With the rapid development and application of the mobile Internet, it is necessary to analyze and classify mobile traffic to meet the needs of users. Due to the difficulty in collecting some application data, the mobile traffic data presents a long-tailed distribution, resulting in a decrease in classification accuracy. In addition, the original GAN is difficult to train, and it is prone to "mode collapse". Therefore, this paper introduces the self-attention mechanism and gradient normalization into the auxiliary classifier generative adversarial network to form SA-ACGAN-GN model to solve the long-tailed distribution and training stability problems of mobile traffic data. This method firstly converts the traffic into images; secondly, to improve the quality of the generated images, the self-attention mechanism is introduced into the ACGAN model to obtain the global geometric features of the images; finally, the gradient normalization strategy is added to SA-ACGAN to further improve the data augmentation effect and improve the training stability. It can be seen from the cross-validation experimental data that, on the basis of using the same classifier, the SA-ACGAN-GN algorithm proposed in this paper, compared with other comparison algorithms, has the best precision reaching 93.8%; after adding gradient normalization, during the training process of the model, the classification loss decreases rapidly and the loss curve fluctuates less, indicating that the method proposed in this paper can not only effectively improve the long-tail problem of the dataset, but also enhance the stability of the model training.
Citation: Xingyu Gong, Ling Jia, Na Li. Research on mobile traffic data augmentation methods based on SA-ACGAN-GN[J]. Mathematical Biosciences and Engineering, 2022, 19(11): 11512-11532. doi: 10.3934/mbe.2022536
[1] | Jagan Mohan Jonnalagadda . On a nabla fractional boundary value problem with general boundary conditions. AIMS Mathematics, 2020, 5(1): 204-215. doi: 10.3934/math.2020012 |
[2] | Lakhdar Ragoub, J. F. Gómez-Aguilar, Eduardo Pérez-Careta, Dumitru Baleanu . On a class of Lyapunov's inequality involving λ-Hilfer Hadamard fractional derivative. AIMS Mathematics, 2024, 9(2): 4907-4924. doi: 10.3934/math.2024239 |
[3] | Wei Zhang, Jifeng Zhang, Jinbo Ni . New Lyapunov-type inequalities for fractional multi-point boundary value problems involving Hilfer-Katugampola fractional derivative. AIMS Mathematics, 2022, 7(1): 1074-1094. doi: 10.3934/math.2022064 |
[4] | Jaganmohan Jonnalagadda, Basua Debananda . Lyapunov-type inequalities for Hadamard type fractional boundary value problems. AIMS Mathematics, 2020, 5(2): 1127-1146. doi: 10.3934/math.2020078 |
[5] | Shuqin Zhang, Lei Hu . The existence of solutions and generalized Lyapunov-type inequalities to boundary value problems of differential equations of variable order. AIMS Mathematics, 2020, 5(4): 2923-2943. doi: 10.3934/math.2020189 |
[6] | Dumitru Baleanu, Muhammad Samraiz, Zahida Perveen, Sajid Iqbal, Kottakkaran Sooppy Nisar, Gauhar Rahman . Hermite-Hadamard-Fejer type inequalities via fractional integral of a function concerning another function. AIMS Mathematics, 2021, 6(5): 4280-4295. doi: 10.3934/math.2021253 |
[7] | Jonas Ogar Achuobi, Edet Peter Akpan, Reny George, Austine Efut Ofem . Stability analysis of Caputo fractional time-dependent systems with delay using vector lyapunov functions. AIMS Mathematics, 2024, 9(10): 28079-28099. doi: 10.3934/math.20241362 |
[8] | Chantapish Zamart, Thongchai Botmart, Wajaree Weera, Prem Junsawang . Finite-time decentralized event-triggered feedback control for generalized neural networks with mixed interval time-varying delays and cyber-attacks. AIMS Mathematics, 2023, 8(9): 22274-22300. doi: 10.3934/math.20231136 |
[9] | Yitao Yang, Dehong Ji . Properties of positive solutions for a fractional boundary value problem involving fractional derivative with respect to another function. AIMS Mathematics, 2020, 5(6): 7359-7371. doi: 10.3934/math.2020471 |
[10] | Tingting Guan, Guotao Wang, Haiyong Xu . Initial boundary value problems for space-time fractional conformable differential equation. AIMS Mathematics, 2021, 6(5): 5275-5291. doi: 10.3934/math.2021312 |
With the rapid development and application of the mobile Internet, it is necessary to analyze and classify mobile traffic to meet the needs of users. Due to the difficulty in collecting some application data, the mobile traffic data presents a long-tailed distribution, resulting in a decrease in classification accuracy. In addition, the original GAN is difficult to train, and it is prone to "mode collapse". Therefore, this paper introduces the self-attention mechanism and gradient normalization into the auxiliary classifier generative adversarial network to form SA-ACGAN-GN model to solve the long-tailed distribution and training stability problems of mobile traffic data. This method firstly converts the traffic into images; secondly, to improve the quality of the generated images, the self-attention mechanism is introduced into the ACGAN model to obtain the global geometric features of the images; finally, the gradient normalization strategy is added to SA-ACGAN to further improve the data augmentation effect and improve the training stability. It can be seen from the cross-validation experimental data that, on the basis of using the same classifier, the SA-ACGAN-GN algorithm proposed in this paper, compared with other comparison algorithms, has the best precision reaching 93.8%; after adding gradient normalization, during the training process of the model, the classification loss decreases rapidly and the loss curve fluctuates less, indicating that the method proposed in this paper can not only effectively improve the long-tail problem of the dataset, but also enhance the stability of the model training.
The well-known classical Lyapunov inequality [15] states that, if u is a nontrivial solution of the Hill's equation
u′′(t)+q(t)u(t)=0, a<t<b, | (1.1) |
subject to Dirichlet-type boundary conditions:
u(a)=u(b)=0, | (1.2) |
then
∫ba|q(t)|dt>4b−a, | (1.3) |
where q:[a,b]→R is a real and continuous function.
Later, in 1951, Wintner [24], obtained the following inequality:
∫baq+(t)dt>4b−a, | (1.4) |
where q+(t)=max{q(t),0}.
A more general inequality was given by Hartman and Wintner in [12], that is known as Hartman Wintner-type inequality:
∫ba(t−a)(b−t)q+(t)dt>b−a, | (1.5) |
Since maxt∈[a,b](t−a)(b−t)=(b−a)24, then, (1.5) implies (1.4).
The Lyapunov inequality and its generalizations have many applications in different fields such in oscillation theory, asymptotic theory, disconjugacy, eigenvalue problems.
Recently, many authors have extended the Lyapunov inequality (1.3) for fractional differential equations [1,2,3,4,5,6,7,8,9,10,11,12,13,15,18,20,22,23,24]. For this end, they substituted the ordinary second order derivative in (1.1) by a fractional derivative or a conformable derivative. The first result in which a fractional derivative is used instead of the ordinary derivative in equation (1.1), is the work of Ferreira [6]. He considered the following two-point Riemann-Liouville fractional boundary value problem
Dαa+u(t)+q(t)u(t)=0, a<t<b, 1<α≤2 |
u(a)=u(b)=0. |
And obtained the Lyapunov inequality:
∫ba|q(t)|dt>Γ(α)(4b−a)α−1. |
Then, he studied in [7], the Caputo fractional differential equation
CDαa+u(t)+q(t)u(t)=0, a<t<b, 1<α≤2 |
under Dirichlet boundary conditions (1.2). In this case, the corresponding Lyapunov inequality has the form
∫ba|q(t)|dt>ααΓ(α)((α−1)(b−a))α−1. |
Later Agarwal and Özbekler in [1], complimented and improved the work of Ferreira [6]. More precisely, they proved that if u is a nontrivial solution of the Riemann-Liouville fractional forced nonlinear differential equations of order α∈(0,2]:
Dαa+u(t)+p(t)|u(t)|μ−1u(t)+q(t)|u(t)|γ−1u(t)=f(t), a<t<b, |
satisfying the Dirichlet boundary conditions (1.2), then the following Lyapunov type inequality
(∫ba[p+(t)+q+(t)]dt)(∫ba[μ0p+(t)+γ0q+(t)+|f(t)|]dt)>42α−3Γ2(α)(b−a)2α−2. |
holds, where p, q, f are real-valued functions, 0<γ<1<μ<2, μ0=(2−μ)μμ/(2−μ)22/(μ−2) and γ0=(2−γ)γγ/(2−γ)22/(γ−2).
In 2017, Guezane-Lakoud et al. [11], derived a new Lyapunov type inequality for a boundary value problem involving both left Riemann-Liouville and right Caputo fractional derivatives in presence of natural conditions
−CDαb−Dβa+u(t)+q(t)u(t)=0, a<t<b, 0<α,β≤1 |
u(a)=Dβa+u(b)=0, |
then, they obtained the following Lyapunov inequality:
∫ba|q(t)|dt>(α+β−1)Γ(α)Γ(β)(b−a)α+β−1. |
Recently, Ferreira in [9], derived a Lyapunov-type inequality for a sequential fractional right-focal boundary value problem
CDαa+Dβa+u(t)+q(t)u(t)=0, a<t<b |
u(a)=Dγa+u(b)=0, |
where 0<α,β,γ≤1, 1<α+β≤2, then, they obtained the following Lyapunov inequality:
∫ba(b−s)α+β−γ−1|q(t)|dt>1C, |
where
C=(b−a)γmax{Γ(β−γ+1)Γ(α+β−γ)Γ(β+1),1−αβΓ(α+β)(Γ(β−γ+1)Γ(α+β−1)Γ(α+β−γ)Γ(β))α+β−1α−1, with α<1} |
Note that more generalized Lyapunov type inequalities have been obtained for conformable derivative differential equations in [13]. For more results on Lyapunov-type inequalities for fractional differential equations, we refer to the recent survey of Ntouyas et al. [18].
In this work, we obtain Lyapunov type inequality for the following mixed fractional differential equation involving both right Caputo and left Riemann-Liouville fractional derivatives
−CDαb−Dβa+u(t)+q(t)u(t)=0, a<t<b, | (1.6) |
satisfying the Dirichlet boundary conditions (1.2), here 0<β≤α≤1, 1<α+β≤2, CDαb− denotes right Caputo derivative, Dβa+ denotes the left Riemann-Liouville and q is a continuous function on [a,b].
So far, few authors have considered sequential fractional derivatives, and some Lyapunov type inequalities have been obtained. In this study, we place ourselves in a very general context, in that in each fractional operator, the order of the derivative can be different. Such problems, with both left and right fractional derivatives arise in the study of Euler-Lagrange equations for fractional problems of the calculus of variations [2,16,17]. However, the presence of a mixed left and right Caputo or Riemann-Liouville derivatives of order 0<α<1 leads to great difficulties in the study of the properties of the Green function since in this case it's given as a fractional integral operator.
We recall the concept of fractional integral and derivative of order p>0. For details, we refer the reader to [14,19,21]
The left and right Riemann-Liouville fractional integral of a function g are defined respectively by
Ipa+g(t)=1Γ(p)∫tag(s)(t−s)1−pds,Ipb−g(t)=1Γ(p)∫btg(s)(s−t)1−pds. |
The left and right Caputo derivatives of order p>0, of a function g are respectively defined as follows:
CDpa+g(t)=In−pa+g(n)(t),CDpb−g(t)=(−1)nIn−pb−g(n)(t), |
and the left and right Riemann-Liouville fractional derivatives of order p>0, of a function g\ are respectively defined as follows:
Dpa+g(t)=dndtn(In−pa+g)(t),Dpb−g(t)=(−1)ndndtnIn−pb−g(t), |
where n is the smallest integer greater or equal than p.
We also recall the following properties of fractional operators. Let 0<p<1, then:
1- IpCa+Dpa+f(t)=f(t)−f(a).
2- IpCb−Dpb−f(t)=f(t)−f(b).
3- (Ipa+c)(t)=c(t−a)pΓ(p+1),c∈R
4- Dpa+u(t)=CDpa+u(t), when u(a)=0.
5- Dpb−u(t)=CDpb−u(t), when u(b)=0.
Next we transform the problem (1.6) with (1.2) to an equivalent integral equation.
Lemma 1. Assume that 0<α,β≤1. The function u is a solution to the boundary value problem (1.6) with (1.2) if and only if u satisfies the integral equation
u(t)=∫baG(t,r)q(r)u(r)dr, | (2.1) |
where
G(t,r)=1Γ(α)Γ(β)(∫inf{r,t}a(t−s)β−1(r−s)α−1ds |
−(t−a)β(b−a)β∫ra(b−s)β−1(r−s)α−1ds) | (2.2) |
is the Green's function of problem (1.6) with (1.2).
Proof. Firstly, we apply the right side fractional integral Iαb− to equation (1.6), then the left side fractional integral Iβa+ to the resulting equation and taking into account the properties of Caputo and\Riemann-Liouville fractional derivatives and the fact that Dβa+u(t)=CDβa+u(t), we get
u(t)=Iβa+Iαb−q(t)u(t)+c(t−a)βΓ(β+1). | (2.3) |
In view of the boundary condition u(b)=0, we get
c=−Γ(β+1)(b−a)βIβa+Iαb−q(t)u(t)∣t=b. |
Substituting c in (2.3), it yields
u(t)=Iβa+Iαb−q(t)u(t)−(t−a)β(b−a)βIβa+Iαb−q(t)u(t)∣t=b=1Γ(α)Γ(β)∫ta(t−s)β−1(∫bs(r−s)α−1q(r)u(r)dr)ds−(t−a)βΓ(α)Γ(β)(b−a)β∫ba(b−s)β−1(∫bs(r−s)α−1q(r)u(r)dr)ds. |
Finally, by exchanging the order of integration, we get
u(t)=1Γ(α)Γ(β)∫ta(∫ra(t−s)β−1(r−s)α−1ds)q(r)u(r)dr+1Γ(α)Γ(β)∫bt(∫ta(t−s)β−1(r−s)α−1ds)q(r)u(r)dr−(t−a)βΓ(α)Γ(β)(b−a)β∫ba(∫ra(b−s)β−1(r−s)α−1ds)q(r)u(r)dr, |
thus
u(t)=∫baG(t,r)q(r)u(r)dr, |
with
G(t,r)=1Γ(α)Γ(β){∫ra(t−s)β−1(r−s)α−1ds−(t−a)β(b−a)β∫ra(b−s)β−1(r−s)α−1ds,a≤r≤t≤b,∫ta(t−s)β−1(r−s)α−1ds−(t−a)β(b−a)β∫ra(b−s)β−1(r−s)α−1ds,a≤t≤r≤b. |
that can be written as
G(t,r)=1Γ(α)Γ(β)(∫inf{r,t}a(t−s)β−1(r−s)α−1ds−(t−a)β(b−a)β∫ra(b−s)β−1(r−s)α−1ds). |
Conversely, we can verify that if u satisfies the integral equation (2.1), then u is a solution to the boundary value problem (1.6) with (1.2). The proof is completed.
In the next Lemma we give the property of the Green function G that will be needed in the sequel.
Lemma 2. Assume that 0<β≤α≤1,1<α+β≤2, then the Green function G(t,r) given in (2.2) of problem (1.6) with (1.2) satisfies the following property:
|G(t,r)|≤1Γ(α)Γ(β)(α+β−1)(α+β)(α(b−a)(β+α))α+β−1, |
for all a≤r≤t≤b.
Proof. Firstly, for a≤r≤t≤b, we have G(t,r)≥0. In fact, we have
G(t,r)=1Γ(α)Γ(β)(∫ra(t−s)β−1(r−s)α−1ds−(t−a)β(b−a)β∫ra(b−s)β−1(r−s)α−1ds)≥1Γ(α)Γ(β)(∫ra(b−s)β−1(r−s)α−1ds−(t−a)β(b−a)β∫ra(b−s)β−1(r−s)α−1ds) |
=1Γ(α)Γ(β)(1−(t−a)β(b−a)β)∫ra(b−s)β−1(r−s)α−1ds≥0 | (2.4) |
in addition,
G(t,r)≤1Γ(α)Γ(β)(∫ra(r−s)β−1(r−s)α−1ds−(r−a)β(b−a)β∫ra(b−s)β−1(r−s)α−1ds)≤1Γ(α)Γ(β)((r−a)α+β−1(α+β−1)−(r−a)β(b−a)β∫ra(b−a)β−1(r−s)α−1ds) |
=1Γ(α)Γ(β)((r−a)α+β−1(α+β−1)−(r−a)β+αα(b−a)). | (2.5) |
Thus, from (2.4) and (2.5), we get
0≤G(t,r)≤h(r), a≤r≤t≤b, | (2.6) |
where
h(s):=1Γ(α)Γ(β)((s−a)α+β−1(α+β−1)−(s−a)β+αα(b−a)), |
it is clear that h(s)≥0, for all s∈[a,b].
Now, for a≤t≤r≤b, we have
G(t,r)=1Γ(α)Γ(β)(∫ta(t−s)β−1(r−s)α−1ds−(t−a)β(b−a)β∫ra(b−s)β−1(r−s)α−1ds)≤1Γ(α)Γ(β)(∫ta(t−s)β−1(t−s)α−1ds−(t−a)β(b−a)∫ra(r−s)α−1ds)=1Γ(α)Γ(β)((t−a)α+β−1(α+β−1)−(t−a)β(r−a)αα(b−a)) |
≤1Γ(α)Γ(β)((t−a)α+β−1(α+β−1)−(t−a)β+αα(b−a))=h(t). | (2.7) |
On the other hand,
G(t,r)≥1Γ(α)Γ(β)(r−a)α−1∫ta(t−s)β−1ds−(t−a)β(b−a)β∫ra(r−s)β−1(r−s)α−1ds)≥1Γ(α)Γ(β)((t−a)α(t−a)ββ(b−a)−(t−a)β(b−a)β(r−a)α+β−1(α+β−1))≥1Γ(α)Γ(β)((t−a)α+ββ(b−a)−(t−a)β(r−a)α−1(α+β−1))≥1Γ(α)Γ(β)((t−a)α+ββ(b−a)−(t−a)α+β−1(α+β−1)), |
since β≤α, we get
G(t,r)≥−h(t), a≤t≤r≤b. | (2.8) |
From (2.7) and (2.8) we obtain
|G(t,r)|≤h(t), a≤t≤r≤b. | (2.9) |
Finally, by differentiating the function h, it yields
h′(s)=1Γ(α)Γ(β)(s−a)α+β−2(1−(β+α)(s−a)α(b−a)). |
We can see that h′(s)=0 for s0=a+α(b−a)(β+α)∈(a,b), h′(s)<0 for s>s0 and h′(s)>0 for s<s0. Hence, the function h(s) has a unique maximum given by
maxs∈[a,b]h(s)=h(s0)=1Γ(α)Γ(β)((α(b−a)(β+α))α+β−1(α+β−1)−(α(b−a)(β+α))β+αα(b−a))=1Γ(α)Γ(β)(α+β−1)(α+β)(α(b−a)(β+α))α+β−1. |
From (2.6) and (2.9), we get |G(t,r)|≤h(s0), from which the intended result follows.
Next, we state and prove the Lyapunov type inequality for problem (1.6) with (1.2).
Theorem 3. Assume that 0<β≤α≤1 and 1<α+β≤2. If the fractional boundary value problem (1.6) with (1.2) has a nontrivial continuous solution, then
∫ba|q(r)|dr≥Γ(α)Γ(β)(α+β−1)(α+β)α+β(α(b−a))α+β−1. | (2.10) |
Proof. Let X=C[a,b] be the Banach space endowed with norm ||u||=maxt∈[a,b]|u(t)|. It follows from Lemma 1 that a solution u∈X to the boundary value problem (1.6) with (1.2) satisfies
|u(t)|≤∫ba|G(t,r)||q(r)||u(r)|dr≤‖u‖∫ba|G(t,r)|q(r)dr, |
Now, applying Lemma 2 to equation (2.1), it yields
|u(t)|≤1Γ(α)Γ(β)(α+β−1)(α+β)(α(b−a)(β+α))α+β−1‖u‖∫ba|q(r)|dr |
Hence,
‖u‖≤(α(b−a))α+β−1Γ(α)Γ(β)(α+β−1)(α+β)α+β‖u‖∫ba|q(r)|dr, |
from which the inequality (2.10) follows. Note that the constant in (2.10) is not sharp. The proof is completed.
Remark 4. Note that, according to boundary conditions (1.2), the Caputo derivatives CDαb− and CDβa+ coincide respectively with the Riemann-Liouville derivatives Dαb− and Dβa+. So, equation (1.6) is reduced to the one containing only Caputo derivatives or only Riemann-Liouville derivatives, i.e.,
−CDαCb−Dβa+u(t)+q(t)u(t)=0, a<t<b |
or
−Dαb−Dβa+u(t)+q(t)u(t)=0, a<t<b |
Furthermore, by applying the reflection operator (Qf)(t)=f(a+b−t) and taking into account that QCDαa+=CDαb−Q and QCDβb−=CDβa+Q (see [21]), we can see that, the boundary value problem (1.6) with (1.2) is equivalent to the following problem
−CDαa+Dβb−u(t)+q(t)u(t)=0, a<t<b, |
u(a)=u(b)=0. |
Remark 5. If we take α=β=1, then the Lyapunov type inequality (2.3) is reduced to
∫ba|q(t)|dt≥4b−a. |
The authors thank the anonymous referees for their valuable comments and suggestions that improved this paper.
All authors declare no conflicts of interest in this paper.
[1] |
T. T. T. Nguyen, G. Armitage, A survey of techniques for internet traffic classification using machine learning, IEEE Commun. Surv. Tutorials, 10 (2008), 56–76. https://doi.org/10.1109/SURV.2008.080406 doi: 10.1109/SURV.2008.080406
![]() |
[2] | A. Razaghpanah, A. A. Niaki, N. Vallina-Rodriguez, S. Sundaresan, J. Amann, P. Gill, Studying TLS usage in Android apps, in Proceedings of the 13th International Conference on emerging Networking Experiments and Technologies, (2017), 350–362. https://doi.org/10.1145/3232755.3232779 |
[3] | M. Talyansky, A. Tumarkin. System and method for optimizing inter-node communication in content distribution network, P, US14383062, 2018-01-09. |
[4] | D. Li, Y.F. Zhu, W. Lin. Traffic identification of mobile Apps base on variational autoencoder network. in 2017 13th International Conference on Computational Intelligence and Security (CIS), (2017), 287–291. http://doi.org/10.1109/CIS.2017.00069 |
[5] | G. Aceto, D. Ciuonzo, A. Montieri, A.Pescapé, Mobile encrypted traffic classification using deep learning, in 2018 IEEE/ACM Network Traffic Measurement and Analysis Conference, (2018), 1–8. http://doi.org/10.23919/TMA.2018.8506558 |
[6] |
G. Aceto, D. Ciuonzo, A. Montieri, A. Pescapé, MIMETIC: Mobile encrypted traffic classification using multimodal deep learning, Comput. Networks, 165 (2019). http://doi.org/10.1016/j.comnet.2019.106944 doi: 10.1016/j.comnet.2019.106944
![]() |
[7] |
G. Aceto, D. Ciuonzo, A. Montieri, A. Pescapé, DISTILLER: Encrypted traffic classification via multimodal multitask deep learning, Network Comput. Appl., 183–184 (2021), 102985. https://doi.org/10.1016/j.jnca.2021.102985 doi: 10.1016/j.jnca.2021.102985
![]() |
[8] | C. Liu, L. He, G. Xiong, Z. Cao, Z. Li, FS-Net: A flow sequence network for encrypted traffic classification, in 2019 IEEE Conference on Computer Communications, (2019), 1171–1179. http://doi.org/10.1109/INFOCOM.2019.8737507 |
[9] |
A. Nascita, A. Montieri, G. Aceto, D. Ciuonzo, V. Persico, A. Pescapé, XAI meets mobile traffic classification: Understanding and improving multimodal deep learning architectures, IEEE Trans. Network Serv. Manage., 18 (2021), 4225–4246. http://doi.org/10.1109/TNSM.2021.3098157 doi: 10.1109/TNSM.2021.3098157
![]() |
[10] |
M. Galar, A. Fernandez, E. Barrenechea, H. Bustince, F. Herrera, A review on ensembles for the class imbalance problem: bagging-, boosting-, and hybrid-based approaches. IEEE Trans. Syst. Man Cybern., 42 (2012), 463–484. https://doi.org/10.1109/TSMCC.2011.2161285 doi: 10.1109/TSMCC.2011.2161285
![]() |
[11] |
J. Lee, K. Park, Gan-based imbalanced data intrusion detection system, Pers. Ubiquitous Comput., 25(2021), 121–128. https://doi.org/10.1007/s00779-019-01332-y doi: 10.1007/s00779-019-01332-y
![]() |
[12] |
Y. Hong, U. Hwang, J. Yoo, S. Yoon, How generative adversarial networks and their variants work: An overview, ACM Comput. Surv., 52 (2019), 43. https://doi.org/10.1145/3301282 doi: 10.1145/3301282
![]() |
[13] | Y. L. Wu, H. H. Shuai, Z. R. Tam, H. Y. Chiu, Gradient normalization for generative adversarial networks, in International Conference on Computer Vision, 2021. https://doi.org/10.48550/arXiv.2109.02235 |
[14] | N. Japkowicz, Learning from imbalanced data sets: A comparison of various strategies, in AAAI workshop learning from imbalanced data sets, (2000), 10–15. |
[15] |
N. Chawla, K. Bowyer, L. Hall, W. Kegelmeyer, Smote: Synthetic minority over-sampling technique, Artif. Intell. Res., 16 (2002), 321–357. https://doi.org/10.1613/jair.953 doi: 10.1613/jair.953
![]() |
[16] | H. He, Y. Bai, E. A. Garcia, ADASYN: Adaptive synthetic sampling approach for imbalanced learning., in Proceedings of International Joint Conference on Neural Networks, (2008), 1322–1328. |
[17] |
Z. Xu, D. Shen, T. Nie, Y. Kou, A hybrid sampling algorithm combining M-SMOTE and ENN based on Random forest for medical imbalanced data, Biomed. Inform., (2020). https://doi.org/10.1016/j.jbi.2020.103465 doi: 10.1016/j.jbi.2020.103465
![]() |
[18] |
R. Hasibi, M. Shokri, M. Dehghan, Augmentation scheme for dealing with imbalanced network traffic classification using deep learning, Comput. Sci. Networking Int. Archit., 2019. https://doi.org/10.48550/arXiv.1901.00204 doi: 10.48550/arXiv.1901.00204
![]() |
[19] |
M. A. Arefeen, S. T. Nimi, M. S. Rahman, Neural network-based undersampling techniques, IEEE Trans. Syst. Man Cybern. Syst., 2020. https://doi.org/10.1109/TSMC.2020.3016283 doi: 10.1109/TSMC.2020.3016283
![]() |
[20] |
F. Folino, G. Folino, M. Guarascio, F. S. Pisani, L. Pontieri, On learning effective ensembles of deep neural networks for intrusion detection, Inform. Fusion, 72 (2021), 48–69. https://doi.org/10.1016/j.inffus.2021.02.007 doi: 10.1016/j.inffus.2021.02.007
![]() |
[21] |
P. Bedi, N. Gupta, V. Jindal, I-SiamIDS: an improved Siam-IDS for handling class imbalance in network-based intrusion detection systems, Appl. Intell., 51 (2021), 1133–1151. https://doi.org/10.1007/s10489-020-01886-y doi: 10.1007/s10489-020-01886-y
![]() |
[22] |
N. Gupta, V. Jindal, P. Bedi, CSE-IDS: Using cost-sensitive deep learning and ensemble algorithms to handle class imbalance in network-based intrusion detection systems, Comput. Secur., 112 (2022), 102499. https://doi.org/10.1016/j.cose.2021.102499 doi: 10.1016/j.cose.2021.102499
![]() |
[23] | I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, et al., Generative adversarial nets, in Proceedings of the 27th International Conference on Neural Information Processing Systems, (2014), 2672–2680. |
[24] | L. C. Yang, S. Y. Chou, Y. H. Yang, MidiNet: A convolutional generative adversarial network for symbolic-domain music generation, preprint, arXiv: 1703.10847. |
[25] |
S. Yu, S. Zhang, B. Wang, H. Dun, L. Xu, X. Huang, et al., Generative adversarial network based data augmentation to improve cervical cell classification model, Math. Biosci. Eng., 18 (2021), 1740–1752. https://doi.org/10.3934/mbe.2021090 doi: 10.3934/mbe.2021090
![]() |
[26] |
R. Durai, R. N. Abirami, Identity preserving multi-pose facial expression recognition using fine tuned VGG on the latent space vector of generative adversarial network, Math. Biosci. Eng., 18 (2021), 3699–3717. https://doi.org/10.3934/mbe.2021186 doi: 10.3934/mbe.2021186
![]() |
[27] | H. Zhang, X. Yu, P. Ren, C. Luo, G. Min, Deep adversarial learning in intrusion detection: A data augmentation enhanced framework, preprint, arXiv: 1901.07949. |
[28] |
S. Park, H. Park, Combined oversampling and undersampling method based on slow-start algorithm for imbalanced network traffic, Computing, 103 (2021), 401–424. https://doi.org/10.1007/s00607-020-00854-1 doi: 10.1007/s00607-020-00854-1
![]() |
[29] | L. Vu, B. C. Thanh, U. Nguyen, A deep learning based method for handling imbalanced problem in network traffic classification, in Eighth International Symposium on Information and Communication Technology, (2017), 333–339. https://doi.org/10.1145/3155133.3155175 |
[30] |
G. Douzas, F. Bacao, Effective data generation for imbalanced learning using conditional generative adversarial networks, Expert Syst. Appl., 91 (2017), 464–471. https://doi.org/10.1016/j.eswa.2017.09.030 doi: 10.1016/j.eswa.2017.09.030
![]() |
[31] | M. Arjovsky, S. Chintala, L. Bottou, Wasserstein gan, preprint, arXiv: 1701.07875. |
[32] | I. Gulrajani, F. Ahmed, M. Arjovsky, V. Dumoulin, A. Courville, Improved training of Wasserstein gans, in Proceedings of the 31st International Conference on Neural Information Processing Systems, (2017), 5769–5779. |
[33] | T. Miyato, T. Kataoka, M. Koyama, Y. Yoshida, Spectral normalization for generative adversarial networks, in International Conference on Learning Representations, 2018. |
[34] |
H. W. Ding, L. Y. Chen, L. Dong, Z. W. Fu, X. H. Cui, Imbalanced data classification: A KNN and generative adversarial networks-based hybrid approach for intrusion detection, Future Gener. Comput. Syst., 131 (2022), 240–254. https://doi.org/10.1016/j.future.2022.01.026 doi: 10.1016/j.future.2022.01.026
![]() |
[35] |
M. Zheng, T. Li, R. Zhu, Y. H. Tang, M. J. Tang, L. L. Lin, et al., Conditional Wasserstein generative adversarial network-gradient penalty-based approach to alleviating imbalanced data classification, Inform. Sci., 512 (2020), 1009–1023. https://doi.org/10.1016/j.ins.2019.10.014 doi: 10.1016/j.ins.2019.10.014
![]() |
[36] |
L. M. Xua, X. H. Zeng, Z. W. Huang, W. S. Li, H. Zhang, Low-dose chest X-ray image super-resolution using generative adversarial nets with spectral normalization, Biomed. Signal Proc. Control, 55 (2020), 101600. https://doi.org/10.1016/j.bspc.2019.101600 doi: 10.1016/j.bspc.2019.101600
![]() |
1. | Aidyn Kassymov, Berikbol T. Torebek, Lyapunov-type inequalities for a nonlinear fractional boundary value problem, 2021, 115, 1578-7303, 10.1007/s13398-020-00954-9 | |
2. | Jie Wang, Shuqin Zhang, A Lyapunov-Type Inequality for Partial Differential Equation Involving the Mixed Caputo Derivative, 2020, 8, 2227-7390, 47, 10.3390/math8010047 |