An Algorithm for Weighted Positive Influence Dominating Set Based on Learning Automata
|
|
- Rodger Walker
- 5 years ago
- Views:
Transcription
1 4 th Internatonal Conference on Knowledge-Based Engneerng and Innovaton (KBEI-2017) Dec. 22 th, 2017 (Iran Unversty of Scence and Technology) Tehran, Iran An Algorthm for Weghted Postve Influence Domnatng Set Based on Learnng Automata Mohammad Mehd Dalr Khomam Soft computng laboratory, Department of Computer Engneerng and Informaton Technology Amrkabr Unversty of Technology (Tehran Polytechnc), Tehran, Iran Maryam Amr Haer Department of Computer Engneerng & IT, Amrkabr Unversty of Technology (Tehran Polytechnc) Tehran, Iran Mohammad Reza Meybod Soft computng laboratory, Department of Computer Engneerng and Informaton Technology, Amrkabr Unversty of Technology (Tehran Polytechnc), Tehran, Iran Al Mohammad Saghr Soft computng laboratory, Department of Computer Engneerng and Informaton Technology, Amrkabr Unversty of Technology (Tehran Polytechnc), Tehran, Iran Abstract The problem of Influence Maxmzaton (IM) n socal network refers to determnng a set of nodes that can maxmze the spread of nfluence. IM problem has been appled n many domans such as marketng, advertsng and publc opnon montorng. In recent years, dfferent type of algorthms reported n the lterature. A type of algorthms for ths problem s based on domnatng set. In these algorthms, the IM problem s consdered as a verson of domnatng set problem. A lmtaton of most of these algorthms, the graph of the socal network s assumed to be a un-weghted graph whch s not a realstc assumpton. The other drawback of these algorthms s that they generally omt the crucal characterstcs of the socal networks such as dfferent level of nfluence and dynamc nteracton among ndvduals. In order to solve these problems and based on the realstc nature of the socal network, t seems that the Mnmum Weghted Postve Influence Domnatng Set (MWPIDS) problem can support the mentoned characterstcs because t consder the weght of the graph. In ths paper, a learnng automaton based algorthm s proposed to reveal MWPIDS n the socal network graphs. In the proposed algorthm, each vertex of the socal network graph s equpped wth a learnng automaton that determnes the beeng canddate or non-canddate of the correspondng vertex to be n WPIDS or not. Owng to adaptve decson makng characterstcs of learnng automata, the proposed algorthm sgnfcantly reduces the number of canddate soluton. The proposed algorthm, based on learnng automata, teratvely decreases the weght of the obtaned postve nfluence domnatng. In order to evaluate the proposed algorthm, several experments have been conducted on real socal network datasets whch compared to the state-of-the art methods. Expermental results show the superorty of the proposed algorthm over the prevous algorthms. Keywords Learnng Automata; Socal Network Analyss; Domnatng Set; Weghted Postve Influence Domnatng Set I. INTRODUCTION Due to the popularty of onlne socal networks such as Facebook, Myspace and Lnkdn provde companes new settng for enablng large-scale markng through the powerful word-of-mouth effect. Recently, extensve researches around socal network analyss have been nvestgated. They have shown that onlne users tend to share a new nformaton or trust to a news whch s ganed from ther close socal groups such as frends and acquantances nstead of trust to comments of ordnary people or news from publc meda. Many researchers on socal network analyss have been focused on sgnfcant characterstc of socal networks. In most of the studes, a socal network can usually be represented as a network wth a set of nodes corresponds to onlne users and edges representng a relatonshp between users. One of the sgnfcant characterstc of socal networks s the vral markng that ndcates the tendency of ganng largest nfluence on people through adoptng the product. In [1] s shown that the vral markng s effcent than other marketng polces. From the algorthmc aspect the term of vral marketng s transformed nto the Influence Maxmzaton (IM) that s known as an optmzaton problem. The man goal of the nfluence maxmzaton n a socal network s to fnd small k vertces (referred to as ntal seeds) n the networks, such that under the certan dffuson model, the expected number of vertces nfluenced by k s maxmzed. In ths paper, we propose a learnng automata based algorthm for solvng Mnmum Weghted Postve Influence Domnatng Set (MWPIDS) consderng the characterstcs n the socal networks. On the other hand, the theory of learnng automata has been shown ts potental as an adaptve decson makng unt n the dynamc and complex envronment such as sensor networks[2], peer to peer networks[3] and socal networks[4]. 1 Unversal Scentfc Organzaton KBEI-2016
2 The rest of the paper s organzed as follows: In secton 2 revews related work. In secton 3 the theory of learnng automata s ntroduced brefly. In secton 4 the proposed learnng automata s descrbed. The performance of the proposed algorthm s evaluated through the smulaton experments n secton 5. We make concluson and future works n secton 6. II. RELATED WORKS Clark et al. [5] proved that fndng the Mnmum Domnatng Set and ts varatons are NP-hard and proved exponental tme complexty. Several real practcal applcatons of ths problem are ntroduced n socal networks. For example, n terms of nfluence maxmzaton due to the polcy of product adopton, one person can change the decson by acceptng or refusng an offer. There are many researchs have been focused on Postve Influence Domnatng Set (PIDS), however, there are only a lttle works done on Weghted Postve Influence Domnatng Set. In the followng, we wll revewed the outcomes of these studes n a nutshell: The mlestone work for mnmum postve nfluence domnatng set problem s presented by Wang et al. [6]. They proposed an approxmate greedy algorthm and proved that the problem s APX-hard. Moreover, the author s studed the PIDS(Postve Influence Domnatng Set) problem under the determnstc lnear threshold model, n whch the nfluence from a par of nodes s computed by total ncomng nfluence from ts neghbors exceeds a pre-determned threshold. Recently, Rae et al. [7] proposed a new greedy algorthm whch mproves the tme complexty of the prmtve algorthm for PIDS. Although the greedy algorthm acheves many promsng results, but n some cases t s lmted to fnd proper approxmaton and produce unfavorable results. Zhang et al[8] studed PIDS n power-law graph and proved the greedy algorthm wth constant factor. Dneh et al. [9] studed the PIDS from dfferent ponts, and proposed the Total Postve Influence Domnatng Set (TPIDS) nstead of PIDS and consdered fracton of neghbors nstead of half. In addton, they proved an approxmaton factor for TPIDS and showed that n power-law networks whch follow power-law degree dstrbuton, both PIDS and TPIDS admt constant factor approxmaton. Basuchowdhur et al. [10] nvestgated a more generc problem whch s smlar to fndng mnmum domnatng set, but t s dfferent from the nave algorthm n terms of the number of hops needed to reach all nodes n the network and known as Mnmum k- Hop Domnatng Set. Intutvely, by ncreasng the parameter k correspondng to the number of hops, t ncreases the number of nodes whch s trggered by the seeds. In [11] nvestgated partal postve nfluence domnatng set and appled general threshold model for all nodes n networks. Threshold functon s determned as a value of certan fracton of neghbors whch are n actve state. A remarkable applcaton of PIDS s maxmzng the nfluence among partcpant n networks. Gven a socal network, the problem of nfluence maxmzaton refers to select a set of nodes such that the spread of nfluence based on a gven model s maxmzed. In order to modelng the mcro behavors of the ndvduals when they are exposed to specfc nformaton n networks, spreadng models have been proposed. Two well-known models ncludng ndependent cascadng model (IC) and lnear threshold model (LT) are presented n [12]. Based on spreadng models, Kempe et al. [13] nvestgated the nfluence maxmzaton problem n a socal network for fndng a set of ntal nodes that the spread of nformaton s maxmzed. We Chen et al. [14] nvestgate the IM, and modfed the algorthms to apply n the context of large-scale data mnng n real socal networks. Chen et al. [15] proposed a heurstc based algorthm for nfluence maxmzaton n arborescence whch s called (MIA) to change the scalablty of the algorthm usng the maxmum nfluence path of each par of nodes. Km et al. [16] proposed scalable and parallelzable nfluence approxmaton algorthm by calculatng ndependent paths n networks. Yang et al. [17] extended the nfluence maxmzaton problem and proposed a coordnate descent method to solve the transmsson cost problem. Goyal et al. [18]used the smple path between neghbor nodes to estmate the nfluence spread n networks. Due to the dynamc and complex nature of socal networks, all works whch were revewed n ths paper suffer from two major challenges that descrbed as below. Postve and negatve nteractons through communcaton should be fully captured. Note that varyng the behavor of the users reflected by postve and negatve nteractons affect the decsons of the algorthm whch analyses the socal network. Snce the propertes of users are dfferent from each other, they reflect dfferent levels of nfluence durng the communcaton whch IM gnores them. None of exstng algorthms reported for PIDS and ts varatons consders the menton characterstcs. Our man contrbutons are lsted as follows: We use a realstc model whch s superor to model ude by the pror Influence Maxmzaton algorthm. We desgn an effcent learnng automata based algorthm to solve WPIDS n socal networks. Our proposed algorthm s the frst learnng based algorthm for solvng the WPIDS We expermentally proved that the obtaned results are comparable and compettve wth exstng stateof-the art algorthms. III. STATEMENT OF THE PROBLEM The socal networks can be represented by a graph G ( V, E, W ) where V V, V,..., V } { 1 2 n s a set of nodes, 2
3 E V V s a set of edge and W ndcates the weght functon such that F : v. Mnmum Weghted Postve Influence Domnatng Set (MWPIDS) s a subset of S V such that each node v V s domnated by at least n 2 w v nodes n S wth mnmum weght where n s the degree of node v. The topology of the underlay learnng automata network ' ' ' can be represent by a graph G ( V, E ) n whch ' V { 2 LA1, LA,..., LA n } E LA ' s V underlyng network and connectng the networks, each node s a set of learnng automata n ' ' ' V V s a set of lnks n the underlay network. In socal s mapped to an LA ' n learnng automata network based on one to one functon ' H : V V. A prmary goal of MWPIDS s to learn the select optmal acton among a lst of canddate ones n whch s guded toward the optmal soluton. Fg 1 demonstrates a snapshot of the graph of the socal network. Node w Node Nodej Node k Node n Node Node l m Fg. 1. A Snapshot of a graph of the socal network IV. LEARNING AUTOMATA Learnng automaton (LA) [12, 14] s an abstract model of fnte state machne so that t can perform fnte actons whch randomly chosen and evaluated by a stochastc envronment and response s gven as a reward or penalty to LA. LA uses feedback of envronment and select the next acton. Durng ths process, LA learns how to choose the best acton from a set of allowed actons. Fgure 2 llustrates the relatonshp between the LA and the t s random envronment. LA wth varable structure can be descrbed wth quadruple denotes the fnte set {,, P, T} where { 1,..., r } of actons and,..., } { 1 m s the set of nput values that can be taken by renforcement sgnal and P { p 1,..., p r } P( n 1) T[ ( n), ( n), p( n)} s the probablty vector of each acton. s learnng algorthm where s updated n each teraton. Equatons (1) and (2) llustrate changes n probablty vector accordng to performance evaluaton of α n each step. Automaton updates ts acton probablty set based on equaton (1) for favorable responses as: p ( n 1) p ( n) a[1 p ( n)] (1) p ( n 1) (1 a) p ( n) j j j, And equaton (2) for unfavorable ones: p ( n 1) (1 a) p ( n) b p j ( n 1) ( ) (1 b) p j ( n) r 1 P(n j j, j Where ) s the acton probablty vector at nstant n. r s the number of actons that can be taken by the LA. The learnng rates a and b denote the reward and penalty parameters and determne the amount of ncreases and decreases of the acton probabltes, respectvely. If a=b L learnng algorthm s called lnear reward penalty ( R P ), f b<<a gven learnng algorthm s called lnear reward-ε penalty ( ), and fnally, f b=0 t s called lnear reward LR p nacton ( L ) [24-26]. R I V. THE PROPOSED ALGORITHM In ths secton, a learnng automata based algorthm for fndng MWPIDS problem wll be proposed. In ths algorthm, each node of the socal network s equpped wth a learnng automaton whch consst of two actons canddate and non-canddate. The network of the learnng automata resded n the nodes of the socal network has been utlzed to fnd a proper WPIDS. For the sake of more clarfcaton about the relatonshp between learnng automata and the graph of the WPIDS Fgure 2 s gven. The detaled descrptons of the proposed algorthm are gven n the rest of ths secton. (2) Random Envronment Acton Response Learnng Automata Fg.2. The relatonshp between the LA and ts random envronment 3
4 Node w Node n LA o LA n Node m Node Node l LA Nodej LA j Node k LA k Mean probablty nsoluton 1 n Max( P ) soluton Where Max P ) corresponds to the maxmum probablty ( of learnng Automata and n soluton (3) demonstrated as the number of canddate LA whch s present n soluton. The general structure of proposed algorthm s shown n the followng dagram as Fgure 3. Thus LA learns how to choose the best acton wth for WPIDS n successve teratons. Ths process wll be contnued untl the algorthm reaches to a near optmal soluton. In the next secton the experment smulatons are demonstrated n order to evaluate the proposed algorthm. LA m LA l Fg2. A snapshot of mappng the network of the learnng automata The proposed algorthm conssts of four steps: 1. Intalzaton, 2. Actvaton and acton selecton, 3. Verfcaton and calculatng renforcement sgnal and, 4. Updatng probablty vector. These steps are descrbed as follows: In the ntalzaton, each node n the networks s equpped wth learnng automata that has two actons canddate and non-canddate and ntally are dsabled. At frst all learnng automata are actvated and choosng ts acton randomly. Because of probabltes of all acton n frst teraton are equal, hence each acton s selected wth equal probablty. After an acton selecton step s done the verfcaton and calculatng renforcement sgnal s stars. In ths step, the set of learnng automata whch selects the canddate actons are checked n terms of correctness, whch means that the obtaned set s WPIDS or not. Then ther weghts are calculated. If the weght of WPIDS s less than the former WPIDS then the selected acton corresponds to current learnng automata whch s presented n the set are rewarded else the selects acton are penalzed. Ths process s contnued untl the weght of MPIDS s equal to pre-defned threshold. Fgure 3 denotes the flowchart of the proposed algorthm. Note that, at frst, learnng automata chooses ts acton wth equal probablty and wll gradually converge to an approprate value durng teratons of the learnng process. One of the most mportant parts n all learnng algorthms s how the algorthm s termnated. In ths way, there are varous solutons are suggested. One of the common way s usng a lmtaton on the number of teratons. If the number of teratons exceeds from a predefned threshold, then the algorthm stops. Another approach s computng the mean of the probablty vector. After some teratons, f the mean of the probablty vector s fxed then the algorthm stops. The mean of probabltes s expressed n equaton (3) Intalzaton of the network of LAs Weghted Postve Influence Domnatng Set (WPIDS) Formaton No Start WPIDS Verfcaton and Envronment Evaluaton Termnaton Condton End Yes Fg3. Flow Chart of Proposed Algorthm based on Learnng Automata for WPIDS. VI. SIMULATION RESULTS In order to evaluate the proposed algorthm, DIMACS [19] benchmark s used n experments. DIMACS s a wellknown benchmark. Ths benchmark contans dfferent networks whch are descrbed n Table 1. For the proposed algorthm, all experments have been mplemented usng L R- I learnng algorthm for the learnng automata. Snce the proposed algorthm operate on networks wth vertex weght and there s no dataset for the socal networks wth vertex 4
5 weght, the benchmark s modfed usng method reported by Pullan et al. [20] whch s descrbed as follows. Based on ths method, the weght corresponds to node v can be computed by ( mod 200) +1. Ths method was also used n[21, 22]. Table1. Descrpton of the datasets used for experments. Networks #nodes #Edges D max D avg Brock k C k C k Dsjc k The results are reported n terms of the mean and standard devaton wth the average of n 1000 runs for the proposed algorthm. Note that n denotes the number of nodes n the network. The executon tme of the proposed algorthm s reported n terms of mnute. The metrcs used for the comparson are descrbed as below. Dynamc Threshold (DT) : ths metrc computes the average weght of the PIDS. Mean-Value: Ths metrc s computed usng equaton (3). Number of actvated nodes: Ths metrc computes the number of actvated nodes of the network. The desgn of the experments s gven as follows. The frst experment s conducted to study the mpact of learnng rate on the executon tme and DT. The second experment s desgned to study the behavor of the proposed algorthm wth respect to DT and Mean value durng the process of fndng the soluton for WIPIDS n networks. The thrd experment s conducted to study the mpact of selectng nfluental nodes on the performance of the nfluence maxmzaton algorthms wth respect to the number of actvated nodes. Experment 1 Ths experment s carred out to study the mpact of rate a on the performance of the proposed algorthm performance of the proposed algorthms. For ths purpose, the WPIDS algorthm s evaluated for fve dfferent learnng parameter {0.001, 0.01, 0.0.5, 0.1, 0.5}. As can be seen n Fgure 5 and Fgure 6, the dynamc threshold and executon tme versus dfferent learnng rate are plotted respectvely. The results show that the proposed Algorthm for whch learnng rate s small, works better than when the learnng rate s large due to avodng local mnmum. But choosng a small learnng rate may ncrease the learnng tme. For ths problem, we have shown that expermentally the learnng rate 0.05 can create a trade-off between executon tme of the learnng algorthm and the accuracy of the obtaned soluton s determned by the learnng algorthm. Hence, for the rest of the experment, we select 0.05 as the learnng rate for the learnng algorthm. Dynamc Threashold Executon Tme(Mnute) Brock200 C250 C500 Dsjc Learnng rate (a) Fg 4. Brock200 C250 C500 Dsjc Learnng Rate (a) Fg. 5. Comparson of proposed algorthm n terms executon tme for dfferent learnng rate. Experment 2 Ths experment s nvestgated to study the behavor of the proposed algorthm durng the process of fndng the soluton for WIPIDS n networks. In order to perform ths experment, we plot both DT and Mean-Value versus teratons. We note that DT s scaled between (0, 1) by dvdng DT by the weght of maxmum WPIDS for each network. The results of ths experment for dfferent test graphs are gven n Fg. 6. As t s shown, DT gradually converges to the weght of mnmum postve nfluence domnatng set and at the same tme Mean-Value converges to 1. Ths means that the algorthm gradually tends to fnd the WPIDS wth mnmum. 5
6 2) a nodes. Based on these experment our algorthm s compared wth four well-known algorthm whch s reported for nfluence maxmzaton ncludng:,hgh-degree [12], PageRank[23], DegGreedy[24]and random generaton algorthm on the all datasets. In Hgh-degree algorthm k nodes s obtaned based on the maxmum degree. For PageRank algorthm, we have chosen the k hghest ranked nodes as ntal seed. The Page-rank algorthm computes and ranked nodes based on the sgnfcance. The DegGreedy algorthm try to maxmze the nfluence based on node neghborhoods greedly and proved effcency of the algorthm n terms of scalablty and hgher spread of nfluence. In random nodes are selected unformly at random. For ths experment, the learnng rates of the LA s are set to In addton, for all networks, we assume that the threshold 0. 5 based the Lnear Threshold model. In addton we consdered that all undrected datasets are double weght and drected for nfluence maxmzaton, n whch the weght node s equal to 1/d where d s the degree of node. Fgure 7 shows number of actvated nodes n comparson wth ntal seed sze for all data sets. 6)b 6) c 7)a 6)d Fg 6. The plot of DT and Mean-Value versus teraton number for dfferent Networks 7)b Experment 3 The goal of ths experment s to study the mpact of selectng nfluental nodes on the performance of the nfluence maxmzaton wth respect to number of actvated 6
7 showed that the proposed algorthm leadng to promsng results n terms of the number of actvated nodes for nfluence maxmzaton. Note that, due to the exponental complexty of WMPIDS problem, the algorthm s not supposed to reach to an optmal soluton n reasonable tme, but the results confrm that proposed algorthm produces better solutons than other well-known algorthms n terms of nfluence maxmzaton. 7)C 7)D Fgure 7: Comparng spread of nfluence usng random, hghdegree, PageRank, deggreedy and WPIS algorthms for the datasets As can be seen from fgure 7, t's compared spread of nfluence usng random, hgh-degree, PageRank, DegGreedy and WPIDS algorthms for the all datasets. From the result, we may conclude that, DegGreedy algorthm outperforms than other heurstc algorthms, n addton hgh-degree centralty algorthm whch relyng solely on the structural propertes of a network doesn't make sense for nfluence maxmzaton. On the other hand, we need to also explctly capture the dynamcs of nformaton whch s captured by WPIDS nodes n the networks. VII. CONCLUSION In ths paper, an algorthm for mnmum weghted postve nfluence domnatng set n socal networks usng learnng automata was proposed. Ths algorthm s the frst algorthm for the mnmum weghted postve nfluence domnatng set problem called MWPIDS n the lterature. Based on the proposed algorthm, each node wth the am of a learnng automaton s able to determne tself to the canddate or non-canddate for the MWPIDS. The proposed learnng automata based algorthm, take the advantage of cooperaton of learnng automata wth each other to fnd MWPIDS. In order to evaluate the performance of the proposed algorthm, a number of experments have been conducted on popular networks. Expermental results VIII. REFERENCES [1] Farbank, V., A study nto the effectveness of vral marketng over the nternet. Retreved Aprl, : p [2] Esnaashar, M. and M. Meybod, A cellular learnng automata based clusterng algorthm for wreless sensor networks. Sensor Letters, (5): p [3] Saghr, A.M. and M.R. Meybod, A Self-adaptve Algorthm for Topology Matchng n Unstructured Peer-to-Peer Networks. Journal of Network and Systems Management, (2): p [4] Rezvanan, A., M. Rahmat, and M.R. Meybod, Samplng from complex networks usng dstrbuted learnng automata. Physca A: Statstcal Mechancs and ts Applcatons, : p [5] Clark, B.N., C.J. Colbourn, and D.S. Johnson, Unt dsk graphs. Dscrete mathematcs, (1-3): p [6] Wang, F., et al., On postve nfluence domnatng sets n socal networks. Theoretcal Computer Scence, (3): p [7] Rae, H., N. Yazdan, and M. Asadpour. A new algorthm for postve nfluence domnatng set n socal networks. n Proceedngs of the 2012 Internatonal Conference on Advances n Socal Networks Analyss and Mnng (ASONAM 2012) IEEE Computer Socety. [8] Zhang, W., et al., Postve nfluence domnatng sets n power-law graphs. Socal Network Analyss and Mnng, (1): p [9] Dnh, T.N., et al., On the approxmablty of postve nfluence domnatng set n socal networks. Journal of Combnatoral Optmzaton, (3): p [10] Basuchowdhur, P. and S. Majumder. Fndng nfluental nodes n socal networks usng mnmum k-hop domnatng set. n Internatonal Conference on Appled Algorthms Sprnger. [11] Zhang, Z., et al. Vral marketng wth postve nfluence. n INFOCOM 2017-IEEE Conference on Computer Communcatons, IEEE IEEE. [12] Kempe, D., J. Klenberg, and É. Tardos. Maxmzng the spread of nfluence through a socal network. n Proceedngs of the nnth ACM 7
8 SIGKDD nternatonal conference on Knowledge dscovery and data mnng ACM. [13] Kempe, D., J.M. Klenberg, and É. Tardos, Maxmzng the Spread of Influence through a Socal Network. Theory of Computng, (4): p [14] Chen, W., Y. Wang, and S. Yang. Effcent nfluence maxmzaton n socal networks. n Proceedngs of the 15th ACM SIGKDD nternatonal conference on Knowledge dscovery and data mnng ACM. [15] Chen, W., C. Wang, and Y. Wang. Scalable nfluence maxmzaton for prevalent vral marketng n large-scale socal networks. n Proceedngs of the 16th ACM SIGKDD nternatonal conference on Knowledge dscovery and data mnng ACM. [16] Km, J., S.-K. Km, and H. Yu. Scalable and parallelzable processng of nfluence maxmzaton for large-scale socal networks? n Data Engneerng (ICDE), 2013 IEEE 29th Internatonal Conference on IEEE. [17] Yang, Y., et al. Contnuous nfluence maxmzaton: What dscounts should we offer to socal network users? n Proceedngs of the 2016 Internatonal Conference on Management of Data ACM. [18] Goyal, A., W. Lu, and L.V. Lakshmanan. Smpath: An effcent algorthm for nfluence maxmzaton under the lnear threshold model. n Data Mnng (ICDM), 2011 IEEE 11th Internatonal Conference on IEEE. [19] Tomta, E. and T. Sek, An effcent branch-andbound algorthm for fndng a maxmum clque. DMTCS, : p [20] Pullan, W., Approxmatng the maxmum vertex/edge weghted clque usng local search. Journal of Heurstcs, (2): p [21] Wu, Q. and J.-K. Hao, A revew on algorthms for maxmum clque problems. European Journal of Operatonal Research, (3): p [22] Gouvea, L. and P. Martns, Solvng the maxmum edge-weght clque problem n sparse graphs wth compact formulatons. EURO Journal on Computatonal Optmzaton, (1): p [23] Brn, S. and L. Page, Reprnt of: The anatomy of a large-scale hypertextual web search engne. Computer networks, (18): p [24] Nand, G., U. Sharma, and A. Das, A Novel Hybrd Approach for Influence Maxmzaton n Onlne Socal Networks Based on Node Neghborhoods, n Advances n Electroncs, Communcaton and Computng: ETAEERE-2016, A. Kalam, S. Das, and K. Sharma, Edtors. 2018, Sprnger Sngapore: Sngapore. p
Parallelism for Nested Loops with Non-uniform and Flow Dependences
Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr
More informationNUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS
ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana
More informationProblem Definitions and Evaluation Criteria for Computational Expensive Optimization
Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty
More informationLoad Balancing for Hex-Cell Interconnection Network
Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,
More informationConstructing Minimum Connected Dominating Set: Algorithmic approach
Constructng Mnmum Connected Domnatng Set: Algorthmc approach G.N. Puroht and Usha Sharma Centre for Mathematcal Scences, Banasthal Unversty, Rajasthan 304022 usha.sharma94@yahoo.com Abstract: Connected
More informationX- Chart Using ANOM Approach
ISSN 1684-8403 Journal of Statstcs Volume 17, 010, pp. 3-3 Abstract X- Chart Usng ANOM Approach Gullapall Chakravarth 1 and Chaluvad Venkateswara Rao Control lmts for ndvdual measurements (X) chart are
More informationClassifier Selection Based on Data Complexity Measures *
Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.
More informationFINDING IMPORTANT NODES IN SOCIAL NETWORKS BASED ON MODIFIED PAGERANK
FINDING IMPORTANT NODES IN SOCIAL NETWORKS BASED ON MODIFIED PAGERANK L-qng Qu, Yong-quan Lang 2, Jng-Chen 3, 2 College of Informaton Scence and Technology, Shandong Unversty of Scence and Technology,
More informationA Binarization Algorithm specialized on Document Images and Photos
A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a
More informationThe Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique
//00 :0 AM Outlne and Readng The Greedy Method The Greedy Method Technque (secton.) Fractonal Knapsack Problem (secton..) Task Schedulng (secton..) Mnmum Spannng Trees (secton.) Change Money Problem Greedy
More informationDetermining the Optimal Bandwidth Based on Multi-criterion Fusion
Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn
More informationCluster Analysis of Electrical Behavior
Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School
More informationA MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS
Proceedngs of the Wnter Smulaton Conference M E Kuhl, N M Steger, F B Armstrong, and J A Jones, eds A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Mark W Brantley Chun-Hung
More informationContent Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers
IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth
More informationLearning the Kernel Parameters in Kernel Minimum Distance Classifier
Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department
More informationIdentifying Top-k Most Influential Nodes by using the Topological Diffusion Models in the Complex Networks
(IJACSA) Internatonal Journal of Advanced Computer Scence and Applcatons, Vol. 8, No., 07 Identfyng Top-k Most Influental Nodes by usng the Topologcal Dffuson Models n the Complex Networks Maryam Padar,
More informationIrregular Cellular Learning Automata and Its Application to Clustering in Sensor Networks
Irregular Cellular Learnng Automata and Its Applcaton to Clusterng n Sensor Networks M. Esnaashar 1, M. R. Meybod 1,2 1 Soft Computng Laboratory, Computer Engneerng and Informaton Technology Department
More informationAn Optimal Algorithm for Prufer Codes *
J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,
More informationAn Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation
17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 An Iteratve Soluton Approach to Process Plant Layout usng Mxed
More informationA New Approach For the Ranking of Fuzzy Sets With Different Heights
New pproach For the ankng of Fuzzy Sets Wth Dfferent Heghts Pushpnder Sngh School of Mathematcs Computer pplcatons Thapar Unversty, Patala-7 00 Inda pushpndersnl@gmalcom STCT ankng of fuzzy sets plays
More informationThe Research of Support Vector Machine in Agricultural Data Classification
The Research of Support Vector Machne n Agrcultural Data Classfcaton Le Sh, Qguo Duan, Xnmng Ma, Me Weng College of Informaton and Management Scence, HeNan Agrcultural Unversty, Zhengzhou 45000 Chna Zhengzhou
More informationLoad-Balanced Anycast Routing
Load-Balanced Anycast Routng Chng-Yu Ln, Jung-Hua Lo, and Sy-Yen Kuo Department of Electrcal Engneerng atonal Tawan Unversty, Tape, Tawan sykuo@cc.ee.ntu.edu.tw Abstract For fault-tolerance and load-balance
More informationTerm Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task
Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto
More informationMeta-heuristics for Multidimensional Knapsack Problems
2012 4th Internatonal Conference on Computer Research and Development IPCSIT vol.39 (2012) (2012) IACSIT Press, Sngapore Meta-heurstcs for Multdmensonal Knapsack Problems Zhbao Man + Computer Scence Department,
More informationImprovement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration
Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,
More informationDistributed Middlebox Placement Based on Potential Game
Int. J. Communcatons, Network and System Scences, 2017, 10, 264-273 http://www.scrp.org/ournal/cns ISSN Onlne: 1913-3723 ISSN Prnt: 1913-3715 Dstrbuted Mddlebox Placement Based on Potental Game Yongwen
More informationEVALUATION OF THE PERFORMANCES OF ARTIFICIAL BEE COLONY AND INVASIVE WEED OPTIMIZATION ALGORITHMS ON THE MODIFIED BENCHMARK FUNCTIONS
Academc Research Internatonal ISS-L: 3-9553, ISS: 3-9944 Vol., o. 3, May 0 EVALUATIO OF THE PERFORMACES OF ARTIFICIAL BEE COLOY AD IVASIVE WEED OPTIMIZATIO ALGORITHMS O THE MODIFIED BECHMARK FUCTIOS Dlay
More informationCompiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz
Compler Desgn Sprng 2014 Regster Allocaton Sample Exercses and Solutons Prof. Pedro C. Dnz USC / Informaton Scences Insttute 4676 Admralty Way, Sute 1001 Marna del Rey, Calforna 90292 pedro@s.edu Regster
More informationSum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints
Australan Journal of Basc and Appled Scences, 2(4): 1204-1208, 2008 ISSN 1991-8178 Sum of Lnear and Fractonal Multobjectve Programmng Problem under Fuzzy Rules Constrants 1 2 Sanjay Jan and Kalash Lachhwan
More information6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour
6.854 Advanced Algorthms Petar Maymounkov Problem Set 11 (November 23, 2005) Wth: Benjamn Rossman, Oren Wemann, and Pouya Kheradpour Problem 1. We reduce vertex cover to MAX-SAT wth weghts, such that the
More informationApplying Continuous Action Reinforcement Learning Automata(CARLA) to Global Training of Hidden Markov Models
Applyng Contnuous Acton Renforcement Learnng Automata(CARLA to Global Tranng of Hdden Markov Models Jahanshah Kabudan, Mohammad Reza Meybod, and Mohammad Mehd Homayounpour Department of Computer Engneerng
More informationWishing you all a Total Quality New Year!
Total Qualty Management and Sx Sgma Post Graduate Program 214-15 Sesson 4 Vnay Kumar Kalakband Assstant Professor Operatons & Systems Area 1 Wshng you all a Total Qualty New Year! Hope you acheve Sx sgma
More informationOutline. Type of Machine Learning. Examples of Application. Unsupervised Learning
Outlne Artfcal Intellgence and ts applcatons Lecture 8 Unsupervsed Learnng Professor Danel Yeung danyeung@eee.org Dr. Patrck Chan patrckchan@eee.org South Chna Unversty of Technology, Chna Introducton
More informationLearning-Based Top-N Selection Query Evaluation over Relational Databases
Learnng-Based Top-N Selecton Query Evaluaton over Relatonal Databases Lang Zhu *, Wey Meng ** * School of Mathematcs and Computer Scence, Hebe Unversty, Baodng, Hebe 071002, Chna, zhu@mal.hbu.edu.cn **
More informationA mathematical programming approach to the analysis, design and scheduling of offshore oilfields
17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 A mathematcal programmng approach to the analyss, desgn and
More informationNon-Split Restrained Dominating Set of an Interval Graph Using an Algorithm
Internatonal Journal of Advancements n Research & Technology, Volume, Issue, July- ISS - on-splt Restraned Domnatng Set of an Interval Graph Usng an Algorthm ABSTRACT Dr.A.Sudhakaraah *, E. Gnana Deepka,
More informationVirtual Machine Migration based on Trust Measurement of Computer Node
Appled Mechancs and Materals Onlne: 2014-04-04 ISSN: 1662-7482, Vols. 536-537, pp 678-682 do:10.4028/www.scentfc.net/amm.536-537.678 2014 Trans Tech Publcatons, Swtzerland Vrtual Machne Mgraton based on
More informationAn Efficient Genetic Algorithm Based Approach for the Minimum Graph Bisection Problem
118 An Effcent Genetc Algorthm Based Approach for the Mnmum Graph Bsecton Problem Zh-Qang Chen, Rong-Long WAG and Kozo OKAZAKI Faculty of Engneerng, Unversty of Fuku, Bunkyo 3-9-1,Fuku-sh, Japan 910-8507
More informationS1 Note. Basis functions.
S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type
More informationTsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance
Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for
More informationARTICLE IN PRESS. Signal Processing: Image Communication
Sgnal Processng: Image Communcaton 23 (2008) 754 768 Contents lsts avalable at ScenceDrect Sgnal Processng: Image Communcaton journal homepage: www.elsever.com/locate/mage Dstrbuted meda rate allocaton
More informationUnsupervised Learning
Pattern Recognton Lecture 8 Outlne Introducton Unsupervsed Learnng Parametrc VS Non-Parametrc Approach Mxture of Denstes Maxmum-Lkelhood Estmates Clusterng Prof. Danel Yeung School of Computer Scence and
More informationA Novel Adaptive Descriptor Algorithm for Ternary Pattern Textures
A Novel Adaptve Descrptor Algorthm for Ternary Pattern Textures Fahuan Hu 1,2, Guopng Lu 1 *, Zengwen Dong 1 1.School of Mechancal & Electrcal Engneerng, Nanchang Unversty, Nanchang, 330031, Chna; 2. School
More informationOnline Policies for Opportunistic Virtual MISO Routing in Wireless Ad Hoc Networks
12 IEEE Wreless Communcatons and Networkng Conference: Moble and Wreless Networks Onlne Polces for Opportunstc Vrtual MISO Routng n Wreless Ad Hoc Networks Crstano Tapparello, Stefano Tomasn and Mchele
More informationComparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments
Comparson of Heurstcs for Schedulng Independent Tasks on Heterogeneous Dstrbuted Envronments Hesam Izakan¹, Ath Abraham², Senor Member, IEEE, Václav Snášel³ ¹ Islamc Azad Unversty, Ramsar Branch, Ramsar,
More informationCourse Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms
Course Introducton Course Topcs Exams, abs, Proects A quc loo at a few algorthms 1 Advanced Data Structures and Algorthms Descrpton: We are gong to dscuss algorthm complexty analyss, algorthm desgn technques
More informationHelsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr)
Helsnk Unversty Of Technology, Systems Analyss Laboratory Mat-2.08 Independent research projects n appled mathematcs (3 cr) "! #$&% Antt Laukkanen 506 R ajlaukka@cc.hut.f 2 Introducton...3 2 Multattrbute
More informationIncremental Learning with Support Vector Machines and Fuzzy Set Theory
The 25th Workshop on Combnatoral Mathematcs and Computaton Theory Incremental Learnng wth Support Vector Machnes and Fuzzy Set Theory Yu-Mng Chuang 1 and Cha-Hwa Ln 2* 1 Department of Computer Scence and
More informationA Saturation Binary Neural Network for Crossbar Switching Problem
A Saturaton Bnary Neural Network for Crossbar Swtchng Problem Cu Zhang 1, L-Qng Zhao 2, and Rong-Long Wang 2 1 Department of Autocontrol, Laonng Insttute of Scence and Technology, Benx, Chna bxlkyzhangcu@163.com
More informationClassifying Acoustic Transient Signals Using Artificial Intelligence
Classfyng Acoustc Transent Sgnals Usng Artfcal Intellgence Steve Sutton, Unversty of North Carolna At Wlmngton (suttons@charter.net) Greg Huff, Unversty of North Carolna At Wlmngton (jgh7476@uncwl.edu)
More informationGA-Based Learning Algorithms to Identify Fuzzy Rules for Fuzzy Neural Networks
Seventh Internatonal Conference on Intellgent Systems Desgn and Applcatons GA-Based Learnng Algorthms to Identfy Fuzzy Rules for Fuzzy Neural Networks K Almejall, K Dahal, Member IEEE, and A Hossan, Member
More informationModule Management Tool in Software Development Organizations
Journal of Computer Scence (5): 8-, 7 ISSN 59-66 7 Scence Publcatons Management Tool n Software Development Organzatons Ahmad A. Al-Rababah and Mohammad A. Al-Rababah Faculty of IT, Al-Ahlyyah Amman Unversty,
More informationTECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS. Muradaliyev A.Z.
TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS Muradalyev AZ Azerbajan Scentfc-Research and Desgn-Prospectng Insttute of Energetc AZ1012, Ave HZardab-94 E-mal:aydn_murad@yahoocom Importance of
More informationCHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION
24 CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION The present chapter proposes an IPSO approach for multprocessor task schedulng problem wth two classfcatons, namely, statc ndependent tasks and
More informationRecommended Items Rating Prediction based on RBF Neural Network Optimized by PSO Algorithm
Recommended Items Ratng Predcton based on RBF Neural Network Optmzed by PSO Algorthm Chengfang Tan, Cayn Wang, Yuln L and Xx Q Abstract In order to mtgate the data sparsty and cold-start problems of recommendaton
More informationDistributed Resource Scheduling in Grid Computing Using Fuzzy Approach
Dstrbuted Resource Schedulng n Grd Computng Usng Fuzzy Approach Shahram Amn, Mohammad Ahmad Computer Engneerng Department Islamc Azad Unversty branch Mahallat, Iran Islamc Azad Unversty branch khomen,
More informationHigh-Boost Mesh Filtering for 3-D Shape Enhancement
Hgh-Boost Mesh Flterng for 3-D Shape Enhancement Hrokazu Yagou Λ Alexander Belyaev y Damng We z Λ y z ; ; Shape Modelng Laboratory, Unversty of Azu, Azu-Wakamatsu 965-8580 Japan y Computer Graphcs Group,
More informationAn Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices
Internatonal Mathematcal Forum, Vol 7, 2012, no 52, 2549-2554 An Applcaton of the Dulmage-Mendelsohn Decomposton to Sparse Null Space Bases of Full Row Rank Matrces Mostafa Khorramzadeh Department of Mathematcal
More informationA Fast Content-Based Multimedia Retrieval Technique Using Compressed Data
A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,
More informationAn Entropy-Based Approach to Integrated Information Needs Assessment
Dstrbuton Statement A: Approved for publc release; dstrbuton s unlmted. An Entropy-Based Approach to ntegrated nformaton Needs Assessment June 8, 2004 Wllam J. Farrell Lockheed Martn Advanced Technology
More informationSupport Vector Machines
/9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.
More informationBackpropagation: In Search of Performance Parameters
Bacpropagaton: In Search of Performance Parameters ANIL KUMAR ENUMULAPALLY, LINGGUO BU, and KHOSROW KAIKHAH, Ph.D. Computer Scence Department Texas State Unversty-San Marcos San Marcos, TX-78666 USA ae049@txstate.edu,
More informationUser Authentication Based On Behavioral Mouse Dynamics Biometrics
User Authentcaton Based On Behavoral Mouse Dynamcs Bometrcs Chee-Hyung Yoon Danel Donghyun Km Department of Computer Scence Department of Computer Scence Stanford Unversty Stanford Unversty Stanford, CA
More informationNetwork Intrusion Detection Based on PSO-SVM
TELKOMNIKA Indonesan Journal of Electrcal Engneerng Vol.1, No., February 014, pp. 150 ~ 1508 DOI: http://dx.do.org/10.11591/telkomnka.v1.386 150 Network Intruson Detecton Based on PSO-SVM Changsheng Xang*
More informationBioTechnology. An Indian Journal FULL PAPER. Trade Science Inc.
[Type text] [Type text] [Type text] ISSN : 0974-74 Volume 0 Issue BoTechnology 04 An Indan Journal FULL PAPER BTAIJ 0() 04 [684-689] Revew on Chna s sports ndustry fnancng market based on market -orented
More informationBIN XIA et al: AN IMPROVED K-MEANS ALGORITHM BASED ON CLOUD PLATFORM FOR DATA MINING
An Improved K-means Algorthm based on Cloud Platform for Data Mnng Bn Xa *, Yan Lu 2. School of nformaton and management scence, Henan Agrcultural Unversty, Zhengzhou, Henan 450002, P.R. Chna 2. College
More informationObstacle-Aware Routing Problem in. a Rectangular Mesh Network
Appled Mathematcal Scences, Vol. 9, 015, no. 14, 653-663 HIKARI Ltd, www.m-hkar.com http://dx.do.org/10.1988/ams.015.411911 Obstacle-Aware Routng Problem n a Rectangular Mesh Network Norazah Adzhar Department
More informationEffective Page Recommendation Algorithms Based on. Distributed Learning Automata and Weighted Association. Rules
Effectve Page Recommendaton Algorthms Based on Dstrbuted Learnng Automata and Weghted Assocaton Rules R. Forsat 1*, M. R. Meybod 2 1 Department of Computer Engneerng, Islamc Azad Unversty, Karaj Branch,
More informationSupport Vector Machines
Support Vector Machnes Decson surface s a hyperplane (lne n 2D) n feature space (smlar to the Perceptron) Arguably, the most mportant recent dscovery n machne learnng In a nutshell: map the data to a predetermned
More informationEfficient Distributed File System (EDFS)
Effcent Dstrbuted Fle System (EDFS) (Sem-Centralzed) Debessay(Debsh) Fesehaye, Rahul Malk & Klara Naherstedt Unversty of Illnos-Urbana Champagn Contents Problem Statement, Related Work, EDFS Desgn Rate
More informationA Facet Generation Procedure. for solving 0/1 integer programs
A Facet Generaton Procedure for solvng 0/ nteger programs by Gyana R. Parja IBM Corporaton, Poughkeepse, NY 260 Radu Gaddov Emery Worldwde Arlnes, Vandala, Oho 45377 and Wlbert E. Wlhelm Teas A&M Unversty,
More informationMaximum Variance Combined with Adaptive Genetic Algorithm for Infrared Image Segmentation
Internatonal Conference on Logstcs Engneerng, Management and Computer Scence (LEMCS 5) Maxmum Varance Combned wth Adaptve Genetc Algorthm for Infrared Image Segmentaton Huxuan Fu College of Automaton Harbn
More informationNetwork Coding as a Dynamical System
Network Codng as a Dynamcal System Narayan B. Mandayam IEEE Dstngushed Lecture (jont work wth Dan Zhang and a Su) Department of Electrcal and Computer Engneerng Rutgers Unversty Outlne. Introducton 2.
More informationNeural Network Control for TCP Network Congestion
5 Amercan Control Conference June 8-, 5. Portland, OR, USA FrA3. Neural Network Control for TCP Network Congeston Hyun C. Cho, M. Sam Fadal, Hyunjeong Lee Electrcal Engneerng/6, Unversty of Nevada, Reno,
More informationSHAPE OPTIMIZATION OF STRUCTURES BY MODIFIED HARMONY SEARCH
INTERNATIONAL JOURNAL OF OPTIMIZATION IN CIVIL ENGINEERING Int. J. Optm. Cvl Eng., 2011; 3:485-494 SHAPE OPTIMIZATION OF STRUCTURES BY MODIFIED HARMONY SEARCH S. Gholzadeh *,, A. Barzegar and Ch. Gheyratmand
More informationEvaluation of Parallel Processing Systems through Queuing Model
ISSN 2278-309 Vkas Shnde, Internatonal Journal of Advanced Volume Trends 4, n Computer No.2, March Scence - and Aprl Engneerng, 205 4(2), March - Aprl 205, 36-43 Internatonal Journal of Advanced Trends
More informationLECTURE : MANIFOLD LEARNING
LECTURE : MANIFOLD LEARNING Rta Osadchy Some sldes are due to L.Saul, V. C. Raykar, N. Verma Topcs PCA MDS IsoMap LLE EgenMaps Done! Dmensonalty Reducton Data representaton Inputs are real-valued vectors
More informationQuality Improvement Algorithm for Tetrahedral Mesh Based on Optimal Delaunay Triangulation
Intellgent Informaton Management, 013, 5, 191-195 Publshed Onlne November 013 (http://www.scrp.org/journal/m) http://dx.do.org/10.36/m.013.5601 Qualty Improvement Algorthm for Tetrahedral Mesh Based on
More informationOutline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like:
Self-Organzng Maps (SOM) Turgay İBRİKÇİ, PhD. Outlne Introducton Structures of SOM SOM Archtecture Neghborhoods SOM Algorthm Examples Summary 1 2 Unsupervsed Hebban Learnng US Hebban Learnng, Cntd 3 A
More informationPerformance Evaluation of Information Retrieval Systems
Why System Evaluaton? Performance Evaluaton of Informaton Retreval Systems Many sldes n ths secton are adapted from Prof. Joydeep Ghosh (UT ECE) who n turn adapted them from Prof. Dk Lee (Unv. of Scence
More informationOverview. Basic Setup [9] Motivation and Tasks. Modularization 2008/2/20 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION
Overvew 2 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION Introducton Mult- Smulator MASIM Theoretcal Work and Smulaton Results Concluson Jay Wagenpfel, Adran Trachte Motvaton and Tasks Basc Setup
More informationA Robust Method for Estimating the Fundamental Matrix
Proc. VIIth Dgtal Image Computng: Technques and Applcatons, Sun C., Talbot H., Ourseln S. and Adraansen T. (Eds.), 0- Dec. 003, Sydney A Robust Method for Estmatng the Fundamental Matrx C.L. Feng and Y.S.
More informationDESIGNING TRANSMISSION SCHEDULES FOR WIRELESS AD HOC NETWORKS TO MAXIMIZE NETWORK THROUGHPUT
DESIGNING TRANSMISSION SCHEDULES FOR WIRELESS AD HOC NETWORKS TO MAXIMIZE NETWORK THROUGHPUT Bran J. Wolf, Joseph L. Hammond, and Harlan B. Russell Dept. of Electrcal and Computer Engneerng, Clemson Unversty,
More informationPositive Semi-definite Programming Localization in Wireless Sensor Networks
Postve Sem-defnte Programmng Localzaton n Wreless Sensor etworks Shengdong Xe 1,, Jn Wang, Aqun Hu 1, Yunl Gu, Jang Xu, 1 School of Informaton Scence and Engneerng, Southeast Unversty, 10096, anjng Computer
More informationImperialist Competitive Algorithm with Variable Parameters to Determine the Global Minimum of Functions with Several Arguments
Fourth Internatonal Conference Modellng and Development of Intellgent Systems October 8 - November, 05 Lucan Blaga Unversty Sbu - Romana Imperalst Compettve Algorthm wth Varable Parameters to Determne
More informationSLAM Summer School 2006 Practical 2: SLAM using Monocular Vision
SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,
More informationOptimal Workload-based Weighted Wavelet Synopses
Optmal Workload-based Weghted Wavelet Synopses Yoss Matas School of Computer Scence Tel Avv Unversty Tel Avv 69978, Israel matas@tau.ac.l Danel Urel School of Computer Scence Tel Avv Unversty Tel Avv 69978,
More informationCS 534: Computer Vision Model Fitting
CS 534: Computer Vson Model Fttng Sprng 004 Ahmed Elgammal Dept of Computer Scence CS 534 Model Fttng - 1 Outlnes Model fttng s mportant Least-squares fttng Maxmum lkelhood estmaton MAP estmaton Robust
More informationImplementation Naïve Bayes Algorithm for Student Classification Based on Graduation Status
Internatonal Journal of Appled Busness and Informaton Systems ISSN: 2597-8993 Vol 1, No 2, September 2017, pp. 6-12 6 Implementaton Naïve Bayes Algorthm for Student Classfcaton Based on Graduaton Status
More informationUsing Particle Swarm Optimization for Enhancing the Hierarchical Cell Relay Routing Protocol
2012 Thrd Internatonal Conference on Networkng and Computng Usng Partcle Swarm Optmzaton for Enhancng the Herarchcal Cell Relay Routng Protocol Hung-Y Ch Department of Electrcal Engneerng Natonal Sun Yat-Sen
More informationSimulation Based Analysis of FAST TCP using OMNET++
Smulaton Based Analyss of FAST TCP usng OMNET++ Umar ul Hassan 04030038@lums.edu.pk Md Term Report CS678 Topcs n Internet Research Sprng, 2006 Introducton Internet traffc s doublng roughly every 3 months
More informationSmoothing Spline ANOVA for variable screening
Smoothng Splne ANOVA for varable screenng a useful tool for metamodels tranng and mult-objectve optmzaton L. Rcco, E. Rgon, A. Turco Outlne RSM Introducton Possble couplng Test case MOO MOO wth Game Theory
More informationComplex System Reliability Evaluation using Support Vector Machine for Incomplete Data-set
Internatonal Journal of Performablty Engneerng, Vol. 7, No. 1, January 2010, pp.32-42. RAMS Consultants Prnted n Inda Complex System Relablty Evaluaton usng Support Vector Machne for Incomplete Data-set
More informationSubspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;
Subspace clusterng Clusterng Fundamental to all clusterng technques s the choce of dstance measure between data ponts; D q ( ) ( ) 2 x x = x x, j k = 1 k jk Squared Eucldean dstance Assumpton: All features
More informationOptimizing Document Scoring for Query Retrieval
Optmzng Document Scorng for Query Retreval Brent Ellwen baellwe@cs.stanford.edu Abstract The goal of ths project was to automate the process of tunng a document query engne. Specfcally, I used machne learnng
More informationMobileGrid: Capacity-aware Topology Control in Mobile Ad Hoc Networks
MobleGrd: Capacty-aware Topology Control n Moble Ad Hoc Networks Jle Lu, Baochun L Department of Electrcal and Computer Engneerng Unversty of Toronto {jenne,bl}@eecg.toronto.edu Abstract Snce wreless moble
More informationAn Accurate Evaluation of Integrals in Convex and Non convex Polygonal Domain by Twelve Node Quadrilateral Finite Element Method
Internatonal Journal of Computatonal and Appled Mathematcs. ISSN 89-4966 Volume, Number (07), pp. 33-4 Research Inda Publcatons http://www.rpublcaton.com An Accurate Evaluaton of Integrals n Convex and
More informationA Deflected Grid-based Algorithm for Clustering Analysis
A Deflected Grd-based Algorthm for Clusterng Analyss NANCY P. LIN, CHUNG-I CHANG, HAO-EN CHUEH, HUNG-JEN CHEN, WEI-HUA HAO Department of Computer Scence and Informaton Engneerng Tamkang Unversty 5 Yng-chuan
More informationSolving two-person zero-sum game by Matlab
Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by
More informationLecture 5: Multilayer Perceptrons
Lecture 5: Multlayer Perceptrons Roger Grosse 1 Introducton So far, we ve only talked about lnear models: lnear regresson and lnear bnary classfers. We noted that there are functons that can t be represented
More information