Feature Selection Method with Proportionate Fitness Based Binary Particle Swarm Optimization

Size: px
Start display at page:

Download "Feature Selection Method with Proportionate Fitness Based Binary Particle Swarm Optimization"

Transcription

1 Feature Selecton Method wth Proportonate Ftness Based Bnary Partcle Swarm Optmzaton Zhe Zhou, Xng Lu, Png L, and Ln Shang State Key Laboratory of Novel Software Technology, Department of Computer Scence and Technology, Nanjng Unversty, Nanjng , Chna, Abstract. Partcle swarm optmzaton(pso) has been appled on feature selecton wth many mproved results. Tradtonal PSO methods have some drawbacks when dealng wth bnary space, whch may have negatve effects on the selecton result. In ths paper, an algorthm based on ftness proportonate selecton bnary partcle swarm optmzaton(fpsbpso) wll be dscussed n detal amng to overcome the problems of tradtonal PSO methods. FPSBPSO wll be utlzed n the feature subset selecton doman. The performance of feature selecton wll be compared n a benchmark dataset, and expermental results prove that the FPSBPSO-based feature selecton methods can avo premature convergence and mprove the classfcaton accuracy. 1 Introducton Feature Selecton refers to the process of selectng a subset of relevant features n model constructon, whch s often seen n classfcaton problems. A well selected feature subset can boost the classfer s performance and reduce the tme cost at the same tme. Current feature selecton methods nclude frequency-based methods, nformaton-gan-based methods, lexcon-based methods and so on. Beses, some evolutonary algorthms can be utlzed nto feature selecton as well, such as generc algorthms[1] and partcle swarm optmzaton. Partcle swarm optmzaton(pso) s an optmzaton technque frstly ntroduced by Kennedy and Eberhart n 1995 [6], whch was nspred by the socal behavor of brds flockng or fsh schoolng. Snce then, PSO has attracted sgnfcant attenton. Researchers report that compared wth other evolutonary algorthms lke genetc algorthms(ga), PSO s computatonally more effcent [4]. Despte many successful applcatons of PSO, varous varatons of the classc PSO model were proposed amng to acheve better performance. However, they stll share the same basc ea: a swarm s made up by many partcles, each of whch represents a candate soluton. In each teraton, partcles update ther

2 2 poston n the search space accordng to ther own hstorcal nformaton and ther neghbours nformaton. After the teratons reach the end, an optmal soluton s supposed to be found. PSO was orgnally desgned to solve contnuous optmzaton problems. In order to deal wth dscrete problems, Kennedy and Eberhart proposed bnary partcle swarm optmzaton(bpso) [7] n BPSO has attracted much attenton snce nvented, but most researchers have been focusng on the applcaton of BPSO rather than ts theoretcal analyss. Although there are several mproved versons of BPSO, none of them has become a real replacement of BPSO. The orgnal BPSO model s used n most publcatons reportng the applcaton of PSO n dealng wth dscrete problems. The basc ea of PSO s that a partcle can explot ts own hstorcal poston and performance nformaton, as well as the hstorcal poston and performance nformaton of the whole swarm. A partcle tends to move closer to ts own hstorcal best poston and the best poston found by the whole swarm durng teratons. Ths ea enables PSO to fnd an optmal soluton, and t s also the ea shared by most varatons of PSO. However, as we analysed the detals of BPSO, we found that BPSO doesn t follow ths ea, or at least not n the way other PSO varatons do. In fact, a partcle n BPSO wll probably move away from the best poston found by tself or the best poston found by the whole swarm. Ths makes BPSO dffcult to nterpret. Our goal s to fnd a feature subset for a bnary classfcaton problem wth PSO-based methods. In order to make the selected feature subset as powerful as possble, we am to mprove PSO s performance n dscrete space. In order to acheve ths goal, we wll dscuss two major problems of PSO when the search space s dscrete and try to prove that the tradtonal way of calculatng veloctes s the man cause of the two problems. A new way to calculate veloctes based on ftness values s then proposed and based on that we wll propose a new bnary verson of PSO called FPSBPSO. We wll prove the effcency of our method by utlzng the method to perform the feature selecton process n classfcaton problems. If the selected feature subset wth FPSBPSO can return better classfcaton results than that wth tradtonal PSO, our proposed scheme can be regarded as an effcent one. The remander of ths paper s organzed as follows. In secton 2, we revew the development of PSO and BPSO. In secton 3, we show two problems PSO wll have to face n dscrete space, and ntroduce a modfed BPSO method called FPSBPSO. Then we wll utlze FPSBPSO to deal wth the task of feature subset selecton. In secton 4, we dscuss dfferent expermental results whch can prove FPSBPSO s effcency both n optmzaton problems and n feature subset selecton. Conclusons and future work are ncluded n secton 5.

3 3 2 Background 2.1 Contnuous Partcle Swarm Optmzaton PSO was proposed by Kennedy and Eberhart n 1995 [6]. It was nspred by the socal behavor of brds flockng or fsh schoolng. A swarm has some partcles, each partcle has a poston component representng a specfc soluton, and a velocty component representng the drecton of a partcle s movement n the soluton space. PSO s an teratve optmzaton algorthm, and lke other teratve algorthms, t also has manly three steps. The frst step s to ntalze the populaton by generatng each partcle s velocty component and poston component randomly. The second step s to evaluate solutons represented by partcles postons. The fnal step s to update partcles veloctes and then update partcles postons usng the followng formulas. v t+1 = w v t + c 1 rand ( pbest x t ) + c 2 rand ( gbest x t ) (1) x t+1 = x t + v t (2) Where v t+1 and x t+1 represents the velocty component and the poston component of partcle p at the (t+1) th teraton respectvely, c 1 and c 2 are two fxed confence coeffcent, rnd s a unformally dstrbuted random varable rangng from 0 to 1, w s the nerta weght. pbest means the poston of partcle p s personal best whle gbest means the poston of partcle p s global best. The rght se of equaton (1) can be dved nto three parts. w v t represents the prevous drecton, c 1 rand (pbest x t ) represents the tendency of movng towards a partcle s personal best, c 2 rand (gbest x t ) represents the tendency of movng towards the swarm s global best. The three parts cooperate wth each other to gue a partcle s movement. The followng fgure shows the effect of ther cooperaton. Fg. 1. The effect of pbest and gbest. Snce the basc PSO was proposed, researchers have proposed many varatons amng to mprove ts performance. In fact, there was no nerta weght n the

4 4 basc PSO, Sh and Eberhart ntroduced nerta weght nto PSO n 1998 [12], and ths parameter has been adopted n almost every PSO varaton. Other mportant modfcatons to the basc PSO model nclude Bare Bones PSO proposed by Kennedy [5], and the ea to clamp veloctes [3] proposed by Eberhart et al.. Ioan gave a good analyss of PSO s convergence and parameter selecton problem n [14]. 2.2 Bnary Partcle Swarm Optmzaton In order to use PSO to solve dscrete problems, Kennedy proposed BPSO [7], whch became the most wely used PSO model dealng wth combnatoral optmzaton problems. Kennedy manly made two changes to the contnuous PSO, and the frst s the representaton of a partcle s poston. Unlke n the contnuous PSO where a partcle s poston s a set of real values, n BPSO, a partcle s poston becomes a bnary strng lke The second change s that the velocty of a partcle no longer represents the drecton of a partcle s movement, rather, t means the possblty of choong 1 at a specfc bt of a partcle s poston component. And the update fomula of the velocty component becomes the followng one. x = { 1, rand < S(v ) 0, otherwse. (3) where 1 S(v ) = 1 + e v There are many applcatons of BPSO, ncludng usng BPSO to solve feature subset selecton problems [10] [15], the travellng salesman problem [11], the lot szng problem [13], flowshop schedulng problems [9], and many other combnatoral optmzaton problems. (4) 3 Feature Selecton wth FPSBPSO In ths secton, we wll propose a modfed BPSO algorthm called FPSBPSO, whch suts the nature of PSO n dscrete space better. Then we wll perform a feature subset selecton wth the concept of FPSBPSO and test whether our proposed method can be benefcal to the classfcaton process. 3.1 Problems of Tradtonal BPSO In the case of bnary space, the tradtonal velocty component becomes useless because partcles now jump rather than move. In BPSO, veloctes become the probablty of choosng 1 n the next teraton, whch s a reasonable nterpretaton. However, n BPSO, the update formula of veloctes keep unchanged, whch s not reasonable. In classc PSO models, veloctes are calculated usng dstance

5 5 nformaton, for example, the dstance between p and ts global best partcle p gb. It makes sense to use dstance measures to calculate veloctes n contnuous space because partcles are movng from one poston to another poston. However, n bnary space, t does not seem to make sense to follow the same ea. There are actually only three possble dstance values n bnary space, 0, 1, and 1, and t s nature to thnk why these values can determne the possblty of jumpng from 0 to 1 or vce versa, why not use 2, 0, and 2? Or other real numbers? we thnk t s not reasonable to use dstance nformaton to calculate veloctes n dscrete space. There s another problem PSO wll have to face n dscrete space. As we all know, partcles tend to follow a leader of the whole swam or a small group. In contnuous space, the result of ths strategy s that a partcle wll be closer to ts personal best and global best n every dmenson n the next teraton. However, n bnary space, the concept of closer can not work well. We can stll say two partcles are closer when they share the same values on many dmensons, but we can only say two partcles p and p j are closer n the d th dmenson when x d = xd j. Now closer means equalty, whch s a problem. The leader partcle s choosen by comparng every partcle s overall performance, whch means that the leader partcle does not have to make the best choce on every dmenson to be selected as the leader. However, the leader partcle wll effect other partcles choce on every dmenson, causng them to move toward t. In contnuous space, the strategy works okay because partcles wll move to new postons closer to the leader partcle, but not the same poston. Whle n bnary space, snce movng closer to the leader partcle wll decrease the dversty of the swarm because partcles tend to select the same value wth the leader partcle, whch wll lead to premature convergence. 3.2 FPSBPSO: Ftness Proportonate Selecton BPSO The two problems dscussed above greatly affect PSO s performance n dscrete space and we propose solutons to them n ths paper. For the frst problem, we beleve t s better to use performance nformaton rather than dstance nformaton to dece whch state wll be selected. Better performance means hgher possblty to be selected, whch s a reasonable strategy commonly known as ftness proportonate selecton. The soluton to the second problem s based on the frst soluton. In bnary space, x d can only be 0 or 1. For a partcle p, ts global best partcle p gb has the best overall performance, however, ts selecton on the d th bt may not be good. Assume p s global best select 1 on the d th bt. If p s personal best partcle or p tself make the same selecton wth the global best partcle on the d th bt, then the average performance of 1 wll be worse than s sngle performance. So, we use average performance of 0 and 1 to calculate the possblty of selectng 0 and 1, whch can depress the tendency of movng towards the leader partcle to a lower level and thus mantan the dversty of the swarm. Now we descrbe the newly desgned update formulas for poston component and velocty component whch are the essental part of the new model. We adopt p gb

6 6 ftness proportonate selecton to update a partcle s poston component, and that s why we call the new model FPSBPSO. In the followng content, when we refer to nvolved partcles n the follong content, we actually mean the three partcles p, p s personal best partcle p pb newly desgned update formulas are as follows. mr, f n 0 == 0 v t+1 = 1 mr, f n 1 == 0 f 1 f 1 +f 0, otherwse. and p s global best partcle p gb. The (5) { x t+1 0, rand < v t+1 = 1, otherwse. (6), Where n 0 s the number of nvolved partcles wth x = 0 and n 1 s the number of nvolved partcles wth x = 1. f 1 and f 0 are computed as follows: Frst, nvolved partcles are dved nto two sets S 1 and S 0 based on whether they select 1 or 0 at the d th bt. Then, f 1 and f 0 are calculated by averagng the ftness values of partcles n S 1 and S 0 respectvely. v t+1 s the probablty of settng x t+1 to 0. The hgher f 1 s, the hgher probablty x t+1 wll be 0. In ths new method, the update formula for veloctes s dfferent from that of tradtonal BPSO. Frst, n the proposed method, the velocty component s calculated usng the current state nformaton, whle n tradtonal PSO models, prevous veloctes have an mpact on current veloctes. However, abandonng prevous veloctes does not mean that the prevous drecton nformaton s lost n FPSBPSO. We use prevous state to play the role of prevous veloctes n classc PSO. Beses, the new veloctes are deced by partcles ftness values, whle veloctes n tradtonal PSO models are calculated manly based on partcles dstance nformaton, for example, the dstance between p and ts global best partcle p gb. As we have argued n prevous subsecton, t s not reasonable to use dstance nformaton to update veloctes. So we desgned a vote-lke s- trategy n the new model. f 1 and f 0 represents the average performance of 1 and 0 respectvely, and better average performance means that the correspondng bt value wll have hgher chance to be selected. In ths way, we can also solve the second problem we proposed n the prevous subsecton. Our proposed FPSBPSO method has another advantage, whch s that there s only one free parameter n FPSBPSO mr. Fewer parameters means that the model wll be easer to be tuned because the dffculty of fndng a good combnaton of parameter settngs wll ncrease exponentally as the number of parameter ncreases. The parameter mr s ntroduced to prevent the problem of premature convergence. In the proposed model, f x pb = xgb = xt, then xt+1 wll be fxed. For example, f x pb = xgb = xt = 1, then vt+1 wll be 0, whch means x t+1 wll be set to 1 wth probablty 1. Ths wll lead to premature convergence. In ths case, the parameter of mr wll force the bt to mutate at tmes, whch s often seen n generc algorthms. How to select a proper value of mr s an mportant problem n the new model. By comparng the results of FPSBPSO usng dfferent settngs of mr, we found

7 7 two basc regulartes. Frst, f mr s set to 0 or a too small value, premature convergence wll happen. Second, f mr s set to a large value, then t wll be very dffcult for the swarm to convergence. Expermental results show that 0.01 s a good choce for many dfferent problems. However, we suggest that t s better to use early-exploraton-late-explotaton strategy, whch means settng mr to a much larger value n the early stage so that partcles can explore more space and gradually decreasng mr to ensure that the partcles can pay more attenton to explotng the neghbourhood nformaton n the later stage. 3.3 FPSBPSO Based Feature Subset Selecton In most classfcaton tasks, the process of feature selecton must be executed, durng whch a smaller set of all features s ganed n order to reach better classfcaton performance. There are many feature selecton methods, such as nformaton gan based method, lexcon based method, generc algorthm based method and so on. PSO can be used to perform feature selecton as well. Smlar to the generc algorthm based method, every feature selecton result s transformed nto a bt strng n PSO based method. The length of strng s equal to the sze of all dfferent features and the value of each bt represents whether the correspondng feature s selected or not. The detaled algorthm s shown n Table 1. As the man am of feature selecton s to rase the classfcaton accuracy, the ftness value of a partcle s represented by the classfer s accuracy usng the correspondng feature subset. Our am s to use FPSBPSO to perform the feature selecton process to obtan a feature subset whch s supposed to get better classfcaton results than that brought by a tradtonal-bpso-based one. 4 Expermental Results In ths secton, we am to prove FPSBPSO s effectveness n feature selecton. Ths s done through two steps. We frst compare the performance of tradtonal BPSO and FPSBPSO on the MKP problem to show that our modfcaton on tradtonal BPSO s reasonable. Then we utlze FPSBPSO nto the feature selecton n a bnary classfcaton problem to test whether the feature subset can mprove the accuracy of classfcaton. 4.1 Experments for the Problem of MKP Frst we want to prove that FPSBPSO works better than tradtonal BPSO n bnary space by comparng the two algorthms on the problem of MKP. MKP s a an ntensvely studed dscrete optmzaton problem because many real world problems can be formulated as a MKP, for example, cargo loadng, cuttng stock problems. MKP s a constraned optmzaton problem, whch means that there exsts llegal soluton space. In our tests, we dn t try to gnore or avo llegal soluton space, nstead, we add a penalty term nto the ftness functon of

8 8 Table 1. FPSBPSO Based Feature Selecton FPSBPSO Based Feature Selecton 1: Intalze parameters of BPSO 2: Randomly ntalze swarm 3: WHILE stoppng crteron not met DO 4: calculate each partcle s ftness functon 6: For = 1 to swarmsze DO 7: update the lbest of P 8: update the pbest of P 9: END 10: FOR = 1 to swarmsze DO 11: FOR j = 1 to dmenson DO 12: update the velocty of P accordng 13: to equaton (5) 14: update the poston of P accordng 15: to equaton (6) 16: END 17: END 18: END 19: Return the best feature subset found by the swarm FPSBPSO. The fnal ftness functon we used was proposed by Khur et al. n [8]. n f(x) = p x s maxp (7) where 1 s = j m w j x > c j (8) 1 Where p s the value of tem, w j s the cost of tem for knapsack j, and c j s the sze of knapsack j. n s the number of tems and m s the number of constrants. The goal s to maxmze n n p x wth the constrans j, w j x <= 1 c j. We used 7 benchmarks named mknap1 n OR-Lbrary [2] to test FPSBPSO s performance on MKP. Optmums of all the 7 benchmarks are all already known. Detaled nformaton of the 7 benchmark problems are gven n Table 2, where n s the number of tems and m s the number of constrants and the column ttled Best shows the known best of the correspondng problem. For standard BPSO, we use the followng parameter settngs: w = , c = , and V max = 4. For FPSBPSO, mr decreases from 0.1 to as teraton process goes on usng formula gven follows. mr = (1 ter/max ter) (9) 1

9 9 Where ter s the current number of teratons and Max ter s the maxmum number of teratons. Max teraton tmes and populaton sze are 1000 and 40 respectvely. In order to reduce the tme cost of tranng, the teraton process wll mmedately stop f the value of global best has kept unchanged for over 120 tmes. For each problem, we run BPSO and FPSBPSO for 20 tmes. We record the average and best value found durng the 20 tmes. Expermental results are gven n Table 3. As can be seen, BPSO succeeded n fndng the best soluton n 4 out of 7 problems whle FPSBPSO succeeded n all 7 problems. Beses, FPSBPSO has the best average performance n most of the problems, and ts advantage over BPSO wll expand as the dmenson of problems ncreases. In a word, FPSBPSO has better performance than BPSO, especally n hgh-dmensonal problems. Table 2. MKP Benchmark n m Best Known n m Best Known Table 3. MKP Results BPSO FPSBPSO Best Avg. Best Avg Feature Selecton wth FPSBPSO Our next goal s to evaluate FPSBPSO s performance n feature selecton doman. To acheve ths, we use the benchmark dataset of Madelon to perform the bnary classfcaton. We use basc BPSO method and FPSBPSO method to get two dfferent feature subsets, then the two subsets are separately used on the dataset to test the classfcaton accuracy. In both cases, the classfcaton algorthm s the CART decson tree. In ths experment, each algorthm s ran for 5 tmes wth dfferent ntalzaton settngs. The expermental results are shown n Table 4 and Table 5. From Table 4, we can know that the average accuracy of results by BPSO s From Table 5, we can see the average accuracy of results by FPSBPSO s , whch s over that by BPSO by 3%. The accuracy mproves much. However, It may take

10 10 more teraton tme for FPSBPSO to return the feature selecton result, whch can be explaned as that FPSBPSO s more precous than tradtonal BPSO and can avo the premature convergence. Table 4. Expermental results by BPSO tme accuracy teraton tmes average Table 5. Expermental results by FPSBPSO tme accuracy teraton tmes average Conclusons and Future Work In ths paper, we dscussed the feasblty of utlzng PSO based methods to select feature subset for classfcaton problems. Frst, we nvestgated the detal of tradtonal BPSO model and some of ts varatons, from whch we argue that BPSO s not a good bnary verson of PSO as t doesn t follow the essence of PSO n some cases. By analysng two major problems PSO wll face n solvng dscrete problems, we proposed a new bnary verson of PSO called FPSBPSO. We redefned the velocty component n FPSBPSO to make t sut the character of PSO better and more computatonally effcent. Expermental results showed that the feature selecton result from our proposed FPSBPSO based method s superor to the one from tradtonal BPSO, as the former one can reach better classfcaton results than the latter one. Although prelmnary results show that the new model can prove sol performance, there are stll many problems to be dealt wth. Frstly, n the proposed

11 11 scheme we used only three partcles nformaton to calculate the average performance of 1 and 0 at a specfc bt, t remans to be seen f more partcles nformaton could mprove the model s performance. Secondly, we have only e- valuated the model s performance on only one benchmark dataset, whch s not suffcent enough. More experments need to be taken to prove the valty and stablty of the FPSBPSO based feature selecton method. In the future, we wll manly focus on the two problems. References 1. A. Abbas, H. Chen, and A. Salem. Sentment analyss n multple languages: Feature selecton for opnon classfcaton n web forums. ACM Transactons on Informaton Systems (TOIS), 26(3):12, J. E. Beasley. Or-lbrary: dstrbutng test problems by electronc mal. Journal of the Operatonal Research Socety, 41(11): , R. Eberhart, P. Smpson, and R. Dobbns. Computatonal ntellgence PC tools. Academc Press Professonal, Inc., San Dego, CA, USA, R. Hassan, B. Cohanm, and O. de Weck. A comparson of partcle swarm optmzaton and the genetc algorthm. In Proceedngs of the 1st AIAA multscplnary desgn optmzaton specalst conference, Aprl J. Kennedy. Bare bones partcle swarms. In Proceedngs of IEEE Swarm Intellgence Symposum, pages 80 87, Aprl J. Kennedy and R. Eberhart. Partcle swarm optmzaton. In Proceedngs of IEEE Internatonal Conference on Neural Networks, volume 4, pages , J. Kennedy and R. Eberhart. A dscrete bnary verson of the partcle swarm optmzaton. In Proceedngs of IEEE Internatonal Conference on Systems, Man, and Cybernetcs, Computatonal Cybernetcs and Smulaton, volume 5, pages , October S. Khur, T. Back, and J. Hetkotter. The zero/one multple knapsack problem and genetc algorthms. In Proceedngs of the 1994 ACM Symposum on Appled Computng, pages , Aprl C.-J. Lao, C.-T. Tseng, and P. Luarn. A dscrete verson of partcle swarm optmzaton for flowshop schedulng problems. Computers & Operatons Research, 34(10): , X. Lu and L. Shang. A fast wrapper feature subset selecton method based on bnary partcle swarm optmzaton. In Proceedngs of IEEE Congress on Evolutonary Computaton, pages , June X. Sh, Y. Lang, H. Lee, C. Lu, and Q. Wang. Partcle swarm optmzaton-based algorthms for tsp and generalzed tsp. Informaton Processng Letters, 103(5): , March Y. Sh and R. Eberhart. A modfed partcle swarm optmzer. In Proceedngs of IEEE World Congress on Computatonal Intellgence, pages 69 73, May M. F. Tasgetren and Y.-C. Lang. A bnary partcle swarm optmzaton algorthm for lot szng problem. Journal of Economc and Socal Research, 5(2):1 20, I. C. Trelea. The partcle swarm optmzaton algorthm: convergence analyss and parameter selecton. Inf. Process. Lett., 85(6): , B. Xue, M. Zhang, and W. N. Browne. Novel ntalsaton and updatng mechansms n pso for feature selecton n classfcaton. In EvoApplcatons, pages , 2013.

The Development of Web Log Mining Based on Improve-K-Means Clustering Analysis

The Development of Web Log Mining Based on Improve-K-Means Clustering Analysis The Development of Web Log Mnng Based on Improve-K-Means Clusterng Analyss TngZhong Wang * College of Informaton Technology, Luoyang Normal Unversty, Luoyang, 471022, Chna wangtngzhong2@sna.cn Abstract.

More information

A Binary Quantum-behaved Particle Swarm Optimization Algorithm with Cooperative Approach

A Binary Quantum-behaved Particle Swarm Optimization Algorithm with Cooperative Approach IJCSI Internatonal Journal of Computer Scence Issues, Vol., Issue, No, January 3 ISSN (Prnt): 694-784 ISSN (Onlne): 694-84 www.ijcsi.org A Bnary Quantum-behave Partcle Swarm Optmzaton Algorthm wth Cooperatve

More information

SOLVING CARDINALITY CONSTRAINED PORTFOLIO OPTIMIZATION PROBLEM BY BINARY PARTICLE SWARM OPTIMIZATION ALGORITHM

SOLVING CARDINALITY CONSTRAINED PORTFOLIO OPTIMIZATION PROBLEM BY BINARY PARTICLE SWARM OPTIMIZATION ALGORITHM SOLVIG CARDIALITY COSTRAIED PORTFOLIO OPTIMIZATIO PROBLEM BY BIARY PARTICLE SWARM OPTIMIZATIO ALGORITHM Aleš Kresta Klíčová slova: optmalzace portfola, bnární algortmus rojení částc Key words: portfolo

More information

Patterns Antennas Arrays Synthesis Based on Adaptive Particle Swarm Optimization and Genetic Algorithms

Patterns Antennas Arrays Synthesis Based on Adaptive Particle Swarm Optimization and Genetic Algorithms IJCSI Internatonal Journal of Computer Scence Issues, Vol. 1, Issue 1, No 2, January 213 ISSN (Prnt): 1694-784 ISSN (Onlne): 1694-814 www.ijcsi.org 21 Patterns Antennas Arrays Synthess Based on Adaptve

More information

A hybrid global optimization algorithm based on parallel chaos optimization and outlook algorithm

A hybrid global optimization algorithm based on parallel chaos optimization and outlook algorithm Avalable onlne www.ocpr.com Journal of Chemcal and Pharmaceutcal Research, 2014, 6(7):1884-1889 Research Artcle ISSN : 0975-7384 CODEN(USA) : JCPRC5 A hybrd global optmzaton algorthm based on parallel

More information

Investigation of Modified Bee Colony Algorithm with Particle and Chaos Theory

Investigation of Modified Bee Colony Algorithm with Particle and Chaos Theory Internatonal Journal of Control and Automaton, pp. 311-3 http://dx.do.org/10.1457/jca.015.8..30 Investgaton of Modfed Bee Colony Algorthm wth Partcle and Chaos Theory Guo Cheng Shangluo College, Zhangye,

More information

Intelligent Method for Cloud Task Scheduling Based on Particle Swarm Optimization Algorithm

Intelligent Method for Cloud Task Scheduling Based on Particle Swarm Optimization Algorithm Unversty of Nzwa, Oman December 9-11, 2014 Page 39 THE INTERNATIONAL ARAB CONFERENCE ON INFORMATION TECHNOLOGY (ACIT2014) Intellgent Method for Cloud Task Schedulng Based on Partcle Swarm Optmzaton Algorthm

More information

Document Clustering Analysis Based on Hybrid PSO+K-means Algorithm

Document Clustering Analysis Based on Hybrid PSO+K-means Algorithm Document Clusterng Analyss Based on Hybrd PSO+K-means Algorthm Xaohu Cu, Thomas E. Potok Appled Software Engneerng Research Group, Computatonal Scences and Engneerng Dvson, Oak Rdge Natonal Laboratory,

More information

A Binary Particle Swarm Optimization Algorithm for Lot Sizing Problem

A Binary Particle Swarm Optimization Algorithm for Lot Sizing Problem Journal o Economc and Socal Research 5 (2), -2 A Bnary Partcle Swarm Optmzaton Algorthm or Lot Szng Problem M. Fath Taşgetren & Yun-Cha Lang Abstract. Ths paper presents a bnary partcle swarm optmzaton

More information

On the Optimal Control of a Cascade of Hydro-Electric Power Stations

On the Optimal Control of a Cascade of Hydro-Electric Power Stations On the Optmal Control of a Cascade of Hydro-Electrc Power Statons M.C.M. Guedes a, A.F. Rbero a, G.V. Smrnov b and S. Vlela c a Department of Mathematcs, School of Scences, Unversty of Porto, Portugal;

More information

Ants Can Schedule Software Projects

Ants Can Schedule Software Projects Ants Can Schedule Software Proects Broderck Crawford 1,2, Rcardo Soto 1,3, Frankln Johnson 4, and Erc Monfroy 5 1 Pontfca Unversdad Católca de Valparaíso, Chle FrstName.Name@ucv.cl 2 Unversdad Fns Terrae,

More information

An Interest-Oriented Network Evolution Mechanism for Online Communities

An Interest-Oriented Network Evolution Mechanism for Online Communities An Interest-Orented Network Evoluton Mechansm for Onlne Communtes Cahong Sun and Xaopng Yang School of Informaton, Renmn Unversty of Chna, Bejng 100872, P.R. Chna {chsun,yang}@ruc.edu.cn Abstract. Onlne

More information

A COPMARISON OF PARTICLE SWARM OPTIMIZATION AND THE GENETIC ALGORITHM

A COPMARISON OF PARTICLE SWARM OPTIMIZATION AND THE GENETIC ALGORITHM A COPMARISON OF PARTICLE SWARM OPTIMIZATION AND THE GENETIC ALGORITHM Rana Hassan * Babak Cohanm Olver de Weck Massachusetts Insttute of Technology, Cambrdge, MA, 39 Gerhard Venter Vanderplaats Research

More information

Module 2 LOSSLESS IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur

Module 2 LOSSLESS IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur Module LOSSLESS IMAGE COMPRESSION SYSTEMS Lesson 3 Lossless Compresson: Huffman Codng Instructonal Objectves At the end of ths lesson, the students should be able to:. Defne and measure source entropy..

More information

A New Task Scheduling Algorithm Based on Improved Genetic Algorithm

A New Task Scheduling Algorithm Based on Improved Genetic Algorithm A New Task Schedulng Algorthm Based on Improved Genetc Algorthm n Cloud Computng Envronment Congcong Xong, Long Feng, Lxan Chen A New Task Schedulng Algorthm Based on Improved Genetc Algorthm n Cloud Computng

More information

Feature selection for intrusion detection. Slobodan Petrović NISlab, Gjøvik University College

Feature selection for intrusion detection. Slobodan Petrović NISlab, Gjøvik University College Feature selecton for ntruson detecton Slobodan Petrovć NISlab, Gjøvk Unversty College Contents The feature selecton problem Intruson detecton Traffc features relevant for IDS The CFS measure The mrmr measure

More information

ECE544NA Final Project: Robust Machine Learning Hardware via Classifier Ensemble

ECE544NA Final Project: Robust Machine Learning Hardware via Classifier Ensemble 1 ECE544NA Fnal Project: Robust Machne Learnng Hardware va Classfer Ensemble Sa Zhang, szhang12@llnos.edu Dept. of Electr. & Comput. Eng., Unv. of Illnos at Urbana-Champagn, Urbana, IL, USA Abstract In

More information

IMPACT ANALYSIS OF A CELLULAR PHONE

IMPACT ANALYSIS OF A CELLULAR PHONE 4 th ASA & μeta Internatonal Conference IMPACT AALYSIS OF A CELLULAR PHOE We Lu, 2 Hongy L Bejng FEAonlne Engneerng Co.,Ltd. Bejng, Chna ABSTRACT Drop test smulaton plays an mportant role n nvestgatng

More information

Improved SVM in Cloud Computing Information Mining

Improved SVM in Cloud Computing Information Mining Internatonal Journal of Grd Dstrbuton Computng Vol.8, No.1 (015), pp.33-40 http://dx.do.org/10.1457/jgdc.015.8.1.04 Improved n Cloud Computng Informaton Mnng Lvshuhong (ZhengDe polytechnc college JangSu

More information

Gender Classification for Real-Time Audience Analysis System

Gender Classification for Real-Time Audience Analysis System Gender Classfcaton for Real-Tme Audence Analyss System Vladmr Khryashchev, Lev Shmaglt, Andrey Shemyakov, Anton Lebedev Yaroslavl State Unversty Yaroslavl, Russa vhr@yandex.ru, shmaglt_lev@yahoo.com, andrey.shemakov@gmal.com,

More information

SCHEDULING OF CONSTRUCTION PROJECTS BY MEANS OF EVOLUTIONARY ALGORITHMS

SCHEDULING OF CONSTRUCTION PROJECTS BY MEANS OF EVOLUTIONARY ALGORITHMS SCHEDULING OF CONSTRUCTION PROJECTS BY MEANS OF EVOLUTIONARY ALGORITHMS Magdalena Rogalska 1, Wocech Bożeko 2,Zdzsław Heduck 3, 1 Lubln Unversty of Technology, 2- Lubln, Nadbystrzycka 4., Poland. E-mal:rogalska@akropols.pol.lubln.pl

More information

"Research Note" APPLICATION OF CHARGE SIMULATION METHOD TO ELECTRIC FIELD CALCULATION IN THE POWER CABLES *

Research Note APPLICATION OF CHARGE SIMULATION METHOD TO ELECTRIC FIELD CALCULATION IN THE POWER CABLES * Iranan Journal of Scence & Technology, Transacton B, Engneerng, ol. 30, No. B6, 789-794 rnted n The Islamc Republc of Iran, 006 Shraz Unversty "Research Note" ALICATION OF CHARGE SIMULATION METHOD TO ELECTRIC

More information

Face Verification Problem. Face Recognition Problem. Application: Access Control. Biometric Authentication. Face Verification (1:1 matching)

Face Verification Problem. Face Recognition Problem. Application: Access Control. Biometric Authentication. Face Verification (1:1 matching) Face Recognton Problem Face Verfcaton Problem Face Verfcaton (1:1 matchng) Querymage face query Face Recognton (1:N matchng) database Applcaton: Access Control www.vsage.com www.vsoncs.com Bometrc Authentcaton

More information

What is Candidate Sampling

What is Candidate Sampling What s Canddate Samplng Say we have a multclass or mult label problem where each tranng example ( x, T ) conssts of a context x a small (mult)set of target classes T out of a large unverse L of possble

More information

Study on Model of Risks Assessment of Standard Operation in Rural Power Network

Study on Model of Risks Assessment of Standard Operation in Rural Power Network Study on Model of Rsks Assessment of Standard Operaton n Rural Power Network Qngj L 1, Tao Yang 2 1 Qngj L, College of Informaton and Electrcal Engneerng, Shenyang Agrculture Unversty, Shenyang 110866,

More information

The Greedy Method. Introduction. 0/1 Knapsack Problem

The Greedy Method. Introduction. 0/1 Knapsack Problem The Greedy Method Introducton We have completed data structures. We now are gong to look at algorthm desgn methods. Often we are lookng at optmzaton problems whose performance s exponental. For an optmzaton

More information

A Genetic Programming Based Stock Price Predictor together with Mean-Variance Based Sell/Buy Actions

A Genetic Programming Based Stock Price Predictor together with Mean-Variance Based Sell/Buy Actions Proceedngs of the World Congress on Engneerng 28 Vol II WCE 28, July 2-4, 28, London, U.K. A Genetc Programmng Based Stock Prce Predctor together wth Mean-Varance Based Sell/Buy Actons Ramn Rajaboun and

More information

Luby s Alg. for Maximal Independent Sets using Pairwise Independence

Luby s Alg. for Maximal Independent Sets using Pairwise Independence Lecture Notes for Randomzed Algorthms Luby s Alg. for Maxmal Independent Sets usng Parwse Independence Last Updated by Erc Vgoda on February, 006 8. Maxmal Independent Sets For a graph G = (V, E), an ndependent

More information

Project Networks With Mixed-Time Constraints

Project Networks With Mixed-Time Constraints Project Networs Wth Mxed-Tme Constrants L Caccetta and B Wattananon Western Australan Centre of Excellence n Industral Optmsaton (WACEIO) Curtn Unversty of Technology GPO Box U1987 Perth Western Australa

More information

The OC Curve of Attribute Acceptance Plans

The OC Curve of Attribute Acceptance Plans The OC Curve of Attrbute Acceptance Plans The Operatng Characterstc (OC) curve descrbes the probablty of acceptng a lot as a functon of the lot s qualty. Fgure 1 shows a typcal OC Curve. 10 8 6 4 1 3 4

More information

Single and multiple stage classifiers implementing logistic discrimination

Single and multiple stage classifiers implementing logistic discrimination Sngle and multple stage classfers mplementng logstc dscrmnaton Hélo Radke Bttencourt 1 Dens Alter de Olvera Moraes 2 Vctor Haertel 2 1 Pontfíca Unversdade Católca do Ro Grande do Sul - PUCRS Av. Ipranga,

More information

Sciences Shenyang, Shenyang, China.

Sciences Shenyang, Shenyang, China. Advanced Materals Research Vols. 314-316 (2011) pp 1315-1320 (2011) Trans Tech Publcatons, Swtzerland do:10.4028/www.scentfc.net/amr.314-316.1315 Solvng the Two-Obectve Shop Schedulng Problem n MTO Manufacturng

More information

Software project management with GAs

Software project management with GAs Informaton Scences 177 (27) 238 241 www.elsever.com/locate/ns Software project management wth GAs Enrque Alba *, J. Francsco Chcano Unversty of Málaga, Grupo GISUM, Departamento de Lenguajes y Cencas de

More information

Recurrence. 1 Definitions and main statements

Recurrence. 1 Definitions and main statements Recurrence 1 Defntons and man statements Let X n, n = 0, 1, 2,... be a MC wth the state space S = (1, 2,...), transton probabltes p j = P {X n+1 = j X n = }, and the transton matrx P = (p j ),j S def.

More information

Global Optimization Algorithms with Application to Non-Life Insurance

Global Optimization Algorithms with Application to Non-Life Insurance Global Optmzaton Algorthms wth Applcaton to Non-Lfe Insurance Problems Ralf Kellner Workng Paper Char for Insurance Economcs Fredrch-Alexander-Unversty of Erlangen-Nürnberg Verson: June 202 GLOBAL OPTIMIZATION

More information

Logistic Regression. Lecture 4: More classifiers and classes. Logistic regression. Adaboost. Optimization. Multiple class classification

Logistic Regression. Lecture 4: More classifiers and classes. Logistic regression. Adaboost. Optimization. Multiple class classification Lecture 4: More classfers and classes C4B Machne Learnng Hlary 20 A. Zsserman Logstc regresson Loss functons revsted Adaboost Loss functons revsted Optmzaton Multple class classfcaton Logstc Regresson

More information

Forecasting the Demand of Emergency Supplies: Based on the CBR Theory and BP Neural Network

Forecasting the Demand of Emergency Supplies: Based on the CBR Theory and BP Neural Network 700 Proceedngs of the 8th Internatonal Conference on Innovaton & Management Forecastng the Demand of Emergency Supples: Based on the CBR Theory and BP Neural Network Fu Deqang, Lu Yun, L Changbng School

More information

Optimal Choice of Random Variables in D-ITG Traffic Generating Tool using Evolutionary Algorithms

Optimal Choice of Random Variables in D-ITG Traffic Generating Tool using Evolutionary Algorithms Optmal Choce of Random Varables n D-ITG Traffc Generatng Tool usng Evolutonary Algorthms M. R. Mosav* (C.A.), F. Farab* and S. Karam* Abstract: Impressve development of computer networks has been requred

More information

A Load-Balancing Algorithm for Cluster-based Multi-core Web Servers

A Load-Balancing Algorithm for Cluster-based Multi-core Web Servers Journal of Computatonal Informaton Systems 7: 13 (2011) 4740-4747 Avalable at http://www.jofcs.com A Load-Balancng Algorthm for Cluster-based Mult-core Web Servers Guohua YOU, Yng ZHAO College of Informaton

More information

DEFINING %COMPLETE IN MICROSOFT PROJECT

DEFINING %COMPLETE IN MICROSOFT PROJECT CelersSystems DEFINING %COMPLETE IN MICROSOFT PROJECT PREPARED BY James E Aksel, PMP, PMI-SP, MVP For Addtonal Informaton about Earned Value Management Systems and reportng, please contact: CelersSystems,

More information

Calculation of Sampling Weights

Calculation of Sampling Weights Perre Foy Statstcs Canada 4 Calculaton of Samplng Weghts 4.1 OVERVIEW The basc sample desgn used n TIMSS Populatons 1 and 2 was a two-stage stratfed cluster desgn. 1 The frst stage conssted of a sample

More information

8.5 UNITARY AND HERMITIAN MATRICES. The conjugate transpose of a complex matrix A, denoted by A*, is given by

8.5 UNITARY AND HERMITIAN MATRICES. The conjugate transpose of a complex matrix A, denoted by A*, is given by 6 CHAPTER 8 COMPLEX VECTOR SPACES 5. Fnd the kernel of the lnear transformaton gven n Exercse 5. In Exercses 55 and 56, fnd the mage of v, for the ndcated composton, where and are gven by the followng

More information

How Sets of Coherent Probabilities May Serve as Models for Degrees of Incoherence

How Sets of Coherent Probabilities May Serve as Models for Degrees of Incoherence 1 st Internatonal Symposum on Imprecse Probabltes and Ther Applcatons, Ghent, Belgum, 29 June 2 July 1999 How Sets of Coherent Probabltes May Serve as Models for Degrees of Incoherence Mar J. Schervsh

More information

LSSVM-ABC Algorithm for Stock Price prediction Osman Hegazy 1, Omar S. Soliman 2 and Mustafa Abdul Salam 3

LSSVM-ABC Algorithm for Stock Price prediction Osman Hegazy 1, Omar S. Soliman 2 and Mustafa Abdul Salam 3 LSSVM-ABC Algorthm for Stock Prce predcton Osman Hegazy 1, Omar S. Solman 2 and Mustafa Abdul Salam 3 1, 2 (Faculty of Computers and Informatcs, Caro Unversty, Egypt) 3 (Hgher echnologcal Insttute (H..I),

More information

Ant Colony Optimization for Economic Generator Scheduling and Load Dispatch

Ant Colony Optimization for Economic Generator Scheduling and Load Dispatch Proceedngs of the th WSEAS Int. Conf. on EVOLUTIONARY COMPUTING, Lsbon, Portugal, June 1-18, 5 (pp17-175) Ant Colony Optmzaton for Economc Generator Schedulng and Load Dspatch K. S. Swarup Abstract Feasblty

More information

CS 2750 Machine Learning. Lecture 3. Density estimation. CS 2750 Machine Learning. Announcements

CS 2750 Machine Learning. Lecture 3. Density estimation. CS 2750 Machine Learning. Announcements Lecture 3 Densty estmaton Mlos Hauskrecht mlos@cs.ptt.edu 5329 Sennott Square Next lecture: Matlab tutoral Announcements Rules for attendng the class: Regstered for credt Regstered for audt (only f there

More information

An ACO Algorithm for. the Graph Coloring Problem

An ACO Algorithm for. the Graph Coloring Problem Int. J. Contemp. Math. Scences, Vol. 3, 2008, no. 6, 293-304 An ACO Algorthm for the Graph Colorng Problem Ehsan Salar and Kourosh Eshgh Department of Industral Engneerng Sharf Unversty of Technology,

More information

Performance Management and Evaluation Research to University Students

Performance Management and Evaluation Research to University Students 631 A publcaton of CHEMICAL ENGINEERING TRANSACTIONS VOL. 46, 2015 Guest Edtors: Peyu Ren, Yancang L, Hupng Song Copyrght 2015, AIDIC Servz S.r.l., ISBN 978-88-95608-37-2; ISSN 2283-9216 The Italan Assocaton

More information

A Performance Analysis of View Maintenance Techniques for Data Warehouses

A Performance Analysis of View Maintenance Techniques for Data Warehouses A Performance Analyss of Vew Mantenance Technques for Data Warehouses Xng Wang Dell Computer Corporaton Round Roc, Texas Le Gruenwald The nversty of Olahoma School of Computer Scence orman, OK 739 Guangtao

More information

Optimal Provisioning of Resource in a Cloud Service

Optimal Provisioning of Resource in a Cloud Service ISSN (Onlne): 169-081 95 Optmal Provsonng of Resource n a Cloud Servce Yee Mng Chen 1 Shn-Yng Tsa Department of Industral Engneerng and Management Yuan Ze Unversty 135 Yuan-Tung Rd. Chung-L Tao-Yuan Tawan

More information

1 Example 1: Axis-aligned rectangles

1 Example 1: Axis-aligned rectangles COS 511: Theoretcal Machne Learnng Lecturer: Rob Schapre Lecture # 6 Scrbe: Aaron Schld February 21, 2013 Last class, we dscussed an analogue for Occam s Razor for nfnte hypothess spaces that, n conjuncton

More information

Mooring Pattern Optimization using Genetic Algorithms

Mooring Pattern Optimization using Genetic Algorithms 6th World Congresses of Structural and Multdscplnary Optmzaton Ro de Janero, 30 May - 03 June 005, Brazl Moorng Pattern Optmzaton usng Genetc Algorthms Alonso J. Juvnao Carbono, Ivan F. M. Menezes Luz

More information

Maintenance Scheduling by using the Bi-Criterion Algorithm of Preferential Anti-Pheromone

Maintenance Scheduling by using the Bi-Criterion Algorithm of Preferential Anti-Pheromone Leonardo ournal of Scences ISSN 583-0233 Issue 2, anuary-une 2008 p. 43-64 Mantenance Schedulng by usng the B-Crteron Algorthm of Preferental Ant-Pheromone Trantafyllos MYTAKIDIS and Arstds VLACHOS Department

More information

INVESTIGATION OF VEHICULAR USERS FAIRNESS IN CDMA-HDR NETWORKS

INVESTIGATION OF VEHICULAR USERS FAIRNESS IN CDMA-HDR NETWORKS 21 22 September 2007, BULGARIA 119 Proceedngs of the Internatonal Conference on Informaton Technologes (InfoTech-2007) 21 st 22 nd September 2007, Bulgara vol. 2 INVESTIGATION OF VEHICULAR USERS FAIRNESS

More information

LITERATURE REVIEW: VARIOUS PRIORITY BASED TASK SCHEDULING ALGORITHMS IN CLOUD COMPUTING

LITERATURE REVIEW: VARIOUS PRIORITY BASED TASK SCHEDULING ALGORITHMS IN CLOUD COMPUTING LITERATURE REVIEW: VARIOUS PRIORITY BASED TASK SCHEDULING ALGORITHMS IN CLOUD COMPUTING 1 MS. POOJA.P.VASANI, 2 MR. NISHANT.S. SANGHANI 1 M.Tech. [Software Systems] Student, Patel College of Scence and

More information

Research Article Enhanced Two-Step Method via Relaxed Order of α-satisfactory Degrees for Fuzzy Multiobjective Optimization

Research Article Enhanced Two-Step Method via Relaxed Order of α-satisfactory Degrees for Fuzzy Multiobjective Optimization Hndaw Publshng Corporaton Mathematcal Problems n Engneerng Artcle ID 867836 pages http://dxdoorg/055/204/867836 Research Artcle Enhanced Two-Step Method va Relaxed Order of α-satsfactory Degrees for Fuzzy

More information

Data Broadcast on a Multi-System Heterogeneous Overlayed Wireless Network *

Data Broadcast on a Multi-System Heterogeneous Overlayed Wireless Network * JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 24, 819-840 (2008) Data Broadcast on a Mult-System Heterogeneous Overlayed Wreless Network * Department of Computer Scence Natonal Chao Tung Unversty Hsnchu,

More information

Mining Feature Importance: Applying Evolutionary Algorithms within a Web-based Educational System

Mining Feature Importance: Applying Evolutionary Algorithms within a Web-based Educational System Mnng Feature Importance: Applyng Evolutonary Algorthms wthn a Web-based Educatonal System Behrouz MINAEI-BIDGOLI 1, and Gerd KORTEMEYER 2, and Wllam F. PUNCH 1 1 Genetc Algorthms Research and Applcatons

More information

Learning with Imperfections A Multi-Agent Neural-Genetic Trading System. with Differing Levels of Social Learning

Learning with Imperfections A Multi-Agent Neural-Genetic Trading System. with Differing Levels of Social Learning Proceedngs of the 4 IEEE Conference on Cybernetcs and Intellgent Systems Sngapore, 1-3 December, 4 Learnng wth Imperfectons A Mult-Agent Neural-Genetc Tradng System wth Dfferng Levels of Socal Learnng

More information

A Hierarchical Anomaly Network Intrusion Detection System using Neural Network Classification

A Hierarchical Anomaly Network Intrusion Detection System using Neural Network Classification IDC IDC A Herarchcal Anomaly Network Intruson Detecton System usng Neural Network Classfcaton ZHENG ZHANG, JUN LI, C. N. MANIKOPOULOS, JAY JORGENSON and JOSE UCLES ECE Department, New Jersey Inst. of Tech.,

More information

An Analysis of Central Processor Scheduling in Multiprogrammed Computer Systems

An Analysis of Central Processor Scheduling in Multiprogrammed Computer Systems STAN-CS-73-355 I SU-SE-73-013 An Analyss of Central Processor Schedulng n Multprogrammed Computer Systems (Dgest Edton) by Thomas G. Prce October 1972 Techncal Report No. 57 Reproducton n whole or n part

More information

Laddered Multilevel DC/AC Inverters used in Solar Panel Energy Systems

Laddered Multilevel DC/AC Inverters used in Solar Panel Energy Systems Proceedngs of the nd Internatonal Conference on Computer Scence and Electroncs Engneerng (ICCSEE 03) Laddered Multlevel DC/AC Inverters used n Solar Panel Energy Systems Fang Ln Luo, Senor Member IEEE

More information

Forecasting the Direction and Strength of Stock Market Movement

Forecasting the Direction and Strength of Stock Market Movement Forecastng the Drecton and Strength of Stock Market Movement Jngwe Chen Mng Chen Nan Ye cjngwe@stanford.edu mchen5@stanford.edu nanye@stanford.edu Abstract - Stock market s one of the most complcated systems

More information

Can Auto Liability Insurance Purchases Signal Risk Attitude?

Can Auto Liability Insurance Purchases Signal Risk Attitude? Internatonal Journal of Busness and Economcs, 2011, Vol. 10, No. 2, 159-164 Can Auto Lablty Insurance Purchases Sgnal Rsk Atttude? Chu-Shu L Department of Internatonal Busness, Asa Unversty, Tawan Sheng-Chang

More information

A Novel Methodology of Working Capital Management for Large. Public Constructions by Using Fuzzy S-curve Regression

A Novel Methodology of Working Capital Management for Large. Public Constructions by Using Fuzzy S-curve Regression Novel Methodology of Workng Captal Management for Large Publc Constructons by Usng Fuzzy S-curve Regresson Cheng-Wu Chen, Morrs H. L. Wang and Tng-Ya Hseh Department of Cvl Engneerng, Natonal Central Unversty,

More information

Vision Mouse. Saurabh Sarkar a* University of Cincinnati, Cincinnati, USA ABSTRACT 1. INTRODUCTION

Vision Mouse. Saurabh Sarkar a* University of Cincinnati, Cincinnati, USA ABSTRACT 1. INTRODUCTION Vson Mouse Saurabh Sarkar a* a Unversty of Cncnnat, Cncnnat, USA ABSTRACT The report dscusses a vson based approach towards trackng of eyes and fngers. The report descrbes the process of locatng the possble

More information

Fault tolerance in cloud technologies presented as a service

Fault tolerance in cloud technologies presented as a service Internatonal Scentfc Conference Computer Scence 2015 Pavel Dzhunev, PhD student Fault tolerance n cloud technologes presented as a servce INTRODUCTION Improvements n technques for vrtualzaton and performance

More information

Vehicle Routing Problem with Time Windows for Reducing Fuel Consumption

Vehicle Routing Problem with Time Windows for Reducing Fuel Consumption 3020 JOURNAL OF COMPUTERS, VOL. 7, NO. 12, DECEMBER 2012 Vehcle Routng Problem wth Tme Wndows for Reducng Fuel Consumpton Jn L School of Computer and Informaton Engneerng, Zhejang Gongshang Unversty, Hangzhou,

More information

Mining Multiple Large Data Sources

Mining Multiple Large Data Sources The Internatonal Arab Journal of Informaton Technology, Vol. 7, No. 3, July 2 24 Mnng Multple Large Data Sources Anmesh Adhkar, Pralhad Ramachandrarao 2, Bhanu Prasad 3, and Jhml Adhkar 4 Department of

More information

Joint Scheduling of Processing and Shuffle Phases in MapReduce Systems

Joint Scheduling of Processing and Shuffle Phases in MapReduce Systems Jont Schedulng of Processng and Shuffle Phases n MapReduce Systems Fangfe Chen, Mural Kodalam, T. V. Lakshman Department of Computer Scence and Engneerng, The Penn State Unversty Bell Laboratores, Alcatel-Lucent

More information

Institute of Informatics, Faculty of Business and Management, Brno University of Technology,Czech Republic

Institute of Informatics, Faculty of Business and Management, Brno University of Technology,Czech Republic Lagrange Multplers as Quanttatve Indcators n Economcs Ivan Mezník Insttute of Informatcs, Faculty of Busness and Management, Brno Unversty of TechnologCzech Republc Abstract The quanttatve role of Lagrange

More information

Answer: A). There is a flatter IS curve in the high MPC economy. Original LM LM after increase in M. IS curve for low MPC economy

Answer: A). There is a flatter IS curve in the high MPC economy. Original LM LM after increase in M. IS curve for low MPC economy 4.02 Quz Solutons Fall 2004 Multple-Choce Questons (30/00 ponts) Please, crcle the correct answer for each of the followng 0 multple-choce questons. For each queston, only one of the answers s correct.

More information

POLYSA: A Polynomial Algorithm for Non-binary Constraint Satisfaction Problems with and

POLYSA: A Polynomial Algorithm for Non-binary Constraint Satisfaction Problems with and POLYSA: A Polynomal Algorthm for Non-bnary Constrant Satsfacton Problems wth and Mguel A. Saldo, Federco Barber Dpto. Sstemas Informátcos y Computacón Unversdad Poltécnca de Valenca, Camno de Vera s/n

More information

The Load Balancing of Database Allocation in the Cloud

The Load Balancing of Database Allocation in the Cloud , March 3-5, 23, Hong Kong The Load Balancng of Database Allocaton n the Cloud Yu-lung Lo and Mn-Shan La Abstract Each database host n the cloud platform often has to servce more than one database applcaton

More information

Open Access A Load Balancing Strategy with Bandwidth Constraint in Cloud Computing. Jing Deng 1,*, Ping Guo 2, Qi Li 3, Haizhu Chen 1

Open Access A Load Balancing Strategy with Bandwidth Constraint in Cloud Computing. Jing Deng 1,*, Ping Guo 2, Qi Li 3, Haizhu Chen 1 Send Orders for Reprnts to reprnts@benthamscence.ae The Open Cybernetcs & Systemcs Journal, 2014, 8, 115-121 115 Open Access A Load Balancng Strategy wth Bandwdth Constrant n Cloud Computng Jng Deng 1,*,

More information

Conversion between the vector and raster data structures using Fuzzy Geographical Entities

Conversion between the vector and raster data structures using Fuzzy Geographical Entities Converson between the vector and raster data structures usng Fuzzy Geographcal Enttes Cdála Fonte Department of Mathematcs Faculty of Scences and Technology Unversty of Combra, Apartado 38, 3 454 Combra,

More information

Hybrid-Learning Methods for Stock Index Modeling

Hybrid-Learning Methods for Stock Index Modeling Hybrd-Learnng Methods for Stock Index Modelng 63 Chapter IV Hybrd-Learnng Methods for Stock Index Modelng Yuehu Chen, Jnan Unversty, Chna Ajth Abraham, Chung-Ang Unversty, Republc of Korea Abstract The

More information

Optimized ready mixed concrete truck scheduling for uncertain factors using bee algorithm

Optimized ready mixed concrete truck scheduling for uncertain factors using bee algorithm Songklanakarn J. Sc. Technol. 37 (2), 221-230, Mar.-Apr. 2015 http://www.sst.psu.ac.th Orgnal Artcle Optmzed ready mxed concrete truck schedulng for uncertan factors usng bee algorthm Nuntana Mayteekreangkra

More information

Comparison of Weighted Sum Fitness Functions for PSO Optimization of Wideband Medium-gain Antennas

Comparison of Weighted Sum Fitness Functions for PSO Optimization of Wideband Medium-gain Antennas 54 ZHOGKU MA, G. A. E. VAEBOSCH, COMPARISO OF WEIGHTE SUM FITESS FUCTIOS FOR PSO Comparson of Weghted Sum Ftness Functons for PSO Optmzaton of Wdeband Medum-gan Antennas Zhongkun MA, Guy A. E. VAEBOSCH

More information

An Alternative Way to Measure Private Equity Performance

An Alternative Way to Measure Private Equity Performance An Alternatve Way to Measure Prvate Equty Performance Peter Todd Parlux Investment Technology LLC Summary Internal Rate of Return (IRR) s probably the most common way to measure the performance of prvate

More information

Implementation of Deutsch's Algorithm Using Mathcad

Implementation of Deutsch's Algorithm Using Mathcad Implementaton of Deutsch's Algorthm Usng Mathcad Frank Roux The followng s a Mathcad mplementaton of Davd Deutsch's quantum computer prototype as presented on pages - n "Machnes, Logc and Quantum Physcs"

More information

Survey on Virtual Machine Placement Techniques in Cloud Computing Environment

Survey on Virtual Machine Placement Techniques in Cloud Computing Environment Survey on Vrtual Machne Placement Technques n Cloud Computng Envronment Rajeev Kumar Gupta and R. K. Paterya Department of Computer Scence & Engneerng, MANIT, Bhopal, Inda ABSTRACT In tradtonal data center

More information

Enabling P2P One-view Multi-party Video Conferencing

Enabling P2P One-view Multi-party Video Conferencing Enablng P2P One-vew Mult-party Vdeo Conferencng Yongxang Zhao, Yong Lu, Changja Chen, and JanYn Zhang Abstract Mult-Party Vdeo Conferencng (MPVC) facltates realtme group nteracton between users. Whle P2P

More information

Credit Limit Optimization (CLO) for Credit Cards

Credit Limit Optimization (CLO) for Credit Cards Credt Lmt Optmzaton (CLO) for Credt Cards Vay S. Desa CSCC IX, Ednburgh September 8, 2005 Copyrght 2003, SAS Insttute Inc. All rghts reserved. SAS Propretary Agenda Background Tradtonal approaches to credt

More information

Joint Optimization of Bid and Budget Allocation in Sponsored Search

Joint Optimization of Bid and Budget Allocation in Sponsored Search Jont Optmzaton of Bd and Budget Allocaton n Sponsored Search Wenan Zhang Shangha Jao Tong Unversty Shangha, 224, P. R. Chna wnzhang@apex.sjtu.edu.cn Yong Yu Shangha Jao Tong Unversty Shangha, 224, P. R.

More information

Performance Analysis and Coding Strategy of ECOC SVMs

Performance Analysis and Coding Strategy of ECOC SVMs Internatonal Journal of Grd and Dstrbuted Computng Vol.7, No. (04), pp.67-76 http://dx.do.org/0.457/jgdc.04.7..07 Performance Analyss and Codng Strategy of ECOC SVMs Zhgang Yan, and Yuanxuan Yang, School

More information

A DATA MINING APPLICATION IN A STUDENT DATABASE

A DATA MINING APPLICATION IN A STUDENT DATABASE JOURNAL OF AERONAUTICS AND SPACE TECHNOLOGIES JULY 005 VOLUME NUMBER (53-57) A DATA MINING APPLICATION IN A STUDENT DATABASE Şenol Zafer ERDOĞAN Maltepe Ünversty Faculty of Engneerng Büyükbakkalköy-Istanbul

More information

Trust Formation in a C2C Market: Effect of Reputation Management System

Trust Formation in a C2C Market: Effect of Reputation Management System Trust Formaton n a C2C Market: Effect of Reputaton Management System Htosh Yamamoto Unversty of Electro-Communcatons htosh@s.uec.ac.jp Kazunar Ishda Tokyo Unversty of Agrculture k-shda@noda.ac.jp Toshzum

More information

ANALYZING THE RELATIONSHIPS BETWEEN QUALITY, TIME, AND COST IN PROJECT MANAGEMENT DECISION MAKING

ANALYZING THE RELATIONSHIPS BETWEEN QUALITY, TIME, AND COST IN PROJECT MANAGEMENT DECISION MAKING ANALYZING THE RELATIONSHIPS BETWEEN QUALITY, TIME, AND COST IN PROJECT MANAGEMENT DECISION MAKING Matthew J. Lberatore, Department of Management and Operatons, Vllanova Unversty, Vllanova, PA 19085, 610-519-4390,

More information

Support Vector Machines

Support Vector Machines Support Vector Machnes Max Wellng Department of Computer Scence Unversty of Toronto 10 Kng s College Road Toronto, M5S 3G5 Canada wellng@cs.toronto.edu Abstract Ths s a note to explan support vector machnes.

More information

Logical Development Of Vogel s Approximation Method (LD-VAM): An Approach To Find Basic Feasible Solution Of Transportation Problem

Logical Development Of Vogel s Approximation Method (LD-VAM): An Approach To Find Basic Feasible Solution Of Transportation Problem INTERNATIONAL JOURNAL OF SCIENTIFIC & TECHNOLOGY RESEARCH VOLUME, ISSUE, FEBRUARY ISSN 77-866 Logcal Development Of Vogel s Approxmaton Method (LD- An Approach To Fnd Basc Feasble Soluton Of Transportaton

More information

A Secure Password-Authenticated Key Agreement Using Smart Cards

A Secure Password-Authenticated Key Agreement Using Smart Cards A Secure Password-Authentcated Key Agreement Usng Smart Cards Ka Chan 1, Wen-Chung Kuo 2 and Jn-Chou Cheng 3 1 Department of Computer and Informaton Scence, R.O.C. Mltary Academy, Kaohsung 83059, Tawan,

More information

Lecture 2: Single Layer Perceptrons Kevin Swingler

Lecture 2: Single Layer Perceptrons Kevin Swingler Lecture 2: Sngle Layer Perceptrons Kevn Sngler kms@cs.str.ac.uk Recap: McCulloch-Ptts Neuron Ths vastly smplfed model of real neurons s also knon as a Threshold Logc Unt: W 2 A Y 3 n W n. A set of synapses

More information

CHOLESTEROL REFERENCE METHOD LABORATORY NETWORK. Sample Stability Protocol

CHOLESTEROL REFERENCE METHOD LABORATORY NETWORK. Sample Stability Protocol CHOLESTEROL REFERENCE METHOD LABORATORY NETWORK Sample Stablty Protocol Background The Cholesterol Reference Method Laboratory Network (CRMLN) developed certfcaton protocols for total cholesterol, HDL

More information

J. Parallel Distrib. Comput.

J. Parallel Distrib. Comput. J. Parallel Dstrb. Comput. 71 (2011) 62 76 Contents lsts avalable at ScenceDrect J. Parallel Dstrb. Comput. journal homepage: www.elsever.com/locate/jpdc Optmzng server placement n dstrbuted systems n

More information

A Dynamic Load Balancing for Massive Multiplayer Online Game Server

A Dynamic Load Balancing for Massive Multiplayer Online Game Server A Dynamc Load Balancng for Massve Multplayer Onlne Game Server Jungyoul Lm, Jaeyong Chung, Jnryong Km and Kwanghyun Shm Dgtal Content Research Dvson Electroncs and Telecommuncatons Research Insttute Daejeon,

More information

Risk-based Fatigue Estimate of Deep Water Risers -- Course Project for EM388F: Fracture Mechanics, Spring 2008

Risk-based Fatigue Estimate of Deep Water Risers -- Course Project for EM388F: Fracture Mechanics, Spring 2008 Rsk-based Fatgue Estmate of Deep Water Rsers -- Course Project for EM388F: Fracture Mechancs, Sprng 2008 Chen Sh Department of Cvl, Archtectural, and Envronmental Engneerng The Unversty of Texas at Austn

More information

An Enhanced Super-Resolution System with Improved Image Registration, Automatic Image Selection, and Image Enhancement

An Enhanced Super-Resolution System with Improved Image Registration, Automatic Image Selection, and Image Enhancement An Enhanced Super-Resoluton System wth Improved Image Regstraton, Automatc Image Selecton, and Image Enhancement Yu-Chuan Kuo ( ), Chen-Yu Chen ( ), and Chou-Shann Fuh ( ) Department of Computer Scence

More information

benefit is 2, paid if the policyholder dies within the year, and probability of death within the year is ).

benefit is 2, paid if the policyholder dies within the year, and probability of death within the year is ). REVIEW OF RISK MANAGEMENT CONCEPTS LOSS DISTRIBUTIONS AND INSURANCE Loss and nsurance: When someone s subject to the rsk of ncurrng a fnancal loss, the loss s generally modeled usng a random varable or

More information

Multiple-Period Attribution: Residuals and Compounding

Multiple-Period Attribution: Residuals and Compounding Multple-Perod Attrbuton: Resduals and Compoundng Our revewer gave these authors full marks for dealng wth an ssue that performance measurers and vendors often regard as propretary nformaton. In 1994, Dens

More information