CN106022134B - A method of setting intrusion detection DCA algorithm weight - Google Patents
A method of setting intrusion detection DCA algorithm weight Download PDFInfo
- Publication number
- CN106022134B CN106022134B CN201610380426.3A CN201610380426A CN106022134B CN 106022134 B CN106022134 B CN 106022134B CN 201610380426 A CN201610380426 A CN 201610380426A CN 106022134 B CN106022134 B CN 106022134B
- Authority
- CN
- China
- Prior art keywords
- signal
- weight
- mature
- semi
- mat
- 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.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/50—Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems
- G06F21/55—Detecting local intrusion or implementing counter-measures
- G06F21/56—Computer malware detection or handling, e.g. anti-virus arrangements
- G06F21/562—Static detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming
Abstract
The purpose of the present invention is to provide a kind of methods, and the weight of DCA algorithm (Dendritic Cell Algorithm, Dendritic Cells algorithm) can be adjusted according to actual network environment, to solve the problems, such as that DCA algorithm detection accuracy is not high.The present invention mainly optimizes setting to weight of the DCA algorithm when synthesizing half ripe signal and mature signal.Optimal weight is obtained using intelligent algorithm then according to known sample information by constructing an optimization problem related with weight variable change.The invention can carry out the setting of DCA algorithm weight according to actual network environment, to improve the intrusion detection performance of DCA algorithm, reduce false detection rate and omission factor.
Description
Technical field
The present invention relates to a kind of methods for setting intrusion detection DCA algorithm weight.
Background technique
Explanation of nouns:
DCA algorithm: (Dendritic Cell Algorithm), Dendritic Cells algorithm.
PAMP signal: (pathogen-associated molecular pattern) pathogen associated molecular pattern letter
Number.
MACV:(mature context antigen value) maturation environmental antigens value.
DCA algorithm has 3 kinds of signals, i.e. costimulatory signal, half ripe signal and mature signal.Every kind of signal all passes through
Safety signal (safe signal), danger signal (danger signal) and the PAMP that corresponding one group of weight acquires the external world
Signal is synthesized into.
For each immature DC cell, by its safety signal (safe signal) collected, danger signal
(danger signal) and PAMP signal synthesize corresponding costimulatory signal, half ripe signal and mature signal.It is synthesized
Formula is as follows:
ocsm=(1+IC) (wP,csm*P+wD,csm*D+wS,csm*S)
osemi=(1+IC) (wP,semi*P+wD,semi*D+wS,semi*S) (1)
omat=(1+IC) (wP,mat*P+wD,mat*D+wS,mat*S)
ocsm、osemi、omatRespectively indicate corresponding costimulatory signal, half ripe signal and mature signal.wP,csmIt indicates
The weight of PAMP signal when synthesizing costimulatory signal;wD,csmIndicate the weight of danger signal when synthesis costimulatory signal;
wS,csmIndicate the weight of safety signal when synthesis costimulatory signal;wP,semiIndicate PAMP signal when synthesis half ripe signal
Weight;wD,semiIndicate the weight of danger signal when synthesis half ripe signal;wS,semiIt indicates to believe safely when synthesis half ripe signal
Number weight;wP,matThe weight of PAMP signal when indicating to synthesize mature signal;wD,matDanger signal when indicating to synthesize mature signal
Weight;wS,matThe weight of safety signal when indicating to synthesize mature signal;P indicates PAMP signal;D indicates danger signal;S is indicated
Safety signal;
Each immature DC cell constantly accumulates costimulatory signal, half ripe signal and mature letter collected
Number, it is as follows:
Ccsm(t)=Ccsm(t-1)+Ocsm(t)
Csemi(t)=Csemi(t-1)+Osemi(t) (2)
Cmat(t)=Cmat(t-1)+Omat(t),
Ccsm(t)、Csemi(t)、Cmat(t) the costimulatory signal value of the accumulation after respectively indicating current acquisition signal, half
Mature signal value and mature signal value.
If the costimulatory signal value of accumulation is greater than or equal to presetting threshold value, stop acquiring signal, and judge
Size between the half ripe signal of accumulation and the mature signal of accumulation, to determine whether band DC cell is that mature or half ripe is thin
Born of the same parents.I.e. if Csemi(t)>Cmat(t), then the immature DC cell differentiation is a semi-matured DC cell, is otherwise divided into
One mature DC cell.
In DCA algorithm, each DC cell can acquire the signal of different type network behavior, and according to corresponding network
The accumulation of behavior progress signal.Therefore for every kind of network behavior, the safety of its behavior can be judged according to MACV, MACV's
Calculation formula is as follows:
If the value of such network behavior is greater than or equal to the threshold value of setting, then it is assumed that an abnormal network behavior, otherwise
For normal network behavior.
DCA algorithm is employed successfully in the fields such as intrusion detection at present, but synthesizes its mature signal and half ripe signal
Weight generally all directlys adopt fixed setting value, cannot be transferred to adjust weight in real time according to actual complex network environment, from
And cause detection accuracy not high.
Summary of the invention
The purpose of the present invention is to provide a kind of methods, and synthesis DCA algorithm can be adjusted according to actual network environment
The weight of middle maturation signal and half ripe signal, to solve the problems, such as that DCA algorithm detection accuracy is not high.
To achieve the goals above, The technical solution adopted by the invention is as follows:
A method of setting intrusion detection DCA algorithm weight utilizes the network row of immature DC cell collection sample
To detect network behavior according to the DC cell after differentiation, setting synthesis half ripe signal according to actual optimization of network environment
With the PAMP signal of mature signal, danger signal, safety signal weight, improve the detection accuracy of DCA algorithm.
For each immature DC cell, costimulatory signal, synthesis half ripe signal and mature signal are synthesized
Method are as follows:
ocsm=(1+IC) (wP,csm*P+wD,csm*D+wS,csm*S)
osemi=(1+IC) (wP,semi*P+wD,semi*D+wS,semi*S) (1)
omat=(1+IC) (wP,mat*P+wD,mat*D+wS,mat*S)
P indicates PAMP signal;D indicates danger signal;S indicates safety signal;ocsmIndicate costimulatory signal, osemiTable
Show half ripe signal, omatIndicate mature signal;wP,csmIndicate the weight of PAMP signal when synthesis costimulatory signal;wD,csmTable
The weight of danger signal when showing synthesis costimulatory signal;wS,csmIndicate the weight of safety signal when synthesis costimulatory signal;
wP,semiIndicate the weight of PAMP signal when synthesis half ripe signal;wD,semiIndicate the power of danger signal when synthesis half ripe signal
Value;wS,semiIndicate the weight of safety signal when synthesis half ripe signal;wP,matIndicate PAMP signal when synthesizing mature signal
Weight;wD,matThe weight of danger signal when indicating to synthesize mature signal;wS,matThe power of safety signal when indicating to synthesize mature signal
Value;IC indicates inflammatory signal.
Each immature DC cell constantly accumulates costimulatory signal, half ripe signal and mature signal collected
For
Ccsm(t)=Ccsm(t-1)+Ocsm(t)
Csemi(t)=Csemi(t-1)+Osemi(t) (2)
Cmat(t)=Cmat(t-1)+Omat(t),
Ccsm(t)、Csemi(t)、Cmat(t) costimulatory signal value, the half ripe signal of current period accumulation are respectively indicated
Value and mature signal value, t indicate current period number;If the costimulatory signal value of accumulation is greater than or equal to presetting threshold
Value then stops acquiring signal;If Csemi(t) > Cmat(t), then the immature DC cell differentiation is a semi-matured DC
Otherwise cell is divided into a mature DC cell, while the DC cell death;
For each network behavior, its MACV value is calculated:
Sample of the selection containing n heterogeneous networks behavior synthesizes half ripe signal and mature signal it is set separately
PAMP signal, danger signal, safety signal weight, establish following optimization problem:
N indicates that a total of different types of network behavior of n kind in selected sample, i are the network row of the i-th seed type
For MACViIndicate the MACV value of i-th of network behavior,I ∈ { 1 ..., n } is the judgement knot of the network behavior of the i-th seed type
Whether fruit is correctly evaluated, and correct judgment is then 1, is otherwise -1;It indicates to obtain and n the correct of network behavior is sentenced
Disconnected maximum times;Synthesize the PAMP signal of half ripe signal and mature signal, danger signal, safety signal respectively optimal power
Value is to makeObtain the value when maximum times correctly judged to n network behavior.
It is further to improve, a type of network behavior is only acquired for each immature DC cell, if should
DC cell death generates a new DC cell then to acquire such network behavior.
Further to improve, the weight for synthesizing costimulatory signal remains unchanged.
It is further to improve, believed using the PAMP that artificial intelligent optimization algorithm adjusts synthesis half ripe signal and mature signal
Number, the weight of danger signal, safety signal;Artificial intelligence optimization's algorithm includes particle swarm optimization algorithm and genetic algorithm
This optimization problem can be solved using the artificial intelligent optimization algorithm such as particle swarm optimization algorithm, genetic algorithm.
Compared with the fixation weight of existing DCA, this method can be more suitable for reality according to actual network environment to determine
The weight of situation, to improve detection accuracy.
Detailed description of the invention
Fig. 1 is the relational graph between weight defined in the present invention and the mature signal of synthesis and half ripe signal;
Fig. 2 is the optimization process of embodiment 1;
Fig. 3 is the optimization process of embodiment 2.
Specific embodiment
Present invention will be further explained below with reference to the attached drawings and examples, and embodiments of the present invention include but is not limited to
The following example.
Embodiment
One group of weight for synthesizing costimulatory signal remains unchanged, and constructs one to synthesize half ripe signal and synthesis
Two groups of weights of mature signal are the optimization problem of variable, to obtain optimal weight.For each immature DC cell,
By its safety signal (safe signal), danger signal (danger signal) and PAMP signal collected, synthesis is corresponding
Costimulatory signal, half ripe signal and mature signal.Define wP,csm、wD,csm、wS,csmFor synthesis costimulatory signal when
One group of weight defines wP,semi、wD,semi、wS,semiTo synthesize one group of weight when half ripe signal, w is definedP,mat、wD,mat、
wS,matTo synthesize one group of weight when mature signal, composite formula is
ocsm=(1+IC) (wP,csm*P+wD,csm*D+wS,csm*S)
osemi=(1+IC) (wP,semi*P+wD,semi*D+wS,semi*S) (4)
omat=(1+IC) (wP,mat*P+wD,mat*D+wS,mat*S)
ocsm、osemi、omatRespectively indicate corresponding costimulatory signal, half ripe signal and mature signal.wP,csmIt indicates
The weight of PAMP signal when synthesizing costimulatory signal;wD,csmIndicate the weight of danger signal when synthesis costimulatory signal;
wS,csmIndicate the weight of safety signal when synthesis costimulatory signal;wP,semiIndicate PAMP signal when synthesis half ripe signal
Weight;wD,semiIndicate the weight of danger signal when synthesis half ripe signal;wS,semiIt indicates to believe safely when synthesis half ripe signal
Number weight;wP,matThe weight of PAMP signal when indicating to synthesize mature signal;wD,matDanger signal when indicating to synthesize mature signal
Weight;wS,matThe weight of safety signal when indicating to synthesize mature signal;P indicates PAMP signal;D indicates danger signal;S is indicated
Safety signal;
Each immature DC cell constantly accumulates costimulatory signal, half ripe signal and mature letter collected
Number, it is as follows:
Ccsm(t)=Ccsm(t-1)+Ocsm(t)
Csemi(t)=Csemi(t-1)+Osemi(t) (5)
Cmat(t)=Cmat(t-1)+Omat(t),
Ccsm(t)、Csemi(t)、Cmat(t) the costimulatory signal value of the accumulation after respectively indicating current acquisition signal, half
Mature signal value and mature signal value.If the costimulatory signal value of accumulation is greater than or equal to presetting threshold value, stop
Signal is acquired, and is broken up.By the size between the half ripe signal of judgement accumulation and the mature signal of accumulation, to determine the DC
Cell is divided into maturation or is divided into blast.I.e. if Csemi(t) > Cmat(t), then the immature DC cell
It is divided into a semi-matured DC cell, is otherwise a mature DC cell.The DC cell death simultaneously.In this algorithm, often
One DC only acquires a type of network behavior, if after the DC cell death, can generate a new DC cell to adopt
Collect such network behavior.
For each network behavior, its MACV value is calculated, is then judged as abnormal if it is greater than or equal to given threshold value
Otherwise behavior is normal network behavior, the calculation formula of MACV is as follows:
Sample of the selection containing n heterogeneous networks behavior synthesizes the weight of mature signal and half ripe signal to set, then
It can establish following optimization problem:
In above formula, n indicates that a total of different types of network behavior of n kind in selected sample, i are the i-th seed type
Network behavior, MACViIndicate the MACV value of i-th of network behavior,I ∈ { 1 ..., n } is the network behavior of the i-th seed type
Whether judging result is correctly evaluated, and correct judgment is then 1, is otherwise -1.Expression is to obtain to n network
The maximum times of behavior correctly judged.
By the w for optimizing formula (4)P,semi、wD,semi、wS,semi、wP,mat、wD,mat、wS,matValue, can obtain different
Mature signal and half ripe signal are synthesized, and then obtains different accumulation maturation signal value and accumulation half ripe signal value, finally
It obtains maximum
By taking 1999 data set of KDD Cup as an example, 1000 samples are picked out from data set first, include 9 kinds of different types
Network behavior.Each sample includes 41 attributes and 1 attack type.Attack type is as antigen, attribute 25,26,29,
38,40 are used to synthesize PAMP signal, and attribute 23,24 is used to synthesize safe signal, and attribute 12,31,32 is used to synthesize danger letter
Number, IC signal is set as 0 in this experiment.Mean value is taken to be used to after the attribute of each sample is normalized to section [0,100] respectively
As PAMP signal, safe signal, danger signal.The threshold value of MACV is set as 0.75, the threshold value of the costimulatory signal of accumulation
It is set as the random number of [100,500].For synthesizing the weight (w of costimulatory signalP,csm,wD,csm,wP,semi) it is fixed be set as (2,
1,2).The group number of the initial weight of synthesis half ripe signal and mature signal is determined according to the actual conditions of algorithm.Wherein one
Group (wP,semi,wD,semi,wS,semi,wP,mat,wD,mat,wS,mat) initial value be set as (0,0,3,2,1, -3), and other group
Initial weight is then by being randomly generated.
Embodiment 1
Optimize weight using particle swarm algorithm.The vector structure of each particle is (wP,semi,wD,semi,wS,semi,wP,mat,
wD,mat,wS,mat), population quantity is set as 30, and maximum number of iterations is 100 times.Particle swarm algorithm parameter c1, c2 is set as 2.
One group of (w in 30 groups of initial weights of populationP,semi,wD,semi,wS,semi,wP,mat,wD,mat,wS,mat) value be set as (0,
0,3,2,1, -3), and other 29 groups initial weights then by being randomly generated, the random number of generation is integer and is located at [- 3,3] model
In enclosing.The fitness function of particle is the objective function of optimization problem, in evolutionary process, the change in displacement range of particle be [- 3,
3], the speed variation of particle is [- 1,1].It is maximum when obtainingEqual to 9, or reach maximum the number of iterations
When, optimization process terminates.Optimization process as shown in Fig. 2, after 13 iteration, weigh required for being by the optimal particle of acquisition
Value is (0,0,3,2,1, -1).
Embodiment 2
Optimize weight using genetic algorithm.wP,semi、wD,semi、wS,semi、wP,mat、wD,mat、wS,matFor the base of chromosome
Cause is transformed into binary form, then the structure of each chromosome is (II (wP,semi),II(wD,semi),II(wS,semi),II
(wP,mat),II(wD,mat),II(wS,mat)), II indicates the binary form of corresponding weight value, and the gene range of binary representation is
[- 3,3], such as 011 indicates that 3,111 indicate -3.The quantity of chromosome is set as 30, and maximum number of iterations is 100 times.Using wheel disc
Back-and-forth method is gambled to obtain selection operator, crossover operator is set as 0.1, and mutation operator is set as 0.05.The fitness function of genetic algorithm
For the objective function of optimization problem, at the end of optimization process, the optimal chromosome of acquisition is required weight.30 groups of dyes
One group of (w in the initial weight of colour solidP,semi,wD,semi,wS,semi,wP,mat,wD,mat,wS,mat) value be set as (0,0,3,2,
1, -3), and other 29 groups initial weights then by being randomly generated, the random number of generation is integer and is located in [- 3,3] range.
The fitness function of genetic algorithm is the objective function of optimization problem, maximum when obtainingEqual to n, or reach maximum
The number of iterations when.Fig. 3 is the optimum results figure of genetic algorithm, when the 16th iteration obtain optimal weight i.e. (0,0,3,2,
1,-1)。
Claims (4)
1. a kind of method for setting intrusion detection DCA algorithm weight, using the network behavior of immature DC cell collection sample,
Network behavior is detected according to the DC cell after differentiation, which is characterized in that according to actual optimization of network environment setting synthesis half
The PAMP signal of mature signal and mature signal, danger signal, safety signal weight, improve the detection accuracy of DCA algorithm;It is right
In each immature DC cell, the method for synthesizing costimulatory signal, synthesis half ripe signal and mature signal are as follows:
P indicates PAMP signal;D indicates danger signal;S indicates safety signal;ocsmIndicate costimulatory signal, osemiIndicate half at
Ripe signal, omatIndicate mature signal;wP,csmIndicate the weight of PAMP signal when synthesis costimulatory signal;wD,csmIndicate synthesis
The weight of danger signal when costimulatory signal;wS,csmIndicate the weight of safety signal when synthesis costimulatory signal;wP,semiTable
The weight of PAMP signal when showing synthesis half ripe signal;wD,semiIndicate the weight of danger signal when synthesis half ripe signal;
wS,semiIndicate the weight of safety signal when synthesis half ripe signal;wP,matIndicate the power of PAMP signal when synthesizing mature signal
Value;wD,matThe weight of danger signal when indicating to synthesize mature signal;wS,matThe power of safety signal when indicating to synthesize mature signal
Value;IC indicates inflammatory signal;
Each immature DC cell constantly accumulates costimulatory signal, half ripe signal and mature signal collected
Ccsm(t)、Csemi(t)、Cmat(t) respectively indicate current period accumulation costimulatory signal value, half ripe signal value and at
Ripe signal value, t indicate current period number;If the costimulatory signal value of accumulation is greater than or equal to presetting threshold value, stop
Only acquire signal;If Csemi(t) > Cmat(t), then the immature DC cell differentiation is a semi-matured DC cell, no
Then it is divided into a mature DC cell, while the DC cell death;
For each network behavior, its MACV value is calculated:
Sample of the selection containing n heterogeneous networks behavior is believed its PAMP for synthesizing half ripe signal and mature signal is set separately
Number, the weight of danger signal, safety signal, establish following optimization problem:
N indicates that a total of different types of network behavior of n kind in selected sample, i are the network behavior of the i-th seed type,
MACViIndicate the MACV value of i-th of network behavior, i ∈ { 1 ..., n } be the network behavior of the i-th seed type judging result whether
Correctly evaluation, otherwise it is -1 that correct judgment, which is then 1,;It indicates to obtain and n network behavior is correctly judged most
Big number;Respectively optimal weight is for the PAMP signal of synthesis half ripe signal and mature signal, danger signal, safety signal
MakeObtain the value when maximum times correctly judged to n network behavior.
2. a kind of method for setting intrusion detection DCA algorithm weight as described in claim 1, which is characterized in that for each
A immature DC cell only acquires a type of network behavior, if the DC cell death, it is thin to generate a new DC
Born of the same parents acquire such network behavior.
3. a kind of method for setting intrusion detection DCA algorithm weight according to claim 1, which is characterized in that for synthesizing
The weight of costimulatory signal remains unchanged.
4. a kind of method for setting intrusion detection DCA algorithm weight according to claim 1, which is characterized in that using artificial
Intelligent optimization algorithm adjusts the weight of the PAMP signal for synthesizing half ripe signal and mature signal, danger signal, safety signal;Institute
Stating artificial intelligence optimization's algorithm includes particle swarm optimization algorithm and genetic algorithm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610380426.3A CN106022134B (en) | 2016-06-01 | 2016-06-01 | A method of setting intrusion detection DCA algorithm weight |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610380426.3A CN106022134B (en) | 2016-06-01 | 2016-06-01 | A method of setting intrusion detection DCA algorithm weight |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106022134A CN106022134A (en) | 2016-10-12 |
CN106022134B true CN106022134B (en) | 2018-12-18 |
Family
ID=57092894
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610380426.3A Expired - Fee Related CN106022134B (en) | 2016-06-01 | 2016-06-01 | A method of setting intrusion detection DCA algorithm weight |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106022134B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108199875A (en) * | 2017-12-29 | 2018-06-22 | 上海上讯信息技术股份有限公司 | A kind of Network Intrusion Detection System and method |
CN110061986B (en) * | 2019-04-19 | 2021-05-25 | 长沙理工大学 | Network intrusion anomaly detection method based on combination of genetic algorithm and ANFIS |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103679025A (en) * | 2013-11-26 | 2014-03-26 | 南京邮电大学 | Malicious code detection method based on dendritic cell algorithm |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9596259B2 (en) * | 2014-11-05 | 2017-03-14 | The Boeing Company | Method for combining multiple signal values in the dendritic cell algorithm |
-
2016
- 2016-06-01 CN CN201610380426.3A patent/CN106022134B/en not_active Expired - Fee Related
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103679025A (en) * | 2013-11-26 | 2014-03-26 | 南京邮电大学 | Malicious code detection method based on dendritic cell algorithm |
Non-Patent Citations (1)
Title |
---|
树突状细胞算法原理及其应用;陈岳兵 等;《计算机工程》;20100420;第36卷(第8期);第173-176页 * |
Also Published As
Publication number | Publication date |
---|---|
CN106022134A (en) | 2016-10-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104809722B (en) | A kind of fault diagnosis method based on infrared thermal imagery | |
CN108632279B (en) | Multilayer anomaly detection method based on network traffic | |
CN108040073A (en) | Malicious attack detection method based on deep learning in information physical traffic system | |
CN106022229B (en) | The abnormal behaviour recognition methods with the Back propagation neural networks of self-adaptive enhancement algorithm is extracted based on video motion information characteristics | |
CN109273096A (en) | A kind of risk management grading evaluation method based on machine learning | |
CN108052968B (en) | QSFLA-SVM perception intrusion detection method | |
CN106022134B (en) | A method of setting intrusion detection DCA algorithm weight | |
CN109726735A (en) | A kind of mobile applications recognition methods based on K-means cluster and random forests algorithm | |
CN106817248A (en) | A kind of APT attack detection methods | |
CN110459267A (en) | A kind of human body composition prediction technique based on improving expert inquiry method | |
CN106992823B (en) | Cognitive radio network spectrum sensing method | |
CN102903007A (en) | Method for optimizing disaggregated model by adopting genetic algorithm | |
CN106960244A (en) | A kind of genetic algorithm and the evolution algorithm of particle cluster algorithm Parallel Fusion | |
CN102184454A (en) | Granulator formula generation method based on neural network system | |
CN109344956A (en) | Based on the SVM parameter optimization for improving Lay dimension flight particle swarm algorithm | |
CN109872773A (en) | Mirco-RNA precursor recognition methods based on the fusion of Adaboost, BP neural network and random forest | |
CN110413601A (en) | A kind of generating set Identification Data screening technique combined based on Gauss Naive Bayes Classifier and Predict error method | |
Epperson | Spatial genetic structure and non‐equilibrium demographics within plant populations | |
CN109525577A (en) | Malware detection method based on HTTP behavior figure | |
CN109615027B (en) | Intelligent prediction method for extracting wind speed characteristics along high-speed railway | |
Croix et al. | Training a multi-criteria decision system and application to the detection of PHP webshells | |
CN114386024A (en) | Power intranet terminal equipment abnormal attack detection method based on ensemble learning | |
CN109948738A (en) | Energy consumption method for detecting abnormality, the apparatus and system of coating drying room | |
CN113780432A (en) | Intelligent detection method for operation and maintenance abnormity of network information system based on reinforcement learning | |
CN113657678A (en) | Power grid power data prediction method based on information freshness |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20210604 Address after: 233000 south head workshop, No.11, building 2, liukmhuayuan, Changqing Township, Yuhui District, Bengbu City, Anhui Province Patentee after: Anhui Xuanwei Technology Co.,Ltd. Address before: 416000 No. 120 Renmin South Road, Jishou City, Xiangxi Tujia and Miao Autonomous Prefecture, Hunan Patentee before: JISHOU University |
|
TR01 | Transfer of patent right | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20181218 |
|
CF01 | Termination of patent right due to non-payment of annual fee |