and predetermined system type and the
One sample delivery data determine the corresponding assets to be identified of the first transmission data in the corresponding relationship of the position in the first space
System type, specifically:
According to the first transmission data in the relative position in the first space and the logistic regression curve, determine that described first passes
Whether the corresponding assets to be identified of transmission of data belong to the system.
3. according to the method described in claim 2, it is characterized by: first space has predetermined logistic regression bent
Line, the logistic regression curve is corresponding with preset system type, and is true according to the first sample of system transmission data
Fixed, specifically:
The first sample for obtaining the known assets in preset time period transmits data, and the first sample transmission data include and institute
State the amount of assets and data packet number of known assets transmission；
First sample transmission data are trained using logistic regression algorithm, determination is corresponding with preset system type
The logistic regression curve.
4. method according to claim 1 or 2 or 3, it is characterised in that: determination the first transmission data are corresponding
After the system type of assets to be identified, the method also includes:
The second transmission data of the assets to be identified in preset time period are obtained, second data include the assets to be identified
The application user type of the number-of-packet and the assets in the system transmitted between the assets of the system；
The second transmission data are mapped into second space, determine the second transmission data in the position of second space；
According to the second transmission data in the position of second space and predetermined Asset Type and the second sample delivery
Data determine the assets class of the corresponding assets to be identified of the second transmission data in the corresponding relationship of the position of second space
Type.
5. according to the method described in claim 4, it is characterized by: the second space has at least one predetermined matter
The heart, the mass center is corresponding with preset Asset Type, and is determined according to the second sample delivery data of the assets；
Correspondingly, according to the second transmission data in the position of second space and predetermined Asset Type and the second sample
This transmission data determine the money of the corresponding assets to be identified of the second transmission data in the corresponding relationship of the position of second space
Type is produced, specifically:
By with the second transmission data in second space apart from the corresponding Asset Type of nearest mass center, as described to be identified
The Asset Type of assets.
6. according to the method described in claim 5, it is characterized by: the second space has at least one predetermined matter
The heart, the mass center is corresponding with preset Asset Type, and is determined according to the second sample delivery data of the assets, specifically
Are as follows:
Obtain preset time period in known assets the second sample delivery data, second data include in the system
The number-of-packet of known assets transmission and the application user type of the known assets；
The second sample delivery data are trained using k-means clustering algorithm, determining and preset Asset Type pair
The mass center answered.
7. according to the method described in claim 4, it is characterized by: the determination the second sample delivery data it is corresponding to
After the Asset Type for identifying assets, the method also includes:
The third for obtaining the assets to be identified in preset time period transmits data, and the third transmission data include described to be identified
The mark of the number-of-packet and the assets in the system transmitted between assets and the assets of Asset Type of the same race；
Third transmission data are mapped into third space, determine the third transmission data in the position in third space；
Data are transmitted in the position in third space and predetermined group type and third sample delivery according to the third
Data determine the collection realm of the corresponding assets to be identified of the third transmission data in the corresponding relationship of the position in third space
Type.
8. according to the method described in claim 7, it is characterized by: the third space has predetermined polynomial fitting bent
Line, the polynomial fitting curve is corresponding with preset group type, and is true according to the third sample delivery data of the cluster
Fixed；
Correspondingly, described that third transmission data are mapped into third space, determine the third transmission data in third sky
Between position, specially obtain third transmission data in the polynomial fitting curve in third space；
It is described that data are transmitted in the position in third space and predetermined group type and third sample according to the third
Data are transmitted in the corresponding relationship of the position in third space, determine the cluster of the corresponding assets to be identified of the third transmission data
Type, specifically:
Determine the polynomial fitting curve of the third transmission data and the fitting coefficient of predetermined polynomial fitting curve；
The corresponding group type of polynomial fitting curve when by fitting coefficient maximum, the group type as assets to be identified.
9. according to the method described in claim 8, it is characterized by: the third space has predetermined polynomial fitting bent
Line, the polynomial fitting curve is corresponding with preset group type, and is true according to the third sample delivery data of the cluster
Fixed, specifically:
The third sample delivery data of the known assets in preset time period are obtained, the third sample delivery data include known
The amount of assets of assets and the outer assets transmission of cluster, and the number-of-packet of transmission；
The third sample delivery data are trained using polynomial fitting algorithm, determination is corresponding with preset group type
The polynomial fitting curve.
10. a kind of computer equipment, including memory, processor, bus and storage are on a memory and can be on a processor
The computer program of operation, which is characterized in that the processor realizes such as claim 1-9 any one when executing described program
Method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710428726.9A CN109040155B (en) | 2017-06-08 | 2017-06-08 | Asset identification method and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710428726.9A CN109040155B (en) | 2017-06-08 | 2017-06-08 | Asset identification method and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109040155A true CN109040155A (en) | 2018-12-18 |
| CN109040155B CN109040155B (en) | 2021-06-08 |
Family
ID=64629559
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710428726.9A Active CN109040155B (en) | 2017-06-08 | 2017-06-08 | Asset identification method and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109040155B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113158001A (en) * | 2021-03-25 | 2021-07-23 | 深圳市联软科技股份有限公司 | Method and system for judging attribution and correlation of network space IP assets |
| CN114124837A (en) * | 2021-10-22 | 2022-03-01 | 南京中新赛克科技有限责任公司 | Asset information discovery system and method based on passive flow |
| EP4222615A1 (en) * | 2020-10-02 | 2023-08-09 | Acentium Inc. | Systems and methods for asset fingerprinting cross reference to related applications |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102214206A (en) * | 2011-04-27 | 2011-10-12 | 百度在线网络技术（北京）有限公司 | Method and equipment for establishing association relation between information entities |
| CN102880543A (en) * | 2011-04-18 | 2013-01-16 | 埃森哲环球服务有限公司 | Test data supply chain manager for an integrated testing platform |
| CN103489092A (en) * | 2013-10-18 | 2014-01-01 | 国家电网公司 | Automated operating device for storage and verification of electric energy meter of electric power system and operating method thereof |
| CN104135068A (en) * | 2014-07-04 | 2014-11-05 | 国家电网公司 | Plug-and-play method of power distribution terminals based on IEC61850 standard |
| US20160092643A1 (en) * | 2014-09-26 | 2016-03-31 | Robert Daniel Hinkle | System and process for providing verification and validation of an individual's identity |
| CN106203343A (en) * | 2015-04-29 | 2016-12-07 | 通用电气公司 | image analysis system and method for automatic asset identification |
| CN106416318A (en) * | 2014-01-23 | 2017-02-15 | 谷歌公司 | Determining data associated with proximate computing devices |
-
2017
- 2017-06-08 CN CN201710428726.9A patent/CN109040155B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102880543A (en) * | 2011-04-18 | 2013-01-16 | 埃森哲环球服务有限公司 | Test data supply chain manager for an integrated testing platform |
| CN102214206A (en) * | 2011-04-27 | 2011-10-12 | 百度在线网络技术（北京）有限公司 | Method and equipment for establishing association relation between information entities |
| CN103489092A (en) * | 2013-10-18 | 2014-01-01 | 国家电网公司 | Automated operating device for storage and verification of electric energy meter of electric power system and operating method thereof |
| CN106416318A (en) * | 2014-01-23 | 2017-02-15 | 谷歌公司 | Determining data associated with proximate computing devices |
| CN104135068A (en) * | 2014-07-04 | 2014-11-05 | 国家电网公司 | Plug-and-play method of power distribution terminals based on IEC61850 standard |
| US20160092643A1 (en) * | 2014-09-26 | 2016-03-31 | Robert Daniel Hinkle | System and process for providing verification and validation of an individual's identity |
| CN106203343A (en) * | 2015-04-29 | 2016-12-07 | 通用电气公司 | image analysis system and method for automatic asset identification |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP4222615A1 (en) * | 2020-10-02 | 2023-08-09 | Acentium Inc. | Systems and methods for asset fingerprinting cross reference to related applications |
| CN113158001A (en) * | 2021-03-25 | 2021-07-23 | 深圳市联软科技股份有限公司 | Method and system for judging attribution and correlation of network space IP assets |
| CN113158001B (en) * | 2021-03-25 | 2024-05-14 | 深圳市联软科技股份有限公司 | Network space IP asset attribution and correlation discrimination method and system |
| CN114124837A (en) * | 2021-10-22 | 2022-03-01 | 南京中新赛克科技有限责任公司 | Asset information discovery system and method based on passive flow |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109040155B (en) | 2021-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111024708B (en) | Method, device, system and equipment for processing product defect detection data | |
| CN111667267B (en) | Block chain transaction risk identification method and device | |
| CN105874743A (en) | Service provider network migration | |
| CN108763389A (en) | The integration method and device of data, storage medium, terminal | |
| CN115345212A (en) | Control method and device based on machine learning model | |
| CN110213070B (en) | Group network identification method, device, server and computer readable storage medium | |
| CN109040155A (en) | Asset identification method and computer equipment | |
| CN110781311B (en) | Enterprise-consistent-person operation system and method | |
| CN112860676B (en) | Data cleaning method applied to big data mining and business analysis and cloud server | |
| CN118138471A (en) | Knowledge-graph-based network model construction method, device and storage medium | |
| CN114142923A (en) | Optical cable fault positioning method, device, equipment and readable medium | |
| CN117172641A (en) | Production logistics management platform based on block chain and digital twin and implementation method | |
| Zhang et al. | A pattern-recognition-based algorithm and case study for clustering and selecting business services | |
| CN104679870B (en) | A kind of method of data acquisition for information system | |
| CN104965846B (en) | Visual human's method for building up in MapReduce platform | |
| Yu et al. | A genetic programming approach to distributed QoS-aware web service composition | |
| Bebensee et al. | Applying dynamic Bayesian networks for automated modeling in ArchiMate: a realization study | |
| CN113169888B (en) | Method for distributing field technicians and technician distribution system | |
| CN108259441A (en) | It is a kind of to prevent URL from accessing the method and device to detour | |
| CN106407212A (en) | Network account category determination method and apparatus, and object clustering method and apparatus | |
| CN116305720A (en) | Multidimensional management data mining method based on constraint | |
| Li | Supply Chain Efficiency and Effectiveness Management Using Decision Support Systems | |
| US11100454B1 (en) | CDD with heuristics for automated variable use-case based constrained logistics route optimization | |
| CN115705511A (en) | Method, device and equipment for determining pickup area and storage medium | |
| CN115660245A (en) | Service arrangement method and device, electronic equipment and storage medium |
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 |