Research article

Analysis of mobility based COVID-19 epidemic model using Federated Multitask Learning


  • Aggregating a massive amount of disease-related data from heterogeneous devices, a distributed learning framework called Federated Learning(FL) is employed. But, FL suffers in distributing the global model, due to the heterogeneity of local data distributions. To overcome this issue, personalized models can be learned by using Federated multitask learning(FMTL). Due to the heterogeneous data from distributed environment, we propose a personalized model learned by federated multitask learning (FMTL) to predict the updated infection rate of COVID-19 in the USA using a mobility-based SEIR model. Furthermore, using a mobility-based SEIR model with an additional constraint we can analyze the availability of beds. We have used the real-time mobility data sets in various states of the USA during the years 2020 and 2021. We have chosen five states for the study and we observe that there exists a correlation among the number of COVID-19 infected cases even though the rate of spread in each case is different. We have considered each US state as a node in the federated learning environment and a linear regression model is built at each node. Our experimental results show that the root-mean-square percentage error for the actual and prediction of COVID-19 cases is low for Colorado state and high for Minnesota state. Using a mobility-based SEIR simulation model, we conclude that it will take at least 400 days to reach extinction when there is no proper vaccination or social distance.

    Citation: M Kumaresan, M Senthil Kumar, Nehal Muthukumar. Analysis of mobility based COVID-19 epidemic model using Federated Multitask Learning[J]. Mathematical Biosciences and Engineering, 2022, 19(10): 9983-10005. doi: 10.3934/mbe.2022466

    Related Papers:

    [1] Edil D. Molina, Paul Bosch, José M. Sigarreta, Eva Tourís . On the variable inverse sum deg index. Mathematical Biosciences and Engineering, 2023, 20(5): 8800-8813. doi: 10.3934/mbe.2023387
    [2] V. R. Kulli, J. A. Méndez-Bermúdez, José M. Rodríguez, José M. Sigarreta . Revan Sombor indices: Analytical and statistical study. Mathematical Biosciences and Engineering, 2023, 20(2): 1801-1819. doi: 10.3934/mbe.2023082
    [3] Xinlin Liu, Viktor Krylov, Su Jun, Natalya Volkova, Anatoliy Sachenko, Galina Shcherbakova, Jacek Woloszyn . Segmentation and identification of spectral and statistical textures for computer medical diagnostics in dermatology. Mathematical Biosciences and Engineering, 2022, 19(7): 6923-6939. doi: 10.3934/mbe.2022326
    [4] A. Newton Licciardi Jr., L.H.A. Monteiro . A network model of social contacts with small-world and scale-free features, tunable connectivity, and geographic restrictions. Mathematical Biosciences and Engineering, 2024, 21(4): 4801-4813. doi: 10.3934/mbe.2024211
    [5] Qingqun Huang, Muhammad Labba, Muhammad Azeem, Muhammad Kamran Jamil, Ricai Luo . Tetrahedral sheets of clay minerals and their edge valency-based entropy measures. Mathematical Biosciences and Engineering, 2023, 20(5): 8068-8084. doi: 10.3934/mbe.2023350
    [6] Wenjuan Guo, Ming Ye, Xining Li, Anke Meyer-Baese, Qimin Zhang . A theta-scheme approximation of basic reproduction number for an age-structured epidemic system in a finite horizon. Mathematical Biosciences and Engineering, 2019, 16(5): 4107-4121. doi: 10.3934/mbe.2019204
    [7] J. A. Méndez-Bermúdez, José M. Rodríguez, José L. Sánchez, José M. Sigarreta . Analytical and computational properties of the variable symmetric division deg index. Mathematical Biosciences and Engineering, 2022, 19(9): 8908-8922. doi: 10.3934/mbe.2022413
    [8] Jair Castro, Ludwin A. Basilio, Gerardo Reyna, Omar Rosario . The differential on operator S(Γ). Mathematical Biosciences and Engineering, 2023, 20(7): 11568-11584. doi: 10.3934/mbe.2023513
    [9] Cheng-Peng Li, Cheng Zhonglin, Mobeen Munir, Kalsoom Yasmin, Jia-bao Liu . M-polynomials and topological indices of linear chains of benzene, napthalene and anthracene. Mathematical Biosciences and Engineering, 2020, 17(3): 2384-2398. doi: 10.3934/mbe.2020127
    [10] Muhammad Akram, Saba Siddique, Majed G. Alharbi . Clustering algorithm with strength of connectedness for m-polar fuzzy network models. Mathematical Biosciences and Engineering, 2022, 19(1): 420-455. doi: 10.3934/mbe.2022021
  • Aggregating a massive amount of disease-related data from heterogeneous devices, a distributed learning framework called Federated Learning(FL) is employed. But, FL suffers in distributing the global model, due to the heterogeneity of local data distributions. To overcome this issue, personalized models can be learned by using Federated multitask learning(FMTL). Due to the heterogeneous data from distributed environment, we propose a personalized model learned by federated multitask learning (FMTL) to predict the updated infection rate of COVID-19 in the USA using a mobility-based SEIR model. Furthermore, using a mobility-based SEIR model with an additional constraint we can analyze the availability of beds. We have used the real-time mobility data sets in various states of the USA during the years 2020 and 2021. We have chosen five states for the study and we observe that there exists a correlation among the number of COVID-19 infected cases even though the rate of spread in each case is different. We have considered each US state as a node in the federated learning environment and a linear regression model is built at each node. Our experimental results show that the root-mean-square percentage error for the actual and prediction of COVID-19 cases is low for Colorado state and high for Minnesota state. Using a mobility-based SEIR simulation model, we conclude that it will take at least 400 days to reach extinction when there is no proper vaccination or social distance.



    Let G be graph having vertex set V=V(G) and edge set E=E(G), then the adjacency matrix associated to the graph G, is

    [A(G)]={1        if   vi   is adjacent to  vj,0                  otherwise.

    Sum of absolute values of the eigenvalues associated to A(G) is known as energy of the graph G denoted by E(G) and the Largest eigenvalue of A(G) is called Spectral radius of the graph G and it is denoted by (G). For details and references relating to Spectral radii, follow [2,3,4]. The motivation of E(G) was initiated by Gutman in 1978 [1] but the idea could not get attention until 2000. Since 2003, rapid development in technology and computer awoke significant interests in these areas. The problem of determining extreme values of spectral radius has been extensively investigated, [5]. Partial solutions to these problems can be traced in [6,7,8,9,10,11]. Fiedler and Nikiforov [6] gave tight sufficient conditions for the existence of Hamilton paths and cycles in terms of the spectral radius of graphs or the complement of graphs. Lu et al. [7] studied sufficient conditions for Hamilton paths in connected graphs and Hamilton cycles in bipartite graphs in terms of the spectral radius of a graph. Some other spectral conditions for Hamilton paths and cycles in graphs have been given in [12,13,14,15,16].

    Horn et al.[17] and Gatmacher [18] used matrix analysis to relate it with graph energies. Balkrishnan, in [19], computed the sharp bounds for energy of a k-regular graph and proved that for n3, there always exists two equi-energetic graphs having order 4n which are not co-spectral. Bapat et al. proved that E(G) can not be an odd integer, [20] whereas Pirzada et al. [21] proved that it can not be square root of an odd integer. Jones [22] discussed E(G) of simple graphs relating it with closuring and algebraic connectivity. Different kinds of matrices energies associated to a graph are presented by Meenakshi et al. [23]. Nikiforov [24] obtained various results related to bounds of energies. Samir et al. [25] constructed 1-splitting and 1-shadow graph of any simple connected graph and proved that adjacency energies of these newly constructed graphs is constant multiple of the energies of the original graph. Samir et al. [26] then generalized the idea of 1-splitting and 2-shadow graph to arbitrary s-splitting and s-shadow graph where s>0 and obtained similar kind of general results for adjacency energies. Liu et al. [27] discussed distance and adjacency energies of multi-level wheel networks. Chu et al. [19] computed Laplacian and signless Laplacian spectra and energies of multi-step wheels.

    In 2015, Liu et al. [29] discussed asymptotic Laplacian energy like invariants of lattices. In 2016, Hosamani et al. [30] presented degree sum energy of a graph and obtained some lower bounds for this energy. In 2018, Basavanagoud et al. [31] computed the characteristic polynomial of the degree square sum matrix of graphs obtained by some graph operations as well as some bounds for spectral radius for square sum eigenvalue and degree square sum energy of graphs. In 2018, Rad et al. [32] presented Zagreb energy and related Estrada index of various graphs. In 2019, Gutman et al. [33] discussed graph energy and its applications, featuring about hundred kinds of graph energies and applications in diverse areas. For further details and basic ideas of graph energies, we refer [20,21,34,35,36]. Interconnection and various applications of graph energy in chemistry of unsaturated hydrocarbons can be traced in [37,38,39]. Applications of different graph energies in crystallography can be found in [40,41], theory of macro molecules in [42,43], protein sequences in [44,45,46], biology in [47], applied network analysis in [47,48,49,50,51,52], problems of air transportation in [48], satellite communications in [50] and constructions of spacecrafts in [52].

    In the present article we produce new results about maximum degree spectral radii and minimum degree spectral radii of m-splitting and m-shadow graphs. In fact we relate these spectral radii of new graph operations with spectral radii of original graphs. The article is organized as follows. Section 2 gives basic definitions and terminologies to lay foundations of our results. In Section 3, we derive maximum degree spectral radii and minimum degree spectral radii of generalized splitting graph constructed on any basic graph. In Section 4 we proceed to find similar results but for generalized shadow graph of the given regular graph.

    In this part we outline main ideas and preliminary facts, for details see [53,54]. The matrix M(G) is the maximum degree matrix of the graph G defined in [53] as

    M(G)={max(di,dj),ifviandvjareadjacent,0,elsewhere.}

    Here di and dj are the degrees of vertices vi and vj respectively. Eigenvalues of maximum degree matrix of the graph G are denoted as η1,η2,ηn. Maximum degree spectral radius is defined as

    M(G)=nmaxi=1|ηi|,

    where η1,η2,...,ηn are the eigenvalues of maximum degree matrix. The matrix MI(G) is called the minimum degree matrix of the graph G is defined in [54] as

    MI(G)={min(di,dj),ifviandvjareadjacent,0,elsewhere.}

    Minimum degree spectral radius is defined as

    MI(G)=nmaxi=1|ηi|,

    where η1,η2,...,ηn are the eigenvalues of minimum degree matrix. If we add a new vertex v to each vertex v of the graph G, v is connected to every vertex that is adjacent to v in G then we obtain the splitting graph (spl1(G)). Take two copies G and G of the graph G, then (sh2(G)) is constructed if we join each vertex in G to the neighbors of the corresponding vertices in G. Let UϵRm×n, VϵRp×q the tensor product (Kronecker product), UV is defined as the matrix.

    UV=(a11V...a1nV............an1V...annV)

    Proposition 1.1. Let UϵMm, VϵMn and α be an eigenvalue of U and η be an eigenvalue of V, then αη is an eigenvalue of UV [25]. Now we move towards the main results.

    In this part we relate maximum degree spectral radius and minimum degree spectral radius of generalized splitting graph with original graph G. Here again we emphasis that G is any regular graph.

    Theorem 1. Let G be any n-regular graph and M(Splm(G)) is themaximum degree spectral radius of m-splitting graph G, then

    M(Splm(G))=M(G)((m+1)(1+1+4m)2).

    Proof. Maximum degree matrix is given by M(G) where

    M(G)={max(di,dj),ifviandvjareadjacent,0,elsewhere.}

    M(Splm(G)) can be written in block matrix form as

    M(Splm(G))={(m+1)Mifi=1,j1andj=1,i1,0elsewhere.}
    =M{(m+1)ifi=1,j1andj=1,i1,0elsewhere.}

    Let

    A={(m+1)ifi=1,j1andj=1,i1,0elsewhere.}m+1

    Now we compute the eigenvalues of A Since matrix A is of rank two, so A has two non-zero eigenvalues, say α1 and α2. Obviously,

    α1+α2=tr(A)=m+1. (3.1)

    Considering

    A2={(m+1)3ifi=1andj=1,(m+1)2elsewhere.}m+1

    Then

    α21+α22=tr(A2)=(m+1)3+m((m+1)2). (3.2)

    Solving Eqs (3.1) and (3.2), we have

    α1=(m+1)(1+1+4m)2.

    and

    α2=(m+1)(11+4m)2.

    So we have,

    specA=(0(m+1)(1+1+4m)2(m+1)(11+4m)2m111) (3.3)

    Using Eq (3.3) we have

    specA=(0(m+1)(1+1+4m)2(m+1)(11+4m)2m111).

    Since M(Splm(G))=M(G)A then by Proposition 1.1, we have

    M(Splm(G))=Maxni=1|(specA))ηi|
    M(Splm(G))=Maxni=1|ηi|[(m+1)(1+1+4m)2]
    M(Splm(G))=M(G)((m+1)(1+1+4m)2).

    In the following corollaries, we obtain the maximum degree spectral radii of splitting graphs of Cn, Kn, Cn,n and crown graph.

    Corollary 2. If n3 and G is a Cn graph, where Cn is cycle graph on n vertices, then

    M(Splm(Cn))=4((m+1)(1+1+4m)2).

    Proof. If G is a cycle graph Cn(n3), then M(Cn)=4. Since cycle graph is 2-regular graph so using Theorem 1 we get the required result.

    Corollary 3. If G is a Kn graph, where Kn is complete graph on n vertices, then

    M(Splm(Kn))=(n1)2((m+1)(1+1+4m)2).

    Proof. If G is a complete graph on n vertices, then M(Kn)=(n1)2. Since complete graph is n1-regular graph so using Theorem 1 we get the required result.

    Corollary 4. If G is a complete bipartite graph Kn,n, then

    M(Splm(Kn,n))=(n)2((m+1)(1+1+4m)2).

    Proof. If G is a complete bipartite graph Kn,n, then M(Kn,n)=(n)2. Using Theorem 1 we get the required result.

    Corollary 5. If G is a crown graph on 2n vertices, then

    M(Splm(G))=(n1)2((m+1)(1+1+4m)2).

    Proof. If G is a crown graph on 2n vertices, then M(G)=(n1)2. Using Theorem 1 we get the required result.

    Theorem 6. Let G be any n-regular graph and MI((Splm(G)) is theminimum degree spectral radius of m-splitting graph G, then

    MI(Splm(G))=MI(G)(m+1+m2+6m+12).

    Proof. Minimum degree matrix is given by

    MI(G)={min(di,dj),ifviandvjareadjacent,0,elsewhere.}

    MI(Splm(G)) can be written in block matrix form as follows:

    MI(Splm(G))={(m+1)MIifi=1andj=1MIifi=1,j2andj=1,i2,0elsewhere.}
    =MI{(m+1)ifi=1andj=11ifi=1,j2andj=1,i2,0elsewhere.}

    Let

    A={(m+1)ifi=1andj=11ifi=1,j2andj=1,i2,0elsewhere.}m+1

    Now we compute the eigenvalues of A Since matrix A is of rank two, so A has two non-zero eigenvalues, say α1 and α2. Obviously,

    α1+α2=tr(A)=m+1. (3.4)

    Considering

    A2={(m+1)2+mifi=1andj=1,m+1ifi=1,j2andj=1,i2,1elsewhere.}m+1

    Then

    α21+α22=tr(A2)=(m+1)2+2m. (3.5)

    Solving Eqs (3.4) and (3.5) we have

    α1=m+1+m2+6m+12.
    α2=m+1m2+6m+12.

    So we have,

    specA=(0m+1+m2+6m+12m+1m2+6m+12.m111) (3.6)

    Using Eq (3.6) we have

    specA=(0m+1+m2+6m+12m+1m2+6m+12.m111).

    Since MI(Splm(G))=MI(G)A then by Proposition 1.1, we have

    MI(Splm(G))=Maxni=1|(specA))ηi|
    MI(Splm(G))=Maxni=1|ηi|[m+1+m2+6m+12]
    MI(Splm(G))=MI(G)(m+1+m2+6m+12).

    In the following corollaries, we obtain the minimum degree spectral radii of splitting graphs of Cn, Kn, Cn,n and crown graph.

    Corollary 7. If n3 and G is a Cn graph, where Cn is cycle graph on n vertices, then

    MI(Splm(Cn))=4(m+1+m2+6m+12).

    Proof. If G is a cycle graph Cn(n3), then MI(Cn)=4. Since cycle graph is 2-regular graph so using Theorem 6 we get the required result.

    Corollary 8. If G is a Kn graph, where Kn is complete graph on n vertices, then

    MI(Splm(Kn))=(n1)2(m+1+m2+6m+12).

    Proof. If G is a complete graph on n vertices, then MI(Kn)=(n1)2. Since complete graph is n1-regular graph so using Theorem 6 we get the required result.

    Corollary 9. If G is a complete bipartite graph Kn,n, then

    MI(Splm(Kn,n))=(n)2(m+1+m2+6m+12).

    Proof. If G is a complete bipartite graph Kn,n, then MI(Kn,n)=(n)2. Using Theorem 6 we get the required result.

    Corollary 10. If G is a crown graph on 2n vertices, then

    MI(Splm(G))=(n1)2(m+1+m2+6m+12).

    Proof. If G is a crown graph on 2n vertices, then MI(G)=(n1)2. Using Theorem 6 we get the required result.

    In this part we relate maximum degree spectral radius and minimum degree spectral radius of generalized Shadow graph with original graph G. Here again we emphasis that G is any regular graph.

    Theorem 11. Let G be any n-regular graph and M(Shm(G)) is themaximum degree spectral radius of m-shadow graph G, then

    M(Shm(G))=M(G)((m)2).

    Proof. Maximum degree matrix is given by

    M(G)={max(di,dj),ifviandvjareadjacent,0,elsewhere.}

    Then M(Shm(G)) can be written in block matrix form as follows:

    M(Shm(G))={(m)Miandj.}
    =M{miandj.}

    Let

    A={miandj.}m

    Now we compute the eigenvalues of A Since matrix A is of rank one, so A has one non-zero eigenvalue, say α1=(m)2. So we have,

    specA=(0(m)2m11) (4.1)

    Using Eq (4.1) we have

    specA=(0(m)2m11).

    Since M(Shm(G))=M(G)A then by Proposition 1.1, we have

    M(Shm(G))=Maxni=1|(specA))ηi|
    M(Shm(G))=Maxni=1|ηi|[(m)2]
    M(Shm(G))=M(G)((m)2).

    In the following corollaries, we obtain the maximum degree spectral radii of shadow graphs of Cn, Kn, Cn,n and crown graph.

    Corollary 12. If n3 and G is a Cn graph, where Cn is cycle graph on n vertices, then

    M(Shm(Cn))=4((m)2).

    Proof. If G is a cycle graph Cn(n3), then M(Cn)=4. Since cycle graph is 2-regular graph so using Theorem 11 we get the required result.

    Corollary 13. If G is a Kn graph, where Kn is complete graph on n vertices, then

    M(Shm(Kn))=(n1)2((m)2).

    Proof. If G is a complete graph on n vertices, then M(Kn)=(n1)2. Since complete graph is n1-regular graph so using Theorem 11, we get the required result.

    Corollary 14. If G is a complete bipartite graph Kn,n, then

    M(Shm(Kn,n))=(n)2((m)2).

    Proof. If G is a complete bipartite graph Kn,n, then M(Kn,n)=(n)2. Using Theorem 11, we get the required result.

    Corollary 15. If G is a crown graph on 2n vertices, then

    M(Shm(G))=(n1)2((m)2).

    Proof. If G is a crown graph on 2n vertices, then M(G)=(n1)2. Using Theorem 11, we get the required result.

    Theorem 16. Let G be any n-regular graph and MI(Shm(G)) is theminimum degree spectral radius of m-shadow graph G, then

    MI(Shm(G))=(m)2MI(G).

    Proof. Minimum degree matrix is given by

    MI={min(di,dj),ifviandvjareadjacent,0,elsewhere.}

    Then MI(Shm(G)) can be written in block matrix form as follows:

    MI(Shm(G))={(m)MIiandj.}
    =MI{miandj.}

    Let

    A={miandj.}m

    Now we compute the eigenvalues of A Since matrix A is of rank one, so A has one non-zero eigenvalue, say α1=(m)2. So we have,

    specA=(0(m)2m11) (4.2)

    Using Eq (4.2) we have

    specA=(0(m)2m11).

    Since MI(Shm(G))=MI(G)A then by Proposition 1.1, we have

    MI(Shm(G))=Maxni=1|(specA))ηi|
    MI(Shm(G))=Maxni=1|ηi|[(m)2]
    MI(Shm(G))=MI(G)((m)2).

    In the following corollaries, we obtain the minimum degree spectral radii of shadow graphs of Cn, Kn, Cn,n and crown graph.

    Corollary 17. If n3 and G is a Cn graph, where Cn is cycle graph on n vertices, then

    MI(Shm(Cn))=4((m)2).

    Proof. If G is a cycle graph Cn(n3), then MI(Cn)=4. Since cycle graph is 2-regular graph so using Theorem 16 we get the required result.

    Corollary 18. If G is a Kn graph, where Kn is complete graph on n vertices, then

    MI(Shm(Kn))=(n1)2((m)2).

    Proof. If G is a complete graph on n vertices, then MI(Kn)=(n1)2. Since complete graph is n1-regular graph so using Theorem 16 we get the required result.

    Corollary 19. If G is a complete bipartite graph Kn,n, then

    MI(Shm(Kn,n))=(n)2((m)2).

    Proof. If G is a complete bipartite graph Kn,n, then MI(Kn,n)=(n)2. Using Theorem 16, we get the required result.

    Corollary 20. If G is a crown graph on 2n vertices, then

    MI(Shm(G))=(n1)2((m)2).

    Proof. If G is a crown graph on 2n vertices, then MI(G)=(n1)2. Using Theorem 16 we get the required result.

    The spectral radius of graph has vast range of applications in computer related areas. It also connects graph theory and chemistry. In this article we have related the spectral radii of the generalized shadow and splitting graph of any regular graph with spectral radius of the given graph. In particular we have proved that the Spectral Radius of the new graph is a multiple of spectral radius of the given regular graph.

    The authors thank the National Key Research and Development Program under Grant 2018YFB0904205, Science and Technology Bureau of ChengDu (2020-YF09-00005-SN), Sichuan Science and Technology program (2021YFH0107) and Erasmus + SHYFTE Project (598649-EPP-1-2018-1-FR-EPPKA2-CBHE-JP).

    The authors declare that there is no conflict of interest.



    [1] D. Cucinotta, M.Vanelli, WHO declares COVID-19 a pandemic, Acta. Biomed., 91 (2020), 157–160. https://doi.org/10.23750/abm.v91i1.9397 doi: 10.23750/abm.v91i1.9397
    [2] T. Fisayo, S. Tsukagoshi, Three waves of the COVID-19 pandemic, Postgrad. Med. J., 97 (2021), 332. https://doi.org/10.1016/j.ijsu.2020.04.018 doi: 10.1016/j.ijsu.2020.04.018
    [3] M. Nicola, Z. Alsafi, C. Sohrabi, A. Kerwan, A. Al-Jabir, C. Iosifidis, et al., The socio-economic implications of the coronavirus pandemic (COVID-19): A review, Int. J. Surg., 78 (2020), 185–193. https://doi.org/10.1016/j.ijsu.2020.04.018 doi: 10.1016/j.ijsu.2020.04.018
    [4] Z. Zheng, Z. Xie, Y. Qin, Exploring the influence of human mobility factors and spread prediction on early COVID-19 in the USA, BMC Public Health, 21 (2021). https://doi.org/10.1186/s12889-021-10682-3 doi: 10.1186/s12889-021-10682-3
    [5] J. Konecny, H. B. McMahan, F. X. Yu, P. Richtárik, A. T. Suresh, D. Bacon, Federated learning: Strategies for improving communication efficiency, in 29th Conference on Neural Information Processing Systems, 2016.
    [6] K. Bonawitz, H. Eichner, W. Grieskamp, D. Huba, A. Ingerman, V. Ivanov, et al., Towards federated learning at scale: System design, Proc. Mach. Learn. Syst., 1 (2019), 374–388.
    [7] S. Azam, J. E. Macías-Díaz, N. Ahmed, I. Khan, M. S. Iqbal, Numerical modeling and theoretical analysis of a nonlinear advection-reaction epidemic system, Comput. Methods Programs Biomed., 193 (2020), 67–83. https://doi.org/10.1016/j.cmpb.2020.105429 doi: 10.1016/j.cmpb.2020.105429
    [8] M. Rafiq, J. E. Macías-Díaz, A. Raza, N. Ahmed, Design of a nonlinear model for the propagation of COVID-19 and its efficient nonstandard computational implementation, Appl. Mathl. Model., 89 (2021), 1835–1846. https://doi.org/10.1016/j.apm.2020.08.082 doi: 10.1016/j.apm.2020.08.082
    [9] J. E. Macías-Díaz, N. Ahmed, M. Rafiq, Analysis and nonstandard numerical design of a discrete three-dimensional hepatitis B epidemic model, Mathematics, 7 (2019). https://doi.org/10.3390/math7121157 doi: 10.3390/math7121157
    [10] J. Guan, Y. Wei, Y. Zhao, F. Chen, Modeling the transmission dynamics of COVID-19 epidemic: a systematic review, J. Biomed. Res., 34 (2020), 422–430. https://doi.org/10.7555/JBR.34.20200119 doi: 10.7555/JBR.34.20200119
    [11] R. ud Din, A. R. Seadawy, K. Shah, A. Ullah, D. Baleanu, Study of global dynamics of COVID-19 via a new mathematical model, Results Phys., 19 (2020). https://doi.org/10.1016/j.rinp.2020.103468 doi: 10.1016/j.rinp.2020.103468
    [12] I. F. Mello, L. Squillante, G. O. Gomes, A. C. Seridonio, M. de Souza, Epidemics, theIsing-model and percolation theory: A comprehensive review focused on Covid-19, Phys. A: Statist. Mech. Appl., 573 (2021). https://doi.org/10.1016/j.physa.2021.125963 doi: 10.1016/j.physa.2021.125963
    [13] H. A. Adekola, I. A. Adekunle, H. O. Egberongbe, S. A. Onitilo, I. N. Abdullahi, Mathematical modeling for infectious viral disease: The COVID-19 perspective, J. Public Aff., 20 (2020). https://doi.org/10.1002/pa.2306 doi: 10.1002/pa.2306
    [14] F. Özköse, M. Yavuz, M. T. Şenel, R. Habbireeh, Fractional order modelling of omicron SARS-CoV-2 variant containing heart attack effect using real data from the United Kingdom, Chaos, Solitons Fractals, 157 (2022), 111954. https://doi.org/10.1016/j.chaos.2022.111954 doi: 10.1016/j.chaos.2022.111954
    [15] P. A. Naik, M. Yavuz, S. Qureshi, Modeling and analysis of COVID-19 epidemics with treatment in fractional derivatives using real data from Pakistan, Eur. Phys. J. Plus, 135 (2020), 795. https://doi.org/10.1140/epjp/s13360-020-00819-5 doi: 10.1140/epjp/s13360-020-00819-5
    [16] B. Dasbasi, Stability analysis of an incommensurate fractional-order SIR model, Math. Modell. Numer. Simul. Appl., 1 (2021). https://doi.org/10.53391/mmnsa.2021.01.005 doi: 10.53391/mmnsa.2021.01.005
    [17] P. Kumar, V. S. Erturk, Dynamics of cholera disease by using two recent fractional numerical methods, Math. Modell. Numer. Simul. Appl., 1 (2021), 102–111. https://doi.org/10.53391/mmnsa.2021.01.010 doi: 10.53391/mmnsa.2021.01.010
    [18] F. Özköse, M. Yavuz, Investigation of interactions between COVID-19 and diabetes with hereditary traits using real data: A case study in Turkey, Comput. Biol. Med., 141 (2022). https://doi.org/10.1016/j.compbiomed.2021.105044 doi: 10.1016/j.compbiomed.2021.105044
    [19] H. Joshi, B. K. Jha, Chaos of calcium diffusion in Parkinson's infectious disease model and treatment mechanism via Hilfer fractional derivative, Math. Modell. Numer. Simul. Appl., 1 (2021), 84–94. https://doi.org/10.53391/mmnsa.2021.01.008 doi: 10.53391/mmnsa.2021.01.008
    [20] F. Özköse, M. T. Şenel, R. Habbireeh, Fractional-order mathematical modelling of cancer cells-cancer stem cells-immune system interaction with chemotherapy, Math. Modell. Numer. Simul. Appl., 1 (2021), 67–83. https://doi.org/10.53391/mmnsa.2021.01.007 doi: 10.53391/mmnsa.2021.01.007
    [21] M. Yavuz, F. Ö. Coşar, F. Günay, F. N. Özdemir, A new mathematical modeling of the COVID-19 pandemic including the vaccination campaign, Open J. Modell. Simul., 9 (2021), 299–321. https://doi.org/10.4236/ojmsi.2021.93020 doi: 10.4236/ojmsi.2021.93020
    [22] S. Allegretti, I. M. Bulai, R. Marino, M. A. Menandro, K. Parisi, Vaccination effect conjoint to fraction of avoided contacts for a Sars-Cov-2 mathematical model, Math. Modell. Numer. Simul. Appl., 1 (2021), 56–66. https://doi.org/10.53391/mmnsa.2021.01.006 doi: 10.53391/mmnsa.2021.01.006
    [23] R. Ikram, A. Khan, M. Zahri, A. Saeed, M. Yavuz, P. Kumam, Extinction and stationary distribution of a stochastic COVID-19 epidemic model with time-delay, Comput. Biol. Med., 141 (2022). https://doi.org/10.1016/j.compbiomed.2021.105115 doi: 10.1016/j.compbiomed.2021.105115
    [24] M. A. Achterberg, B. Prasse, L. Ma, S. Trajanovski, M. Kitsak, P. V. Mieghem, Comparing the accuracy of several network-based COVID-19 prediction algorithms, Int. J. Forecast., 38 (2022), 489–504. https://doi.org/10.1016/j.ijforecast.2020.10.001 doi: 10.1016/j.ijforecast.2020.10.001
    [25] E. Hernández-Pereira, O. Fontenla-Romero, V. Bolón-Canedo, Machine learning techniques to predict different levels of hospital care of CoVid-19, Appl. Intell., 52 (2022), 6413–6431. https://doi.org/10.1007/s10489-021-02743-2 doi: 10.1007/s10489-021-02743-2
    [26] J. Köhler, L. Schwenkel, A. Koch, J. Berberich, P. Pauli, F. Allgöwer, Robust and optimal predictive control of the COVID-19 outbreak, Ann. Rev. Control, 51 (2021), 525–539. https://doi.org/10.1016/j.arcontrol.2020.11.002 doi: 10.1016/j.arcontrol.2020.11.002
    [27] M. M. Morato, S. B. Bastos, D. O. Cajueiro, J. E. Normey-Rico, An optimal predictive control strategy for COVID-19 (SARS-CoV-2) social distancing policies in Brazil, Ann. Rev. Control, 50 (2020), 417–431. https://doi.org/10.1016/j.arcontrol.2020.07.001 doi: 10.1016/j.arcontrol.2020.07.001
    [28] X. Yan, Y. Zou, Optimal and sub-optimal quarantine and isolation control in SARS epidemics, Math. Comput. Model., 47 (2008), 235–245. https://doi.org/10.1016/j.mcm.2007.04.003 doi: 10.1016/j.mcm.2007.04.003
    [29] M. Khouzani, S. S. Venkatesh, S. Sarkar, Market-based control of epidemics, in 2011 49th Annual Allerton Conference on Communication, Control, and Computing (Allerton), (2011), 314–320. https://doi.org/10.1109/Allerton.2011.6120184
    [30] A. Argyriou, T. Evgeniou, M. Pontil, Multi-task feature learning, In NIPS '06, 19 (2006).
    [31] M. Hayhoe, F. Barreras, V. M. Preciado, Multitask learning and nonlinear optimal control of the COVID-19 outbreak: A geometric programming approach, Ann. Rev. Control, 52 (2021), 495–507, https://doi.org/10.1016/j.arcontrol.2021.04.014 doi: 10.1016/j.arcontrol.2021.04.014
    [32] T. Li, A. K. Sahu, A. Talwalkar, V. Smith, Federated learning: Challenges, methods, and future directions, IEEE Sig. Process Mag., 37 (2020), 50–60. https://doi.org/10.1109/MSP.2020.2975749 doi: 10.1109/MSP.2020.2975749
    [33] S. Caldas, S. M. K. Duddu, P. Wu, T. Li, J. Konečný, H. B. McMahan, et al., Leaf: A benchmark for federated settings, 2018. https://doi.org/10.48550/arXiv.1812.01097
    [34] T. Li, A. K. Sahu, M. Zaheer, M. Sanjabi, A. Talwalkar, Federated optimization in heterogeneous networks, Proc. Mach. Learn. Syst., 2 (2020), 429–450.
    [35] S. M. Abdul, S. Taha, M. Ramadan, COVID-19 detection using federated machine learning, PLoS One, 16 (2021). https://doi.org/10.1371/journal.pone.0252573 doi: 10.1371/journal.pone.0252573
    [36] A. Vaid, S. K. Jaladanki, J. Xu, S. Teng, Federated learning of electronic health records to improve mortality prediction in hospitalized patients with COVID-19: Machine learning approach, JMIR Med. Inf., 9 (2021). https://doi.org/10.2196/24207 doi: 10.2196/24207
    [37] M. U. Alam, R. Rahmani, Federated semi-supervised multi-task learning to detect COVID-19 and lungs segmentation marking using chest radiography images and raspberry pi devices: An internet of medical things application, Sensors, 21 (2021). https://doi.org/10.3390/s21155025 doi: 10.3390/s21155025
    [38] V. Smith, C. K. Chiang, M. Sanjabi, A. Talwalkar, Federated multi-task learning, Adv. Neural Inf. Proc. Syst., 30 (2017), 4424–4434,
    [39] A. Argyriou, T. Evgeniou, M. Pontil, A. Argyriou, T. Evgeniou, M. Pontil, Convex multi-task feature learning, Mach. Learn., 73 (2008), 243–272. https://doi.org/10.1007/s10994-007-5040-8 doi: 10.1007/s10994-007-5040-8
    [40] A. Argyriou, C. A. Micchelli, M. Pontil, Y. Ying, A spectral regularization framework for multi-task structure learning, in International Conference on Neural Information, 20 (2007).
    [41] W. O. Kermack, A. G. Mckendrick, Contribution to the mathematical theory of epidemics, Proc. R. Soc. London Ser A, 115 (1927), 700–721. https://doi.org/10.1098/rspa.1927.0118 doi: 10.1098/rspa.1927.0118
    [42] P. van den Driessche, Reproduction numbers of infectious disease models, Infect. Dis. Modell., 2 (2017), 288–303. https://doi.org/10.1016/j.idm.2017.06.002 doi: 10.1016/j.idm.2017.06.002
    [43] See How Your Community is Moving Around Differently due to COVID-19, Google, 2021. Available from: https://www.google.com/covid19/mobility/index.htmlhl=en.
    [44] COVID-19 Data in the United States, Github, 2021. Available from: https://github.com/nytimes/covid-19-data.
    [45] National Population Totals: 2010–2020, US Census Bureau, 2020. Available from: https://www.census.gov/programs-surveys/popest/technical-documentation/research/evaluation-estimates/2020-evaluation-estimates/2010s-totals-national.html.
    [46] COVID-19 Hospital Data, HealthData.gov, 2022.
  • This article has been cited by:

    1. Ahmad Bilal, Mobeen Munir, Randic and reciprocal randic spectral radii and energies of some graph operations, 2022, 10641246, 1, 10.3233/JIFS-221938
    2. Ahmad Bilal, Muhammad Mobeen Munir, ABC energies and spectral radii of some graph operations, 2022, 10, 2296-424X, 10.3389/fphy.2022.1053038
    3. Ahmad Bilal, Muhammad Mobeen Munir, Muhammad Imran Qureshi, Muhammad Athar, ISI spectral radii and ISI energies of graph operations, 2023, 11, 2296-424X, 10.3389/fphy.2023.1149006
    4. Linlin Wang, Sujuan Liu, Han Jiang, Rainbow connections of bioriented graphs, 2024, 10, 24058440, e31426, 10.1016/j.heliyon.2024.e31426
    5. A. R. Nagalakshmi, A. S. Shrikanth, G. K. Kalavathi, K. S. Sreekeshava, The Degree Energy of a Graph, 2024, 12, 2227-7390, 2699, 10.3390/math12172699
    6. Muhammad Mobeen Munir, Urwah Tul Wusqa, Albertson (Alb) spectral radii and Albertson (Alb) energies of graph operation, 2023, 11, 2296-2646, 10.3389/fchem.2023.1267291
    7. Caicai Feng, Muhammad Farhan Hanif, Muhammad Kamran Siddiqui, Mazhar Hussain, Nazir Hussain, On analysis of entropy measure via logarithmic regression model for 2D-honeycomb networks, 2023, 138, 2190-5444, 10.1140/epjp/s13360-023-04547-4
    8. Ahmad Bilal, Muhammad Mobeen Munir, SDD Spectral Radii and SDD Energies of Graph Operations, 2024, 1007, 03043975, 114651, 10.1016/j.tcs.2024.114651
    9. Xueliang Li, Ruiling Zheng, Extremal results on the spectral radius of function-weighted adjacency matrices, 2025, 373, 0166218X, 204, 10.1016/j.dam.2025.04.060
  • Reader Comments
  • © 2022 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(2985) PDF downloads(135) Cited by(6)

Figures and Tables

Figures(18)  /  Tables(6)

Other Articles By Authors

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog