Parallel Learning of Large Fuzzy Cognitive Maps

Size: px
Start display at page:

Download "Parallel Learning of Large Fuzzy Cognitive Maps"

Transcription

1 g Proceedigs of Iteratioal Joit Coferece o Neural Networks, Orlado, Florida, USA, August 2-7, 07 Parallel Learig of Large Fuzzy Cogitive Maps Wojciech Stach, Lukasz Kurga, ad Witold Pedrycz Abstract Fuzzy Cogitive Maps (FCMs) are a class of discrete-time Artificial Neural Networks that are used to model dyamic systems. A recetly itroduced supervised learig method, which is based o real-coded geetic algorithm (RCGA), allows learig high-quality FCMs from historical data. The curret bottleeck of this learig method is its scalability, which origiates from large cotiuous search space (of quadratic size with respect to the size of the FCM) ad computatioal complexity of geetic optimizatio. To this ed, the goal of this paper is to explore parallel ature of geetic algorithms to alleviate the scalability problem. We use the global sigle-populatio master-slave parallelizatio method to speed up the FCMs learig method. We ivestigate the ifluece of differet hardware architectures o the computatioal time of the learig method by executig a wide rage of sythetic ad real-life bechmarkig tests. We aalyze the quality of the proposed parallel learig method i applicatio to both dese ad sparse large FCMs, i.e. maps that cosist of several dozes of cocepts. The parallelizatio is show to provide substatial speed-ups, allowig doublig the size of the FCM that ca be leared by parallelizatio with 8 processors. A. Fuzzy Cogitive Maps I. INTRODUCTION ) Overview roposed i 986 by Bart Kosko [4], Fuzzy Cogitive P Maps (FCMs) form a class of discrete-time Artificial Neural Networks. They represet kowledge i a symbolic maer ad relate states, variables, evets, outputs ad iputs usig a cause ad effect approach. FCMs, whe compared with eural etworks, have several importat advatages such as relative easiess to represet structured kowledge, ad simplicity of the iferece that is computed by umeric matrix operatios [7]. The FCM structure is similar to a recurret artificial eural etwork (RNN), where cocepts are represeted by euros ad causal relatioships by weighted edges coectig the euros. However, i cotrast to FCMs, RNN euros have exteral iputs, whereas i FCMs, the (euros) are oly iterally itercoected. Each of FCM s edges is associated with a weight value that reflects the stregth of the correspodig relatio. This value is usually ormalized to the iterval [,]. Positive values This work was supported i part by the Walter Karplus Summer Research Grat, by the Alberta Igeuity, ad by the Natural Scieces & Egieerig Research Coucil of Caada (NSERC) W. Stach, L. Kurga, ad W. Pedrycz are with the Departmet of Electrical ad Computer Egieerig, Uiversity of Alberta, Edmoto, Caada ( {wstach, lkurga, pedrycz}@ece.ualberta.ca) reflect promotig effect, while egative oes describe ihibitig effect. The value of represets full egative, + full positive ad 0 deotes eutral relatio. Other values correspod to differet itermediate levels of causal effect. The graph represetatio is equivalet to a square matrix, called coectio matrix, which stores all weight values of edges betwee correspodig cocepts. Figure shows a example of FCM that models city health issues [6]. N N2 N3 N4 N5 N6 N7 N N N N N N N Fig.. Example of FCM model ad its equivalet coectio matrix I FCMs, each ode has a value that reflects the degree to which the correspodig cocept i the system is active at a particular iteratio. This value, called activatio level, is a floatig-poit umber betwee 0 (iactive) ad (active). For a give cocept, this value is calculated by takig ito accout the activatio levels at the previous iteratio of all the cocepts that exert ifluece o it: N j {,..., N}, C ( t + ) = f e C () j ij i i= where: C j (t) activatio level of cocept j th at iteratio t e ij stregth of relatio from cocept C i to cocept C j f trasformatio fuctio The trasformatio fuctio is used to reduce ubouded weighted sum to a certai rage. The ormalizatio hiders quatitative aalysis, but, at the same time, it allows for comparisos betwee activatio levels of differet cocepts. A sapshot of activatio levels of all at a particular iteratio defies the system state. It ca be coveietly represeted by a vector, called state vector, which cosists of the activatio values. Iitial state vector refers to the system state at the first iteratio. Successive states are calculated by iterative applicatio of the formula (). Fuzzy Cogitive Maps have bee successfully applied i various domais, icludig egieerig [2][23], medicie [], political sciece [2], ecoomics [], Earth ad evirometal scieces [6], etc X/07/$ IEEE

2 2) Developmet There are two mai groups of approaches to develop Fuzzy Cogitive Maps: () deductive modelig (i.e., it uses a expert kowledge from the domai of applicatio); ad (2) iductive modelig (i.e., it uses learig algorithms to establish FCMs from historical data) A comprehesive overview of these methods ca be foud i [22]. Figure 2 shows a high-level diagram of the FCMs learig method based o real-coded geetic algorithms (RCGA), which is used i this paper. This method has bee recetly itroduced ad thoroughly tested [9][]. RCGA is a floatig-poit extesio to geeric geetic algorithms. Sectio B.2 gives more iformatio o both geetic algorithms ad RCGA. Iitialize ad evaluate populatio Stop? Yes No Mutatio New populatio Fig. 2. High-level diagram of RCGA learig method for FCMs Recombiatio Evaluatio Selectio The RCGA learig method uses iput data to develop a FCM (cadidate FCM), which is capable to mimic the historical data [9]. The iput data is give as time series ad cosist of a sequece of state vectors that describe a give system at differet time poits. Sice FCM ca be fully represeted by its coectio matrix, the learig goal is to establish N 2 parameters, where N deotes the umber of cocepts. They correspod to the weighted values of mutual relatios betwee the cocepts. The RCGA algorithm exploits the iput data to fid these values. The learig objective is to geerate the same state vector sequece from the cadidate FCM for the same iitial state vector, as it is defied i the iput data. At the same time, the cadidate FCM geeralizes the iter-relatios betwee cocept, which are iferred from the iput data. Therefore, it is suitable to perform simulatios from differet iitial state vectors ad, based o their results, to draw coclusios about the modeled system. B. Parallelizatio of Geetic Algorithms ) Itroductio to parallel computig Parallel computig is oe of the popular techiques to speed up the process of solvig complex computatioal problems [8]. It assumes simultaeous executio of the same task o multiple processors to obtai the results faster. The uderlyig assumptio is that a problem beig solved ca be divided ito smaller tasks, which ca be executed simultaeously with a certai level of coordiatio. I parallel computig, it is crucial to use parallel algorithms to take advatage of hardware systems [25]. Parallel algorithms, i cotrary to sequetial algorithms, ca be divided ito parts performed i parallel. Subsequetly, the partial results are put back together to obtai the fial result. The task of fidig a efficiet parallel algorithm to solve a give problem ca be very challegig as we ofte deal with sequetial costrais. For istace, recursive solutios that require a result from previous iteratio to calculate result i ext iteratio are very difficult to parallelize. While implemetig parallel algorithms, commuicatio amog the processors to coordiate the executio of subtasks must be cosidered. Hece, i practice, liear time speed-up vs. the umber of processors is very difficult to obtai. The commuicatio is usually realized either usig shared memory or message passig approach [5]. 2) Overview of geetic algorithms Geetic algorithms (GAs) are a search techique to fid solutios to optimizatio ad search problems [5][7]. They have may advatages, which iclude broad applicability, ease of use, ad global perspective, to ame a few. GAs origiate from evolutioary biology ad use geetically ispired mechaisms, such as mutatio, selectio, ad crossover. These operators are applied to a populatio of chromosomes that is maitaied throughout the etire searchig process. Each chromosome represets a solutio to the problem, ad its quality is quatified by a fitess value that is calculated from the fitess fuctio. The GAs usually start from a radomly geerated populatio ad modify it i subsequet geeratios. I each geeratio, a ew populatio is formed by usig geetic operators, ad by exploitig fitess values of the chromosomes. The idea is to produce better solutios to the problem over the evolvig geeratios. GAs ca be successfully applied to large, complex, ad poorly uderstood search spaces, i which classical optimizatio tools are ofte iappropriate. The detailed iformatio about GAs ca be foud i [5][7][24]. Several extesios to the geeric GAs have bee itroduced. I this paper, a real-coded geetic algorithms (RCGA) approach is used as a part of the method for FCMs learig. RCGA is a floatig-poit extesio to GAs. I RCGA chromosomes are represeted as floatig-poit vectors i cotrast to GAs that use biary vectors. This exteded chromosome represetatio makes RCGA more

3 effective to tackle optimizatio problems with cotiuous variables. Although the geetic operators i RCGA are revised i order to deal with floatig-poit values, the mai priciples of both GA ad RCGA are the same. Comprehesive descriptio of the RCGA is give i [9]. The geetic algorithms ca be parallelized i several differet ways that are compatible with the RCGA learig. Parallelizatio ca be implemeted based o a sigle populatio, or by splittig the populatio ito subpopulatios. I this paper, a sigle populatio method (global sigle-populatio master-slave) is chose. This approach parallelizes the evaluatio of fitess fuctio, which is usually the most time-cosumig part of the GA s optimizatio. The implemetatio is usually doe as a master-slave structure, i which the master process stores the populatio ad the slaves evaluate the fitess. With this feature i mid, the etire populatio is split up ito subsets, which are the assiged to differet available processors. Commuicatio betwee the processes occurs oly whe slaves receive a subset of idividuals to evaluate ad whe they retur the fitess values to the master process. This parallelizatio architecture does ot affect the behavior of the geetic algorithm, sice o additioal restrictios are imposed o the geetic operators such as crossover or selectio. This simplicity motivated our selectio of this parallelizatio approach. Additioal iformatio o differet approaches to parallelizatio of geetic algorithms ca be foud i [][3][4][3]. II. MOTIVATION AND METHODS The bottleeck of the RCGA learig method for FCMs is its scalability, as the umber of parameters to be established grows quadratically with the map size (umber of cocepts). I additio, geetic optimizatio is time cosumig whe employed to problems with large umber of variables. As a result, RCGA method has bee applied to FCMs which cosist of up to cocepts. This restrictio substatially limits the applicability of this learig approach. Sice there are o other quality iductive techiques, learig of larger size FCMs from data was ot possible. At the same time, i some areas such as systems biology, the uderlyig etworks that could be modeled with FCMs are large ad cosist of at least several dozes of. Moreover, recetly proposed extesio to geeric FCMs, called higher-order fuzzy cogitive maps [8], requires eve more parameters to be established durig the learig process. These issues call for developmet of learig approaches for FCMs that would be fast eough to be applied to larger systems. Therefore, this paper has two mai goals: to propose a time-efficiet learig method for FCMs based o RCGA approach. To satisfy this goal, we use a parallelized approach to real-coded geetic algorithm i order to achieve boost i executio time. I this paper, the global sigle-populatio master-slave method for parallelizatio of GAs is used due to its simplicity. to test the proposed parallelized RCGA learig method o a set of large ad diverse FCMs to assess accuracy of the developed FCMs ad the amout of obtaied speedup (whe compared with sequetial learig). This paper icludes results of experimets with sythetic FCMs that cosist of,,, ad cocepts. Additioally, the tests are carried out with a large real-life map. III. EVALUATION A. Data Sets Similarly to experimets reported i [9], we used both sythetic ad real-life data i our experimets. ) Sythetic data The sythetic data for our experimets were obtaied by simulatig radomly geerated FCMs from radom iitial vectors. Four groups of experimets have bee performed with maps that cosist of,,, ad cocepts, respectively. Additioally, for each group we geerated two series of data with two differet map desities (defied as the ratio of the o-zero relatios weights to the total umber of weights), % ad %, respectively. Thus, as a result, we applied 8 differet experimetal setups. Additioally, for each setup, 5 idepedet maps were geerated to assure statistical validity of the results. 2) Real-life data We selected oe of the largest FCMs foud i literature to prepare the iput data. I this case, the FCM was predefied by the domai experts ad cocered factors that affected slurry rheology [2]. The iput map icluded 3 cocepts: gravity, mechaical properties of particles, physiochemical iteractio, hydrodyamic iteractio, effective particle cocetratio, particle-particle cotact, liquid viscosity, effective particle shape, effective particle size, temperature, iter-particle attractio, floc/structure, ad shear rate. The actual FCM ca be foud i [2]. Its desity is 38.5%. We geerated the iput data from the iitial vector, which was used to aalyze the map i the origial paper. B. Evaluatio Criteria The evaluatio measures of the proposed method are threefold: executio time time eeded to complete the learig i-sample error differece betwee the iput data, ad data geerated by simulatig the cadidate FCM from the same iitial state vector as for the iput data. The criterio is defied as a ormalized average error betwee correspodig cocept values at each iteratio betwee the two state vector sequeces [9]. K N error _ iitial = C ˆ C (2) ( K ) N t= = where C (t) is the value of a ode at iteratio t i the

4 iput data, Cˆ is the value of a ode at iteratio t from simulatio of the cadidate FCM, K is the iput data legth, ad N is the umber of out-of-sample error evaluatio of the geeralizatio capabilities of the cadidate FCM. To compute this criterio, both the iput model ad the cadidate FCMs are simulated from te radomly chose iitial state vectors. Subsequetly, the error_iitial value is computed for each of the simulatios to compare state vector sequeces geerated by the iput ad the cadidate FCM, ad a average of these values is computed [9]. P K N p p error _ behavior = C Cˆ (3) P ( K ) N p= t = = where C p (t) is the value of a ode at iteratio t for data geerated by iput FCM started from p th iitial state vector, Cˆ p is the value of a ode at iteratio t for data geerated by cadidate FCM started from p th iitial state vector, K is the iput data legth, ad N is the umber of, ad P is the umber of differet iitial state vectors. The first criterio is used to test the speed-up i executio time, whereas the other two are used to evaluate quality of the developed map. They are cosistet with the criteria reported i [9]. C. Experimetal Setup The hardware used to execute the experimets was a state-of-the-art 2-way IBM p570 server with POWER5 processors. The code has bee writte i C++ usig OpeMP, which is a applicatio programmig iterface that supports multi-platform shared memory multiprocessig. We repeated each experimet with differet cofiguratio of our hardware usig the followig sceario. We started by performig simulatio o a sigle processor, ad the, we doubled the umber of processors i the subsequet experimets, up to eight processors. As a result, we carried out four idepedet simulatios for each experimet. IV. RESULTS Table I summarizes the experimetal results for the sythetic data. The reported values have bee calculated as averages obtaied from 5 idepedet experimets (with differet models) for each setup. The colums correspod to differet experimetal setups i terms of maps sizes (,,, ad ) ad desities (% ad %), the rows correspod to differet hardware cofiguratios (the first colum defies the umber of processors), ad the value i each cell expresses the average value of a correspodig criterio followed by the stadard deviatio. The criteria are labeled o the right had side of the table. TABLE I EXPERIMENTAL RESULTS FOR SYNTHETIC DATA # % % % % % % % % ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±0.04 Time [s] I-sample error Out-of-sample error Table II summarizes the experimetal results for the reallife FCM. The reported values have bee calculated as averages obtaied from 5 idepedet experimets for each setup. TABLE II EXPERIMENTAL RESULTS FOR REAL-LIFE DATA Time # [s] 89.6 ± ± ± ±8.62 I-sample error ± ± ± ±0.009 Out-ofsample error 0.66 ± ± ± ±0.059 Aalysis of the results for both sythetic ad real-life data is preseted i the two followig subsectios. A. Executio time Results i Table I show o sigificat ifluece of desity o the executio time. The ifluece of the map s size ad the umber of processors o the executio time is show i Figure 3. The executio time for the two map desities has bee averaged out for each map size. Additioally, the colums are liearly ormalized to the maximum executio time for each setup (which correspods to the sequetial learig). A umber above each bar shows the executio time (i secods).

5 time 0% % 60% % % 0% processor 2 processors 4 processors 8 processors Fig. 3. Executio time vs. experimetal setup I-sample error % % % processor 2 processors 4 processors 8 processors Fig. 4. I-sample error vs. experimetal setup I geeral, parallel learig o eight processors allows doublig the size of the FCM withi virtually the same time that is required for the sequetial learig. Compariso of the results obtaied from simulatios o a sigle processor (traditioal, sequetial learig) to those o two processors shows almost two-fold time-reductio for each setup (time gai varies betwee 43% ad 47%). However, doublig the umber of processors agai does ot result i the correspodig two-fold time reductio. Whe compared with sequetial learig with a sigle processor, the time gai is, o average, 60% ad 69% for four ad eight processors, respectively. This is due to the fact that the executio time of the sequetial costraits of performig geetic operatios o populatio of chromosomes becomes a more sigificat part of the total executio time. This happes as the fitess fuctio evaluatios performed i parallel are completed faster o larger umber of processors. Therefore, i this RCGA parallelizatio approach we observe decreasig returs alog with icreasig the umber of processors. B. Learig quality Figure 4 shows relatio betwee i-sample error ad the size ad desity of FCMs, as well as the umber of processors. A few iterestig coclusios ca be draw from this figure. Firstly, the quality of learig gradually decreases alog with the icreasig iput map size. The isample error value chages from for FCMs with to for FCMs with. This is due to complexity of the optimizatio problem, which drastically icreases with the icrease of the size of maps. However, eve for the largest ivestigated maps the results are still sigificatly better from the baselie of this problem, which has bee experimetally foud at the value of 0.39 (see [9] for details). Secodly, the experimets with differet hardware setups, i.e. differet umber of processors for a certai map size ad desity, are cosistet i terms of the solutio quality. The stadard deviatio of the i-sample error measured for differet umber of processors for a give setup is very small ad varies from ( %) to 0.00 ( %). This is i spite of the fact that the error values differ, which suggests that the learig method fids sub-optimal solutios of similar quality. The last observatio is also cosistet with the coclusios i [9]. Fially, the i-sample error does ot deped o the iput map s desity. For sparser maps, it is oly slightly lower for ad - FCMs, ad slightly higher for ad - FCMs. Figure 5 shows the relatioship betwee out-of-sample error ad the experimetal setup, i.e. the map size ad desity, ad the umber of processors. Out-of-sample error processor 2 processors 4 processors 8 processors 38.5% % % Fig. 5. Out-of-sample error vs. experimetal setup Figure 5 shows that the out-of-sample error slightly icreases as the size of maps icreases. However, it does ot icrease as rapidly for the larger maps as i case of the isample error. The out-of-sample error is, o average, equal to 0.9 for, 0.32 for, 0.44 for, ad 0.56 for. The error values for maps with are cosistet with values reported i [9]. Agai, we stress that the learig results are still of high quality with respect to the baselie value (0.39). The outof-sample error has larger stadard deviatio for experimets with the same map size ad desity performed o differet umber of processors whe compared with correspodig values for the i-sample error, i.e., they differs betwee ( %) ad ( %). This is due to the fact that sub-optimal solutios (FCM models) for the out-of-sample experimets may differ from each other. Thus, eve if all of them provide high quality results o previously see data (i-sample error); these models ca give differet simulatios for ew iitial vectors (out-of-sample error). Lastly, the relatioship betwee the iput map desity ad the out-of-sample error is more cosistet here tha i case of the i-sample error. O average, better quality is obtaied for deser maps i each case, i.e. for,,, ad.

6 We also aalyze relatio betwee the learig quality ad the umber of processors. The differece i the i-sample error betwee experimets repeated with the same setups (o 2, 4, ad 8 processors, respectively) with respect to the sequetial learig is very small (2-3% o average). I case of the out-of-sample error, these values are larger (6-2%), which is due to sub-optimal solutios foud by the RCGA (see the above paragraph). However, whe compared with the stadard deviatios obtaied for the sequetial setup, the differece i errors betwee the o-parallelized ad the parallelized implemetatios is ot sigificat. V. CONCLUSION This paper proposed a ovel parallel approach to learig of FCMs. The method is based o RCGA learig, which has bee previously reported as beig able to develop highquality FCMs from iput data. Our motivatio was to elimiate the mai drawback of the RCGA based method, which is icapability of dealig with larger maps due to high computatioal complexity. Our focus was to propose a solutio that would combie the quality of the RCGA method with a substatial decrease of the executio time. We have proposed, implemeted ad tested a solutio that parallelizes the geetic algorithm, which is the core of the RCGA approach. The experimetal results show that parallelizatio gives substatial improvemet i the executio time. The parallelized learig of FCMs o eight processors was reported to be up to four times faster tha the sequetial learig. The proposed method allows learig maps that iclude several dozes of cocepts i matter of few hours whe usig eight processors. The parallelized RCGA method for learig FCMs has bee tested o both sythetic ad real-life data. The experimetal results show that this method is able to provide high-quality solutios for large FCMs. Both i-sample ad out-of-sample errors icreased with the icreasig map size; however the error values were still substatially smaller tha the baselie error. VI. FUTURE WORK There are several future research directios based o this project. First, it would be iterestig to measure how much of the computatioal time is devoted to do the fitess evaluatio. Secod, a compariso betwee differet methods of RCGA parallelizatio to select the best approach will be ivestigated. Third, we pla to propose a alterative approach to speed up the learig process by exploitig iheret characteristics of FCMs, e.g. by dividig the iput data ito subsets, performig idepedet learig o each subset, ad, fially, mergig the sub-models. Last but ot least, we pla to apply oe of these learig methods to reallife problems, e.g. i the systems biology field. REFERENCES [] E. Alba, F. Lua, A. J. Nebro, ad J. M. Troya, Parallel heterogeeous geetic algorithms for cotiuous optimizatio, Parallel Computig, vol. 30, o. 5 6, pp , 04 [2] G. A. Baii, ad R. A. Bearma, Applicatio of fuzzy cogitive maps to factors affectig slurry rheology, It. Joural of Mieral Processig, vol. 52, o. 4, 998 [3] E. Catú Paz, A survey of parallel geetic algorithms, Calculateurs Parallèles, vol., o. 2, pp. 4 7, 998 [4] E. Catú Paz, Efficiet ad Accurate Parallel Geetic Algorithms, Kluwer Academic Publishers, 00 [5] K. Deb, A Itroductio to Geetic Algorithms, SADHANA, 999 [6] R. Giordao, G. Passarella, V. F. Uricchio, ad M. Vurro, Fuzzy cogitive maps for issue idetificatio i a water resources coflict resolutio system, Physics ad Chemistry of the Earth, vol. 30, o. 6 7 (Special Issue), pp , 05 [7] D. E. Goldberg, Geetic Algorithms i Search, Optimizatio, ad Machie Learig, Addiso Wesley, 989 [8] A. Grama, A. Gupta, G. Karypis, ad V. Kumar, Itroductio to Parallel Computig, Addiso Wesley, 03 [9] F. Herrera, M. Lozao, ad J. L. Verdegay, Tacklig real coded geetic algorithms: operators ad tools for behavioural aalysis, Artificial Itelligece Review, vol. 2, o. 4, pp , 998 [] P. R. Iocet, ad R. I. Joh, Computer aided fuzzy medical diagosis, Iformatio Scieces, vol. 62, o. 2, pp. 8 4, 04 [] D. Kardaras, ad G. Metzas, Usig fuzzy cogitive maps to model ad aalyse busiess performace assessmet, i Advaces i Idustrial Egieerig Applicatios ad Practice II, J. Che, ad A. Mital, (Eds), pp , 997 [2] M. Kha, ad M. Quaddus, Group decisio support usig fuzzy cogitive maps for causal reasoig, Group Decisio ad Negotiatio Joural, vol. 3, o. 5, pp , 04 [3] Z. Kofrsť, Parallel geetic algorithms: advaces, computig treds, applicatios ad perspectives, It. Parallel ad Distributed Processig Symposium, vol. 8, pp , 04 [4] B. Kosko, Fuzzy cogitive maps, It. Joural of Ma Machie Studies, vol. 24, pp , 986 [5] T. J. LeBlac, ad E. P. Markatos, Shared memory vs. message passig i shared memory multiprocessors, IEEE Symposium o Parallel ad Distributed Processig, pp , 992 [6] K. C. Lee, W. J. Lee, O. B. Kwo, J. H. Ha, ad P. I. Yu, Strategic plaig simulatio based o fuzzy cogitive map kowledge ad differetial game, Simulatio, vol. 7, o. 5, pp , 998 [7] E. I. Papageorgiou, C. D. Stylios, ad P. P. Groumpos, Fuzzy cogitive map learig based o oliear Hebbia rule, I: T. D. Gedeo, ad L. C. C. Fug, (Eds.), Lecture Notes i Artificial Itelligece, Spriger Verlag, vol. 2903, pp , 03. [8] W. Stach, L. Kurga, ad W. Pedrycz, Higher order fuzzy cogitive maps, North America Fuzzy Iformatio Processig Society Coferece (NAFIPS 06), 06 [9] W. Stach, L. Kurga, W. Pedrycz, ad M. Reformat, Geetic learig of fuzzy cogitive maps, Fuzzy Sets ad Systems, vol. 53, o. 3, pp. 37, 05 [] W. Stach, L. Kurga, W. Pedrycz, ad M. Reformat, Learig fuzzy cogitive maps with required precisio usig geetic algorithm approach, Electroics Letters, vol., o. 24, pp. 59 5, 04 [2] W. Stach, L. Kurga, W. Pedrycz, ad M. Reformat, Parallel fuzzy cogitive maps as a tool for modelig software developmet project, North America Fuzzy Iformatio Processig Society Coferece (NAFIPS 04), pp , 04 [22] W. Stach, L. A. Kurga, ad W. Pedrycz, A survey of fuzzy cogitive map learig methods, I: P. Grzegorzewski, M. Krawczak, ad S. Zadrozy, (Eds.), Issues i Soft Computig: Theory ad Applicatios, Exit, pp. 7 84, 05 [23] M. A. Stybliski, ad B. D. Meyer, Sigal flow graphs versus fuzzy cogitive maps i applicatio to qualitative circuit aalysis, It. Joural of Ma Machie Studies, vol. 35, pp , 99 [24] D. Whitley, A geetic algorithm tutorial, Statistics ad Computig, vol. 4, pp , 994 [25] C. Xavier, ad S. S. Iyegar, Itroductio to Parallel Algorithms, Wiley Itersciece, 998

Data-Driven Nonlinear Hebbian Learning Method for Fuzzy Cognitive Maps

Data-Driven Nonlinear Hebbian Learning Method for Fuzzy Cognitive Maps Data-Drive Noliear Hebbia Learig Method for Fuzzy Cogitive Maps Wociech Stach, Lukasz Kurga, ad Witold Pedrycz Abstract Fuzzy Cogitive Maps (FCMs) are a coveiet tool for modelig of dyamic systems by meas

More information

Ones Assignment Method for Solving Traveling Salesman Problem

Ones Assignment Method for Solving Traveling Salesman Problem Joural of mathematics ad computer sciece 0 (0), 58-65 Oes Assigmet Method for Solvig Travelig Salesma Problem Hadi Basirzadeh Departmet of Mathematics, Shahid Chamra Uiversity, Ahvaz, Ira Article history:

More information

An Improved Shuffled Frog-Leaping Algorithm for Knapsack Problem

An Improved Shuffled Frog-Leaping Algorithm for Knapsack Problem A Improved Shuffled Frog-Leapig Algorithm for Kapsack Problem Zhoufag Li, Ya Zhou, ad Peg Cheg School of Iformatio Sciece ad Egieerig Hea Uiversity of Techology ZhegZhou, Chia lzhf1978@126.com Abstract.

More information

A New Morphological 3D Shape Decomposition: Grayscale Interframe Interpolation Method

A New Morphological 3D Shape Decomposition: Grayscale Interframe Interpolation Method A ew Morphological 3D Shape Decompositio: Grayscale Iterframe Iterpolatio Method D.. Vizireau Politehica Uiversity Bucharest, Romaia ae@comm.pub.ro R. M. Udrea Politehica Uiversity Bucharest, Romaia mihea@comm.pub.ro

More information

ISSN (Print) Research Article. *Corresponding author Nengfa Hu

ISSN (Print) Research Article. *Corresponding author Nengfa Hu Scholars Joural of Egieerig ad Techology (SJET) Sch. J. Eg. Tech., 2016; 4(5):249-253 Scholars Academic ad Scietific Publisher (A Iteratioal Publisher for Academic ad Scietific Resources) www.saspublisher.com

More information

A SOFTWARE MODEL FOR THE MULTILAYER PERCEPTRON

A SOFTWARE MODEL FOR THE MULTILAYER PERCEPTRON A SOFTWARE MODEL FOR THE MULTILAYER PERCEPTRON Roberto Lopez ad Eugeio Oñate Iteratioal Ceter for Numerical Methods i Egieerig (CIMNE) Edificio C1, Gra Capitá s/, 08034 Barceloa, Spai ABSTRACT I this work

More information

Copyright 2016 Ramez Elmasri and Shamkant B. Navathe

Copyright 2016 Ramez Elmasri and Shamkant B. Navathe Copyright 2016 Ramez Elmasri ad Shamkat B. Navathe CHAPTER 18 Strategies for Query Processig Copyright 2016 Ramez Elmasri ad Shamkat B. Navathe Itroductio DBMS techiques to process a query Scaer idetifies

More information

ANN WHICH COVERS MLP AND RBF

ANN WHICH COVERS MLP AND RBF ANN WHICH COVERS MLP AND RBF Josef Boští, Jaromír Kual Faculty of Nuclear Scieces ad Physical Egieerig, CTU i Prague Departmet of Software Egieerig Abstract Two basic types of artificial eural etwors Multi

More information

Fuzzy Rule Selection by Data Mining Criteria and Genetic Algorithms

Fuzzy Rule Selection by Data Mining Criteria and Genetic Algorithms Fuzzy Rule Selectio by Data Miig Criteria ad Geetic Algorithms Hisao Ishibuchi Dept. of Idustrial Egieerig Osaka Prefecture Uiversity 1-1 Gakue-cho, Sakai, Osaka 599-8531, JAPAN E-mail: hisaoi@ie.osakafu-u.ac.jp

More information

Improving Template Based Spike Detection

Improving Template Based Spike Detection Improvig Template Based Spike Detectio Kirk Smith, Member - IEEE Portlad State Uiversity petra@ee.pdx.edu Abstract Template matchig algorithms like SSE, Covolutio ad Maximum Likelihood are well kow for

More information

Analysis of Server Resource Consumption of Meteorological Satellite Application System Based on Contour Curve

Analysis of Server Resource Consumption of Meteorological Satellite Application System Based on Contour Curve Advaces i Computer, Sigals ad Systems (2018) 2: 19-25 Clausius Scietific Press, Caada Aalysis of Server Resource Cosumptio of Meteorological Satellite Applicatio System Based o Cotour Curve Xiagag Zhao

More information

3D Model Retrieval Method Based on Sample Prediction

3D Model Retrieval Method Based on Sample Prediction 20 Iteratioal Coferece o Computer Commuicatio ad Maagemet Proc.of CSIT vol.5 (20) (20) IACSIT Press, Sigapore 3D Model Retrieval Method Based o Sample Predictio Qigche Zhag, Ya Tag* School of Computer

More information

Heuristic Approaches for Solving the Multidimensional Knapsack Problem (MKP)

Heuristic Approaches for Solving the Multidimensional Knapsack Problem (MKP) Heuristic Approaches for Solvig the Multidimesioal Kapsack Problem (MKP) R. PARRA-HERNANDEZ N. DIMOPOULOS Departmet of Electrical ad Computer Eg. Uiversity of Victoria Victoria, B.C. CANADA Abstract: -

More information

Neural Networks A Model of Boolean Functions

Neural Networks A Model of Boolean Functions Neural Networks A Model of Boolea Fuctios Berd Steibach, Roma Kohut Freiberg Uiversity of Miig ad Techology Istitute of Computer Sciece D-09596 Freiberg, Germay e-mails: steib@iformatik.tu-freiberg.de

More information

Fast Fourier Transform (FFT) Algorithms

Fast Fourier Transform (FFT) Algorithms Fast Fourier Trasform FFT Algorithms Relatio to the z-trasform elsewhere, ozero, z x z X x [ ] 2 ~ elsewhere,, ~ e j x X x x π j e z z X X π 2 ~ The DFS X represets evely spaced samples of the z- trasform

More information

Bayesian approach to reliability modelling for a probability of failure on demand parameter

Bayesian approach to reliability modelling for a probability of failure on demand parameter Bayesia approach to reliability modellig for a probability of failure o demad parameter BÖRCSÖK J., SCHAEFER S. Departmet of Computer Architecture ad System Programmig Uiversity Kassel, Wilhelmshöher Allee

More information

An Estimation of Distribution Algorithm for solving the Knapsack problem

An Estimation of Distribution Algorithm for solving the Knapsack problem Vol.4,No.5, 214 Published olie: May 25, 214 DOI: 1.7321/jscse.v4.5.1 A Estimatio of Distributio Algorithm for solvig the Kapsack problem 1 Ricardo Pérez, 2 S. Jös, 3 Arturo Herádez, 4 Carlos A. Ochoa *1,

More information

BASED ON ITERATIVE ERROR-CORRECTION

BASED ON ITERATIVE ERROR-CORRECTION A COHPARISO OF CRYPTAALYTIC PRICIPLES BASED O ITERATIVE ERROR-CORRECTIO Miodrag J. MihaljeviC ad Jova Dj. GoliC Istitute of Applied Mathematics ad Electroics. Belgrade School of Electrical Egieerig. Uiversity

More information

Optimization for framework design of new product introduction management system Ma Ying, Wu Hongcui

Optimization for framework design of new product introduction management system Ma Ying, Wu Hongcui 2d Iteratioal Coferece o Electrical, Computer Egieerig ad Electroics (ICECEE 2015) Optimizatio for framework desig of ew product itroductio maagemet system Ma Yig, Wu Hogcui Tiaji Electroic Iformatio Vocatioal

More information

Solving Fuzzy Assignment Problem Using Fourier Elimination Method

Solving Fuzzy Assignment Problem Using Fourier Elimination Method Global Joural of Pure ad Applied Mathematics. ISSN 0973-768 Volume 3, Number 2 (207), pp. 453-462 Research Idia Publicatios http://www.ripublicatio.com Solvig Fuzzy Assigmet Problem Usig Fourier Elimiatio

More information

Copyright 2016 Ramez Elmasri and Shamkant B. Navathe

Copyright 2016 Ramez Elmasri and Shamkant B. Navathe Copyright 2016 Ramez Elmasri ad Shamkat B. Navathe CHAPTER 19 Query Optimizatio Copyright 2016 Ramez Elmasri ad Shamkat B. Navathe Itroductio Query optimizatio Coducted by a query optimizer i a DBMS Goal:

More information

Our second algorithm. Comp 135 Machine Learning Computer Science Tufts University. Decision Trees. Decision Trees. Decision Trees.

Our second algorithm. Comp 135 Machine Learning Computer Science Tufts University. Decision Trees. Decision Trees. Decision Trees. Comp 135 Machie Learig Computer Sciece Tufts Uiversity Fall 2017 Roi Khardo Some of these slides were adapted from previous slides by Carla Brodley Our secod algorithm Let s look at a simple dataset for

More information

Pattern Recognition Systems Lab 1 Least Mean Squares

Pattern Recognition Systems Lab 1 Least Mean Squares Patter Recogitio Systems Lab 1 Least Mea Squares 1. Objectives This laboratory work itroduces the OpeCV-based framework used throughout the course. I this assigmet a lie is fitted to a set of poits usig

More information

A Study on the Performance of Cholesky-Factorization using MPI

A Study on the Performance of Cholesky-Factorization using MPI A Study o the Performace of Cholesky-Factorizatio usig MPI Ha S. Kim Scott B. Bade Departmet of Computer Sciece ad Egieerig Uiversity of Califoria Sa Diego {hskim, bade}@cs.ucsd.edu Abstract Cholesky-factorizatio

More information

Sectio 4, a prototype project of settig field weight with AHP method is developed ad the experimetal results are aalyzed. Fially, we coclude our work

Sectio 4, a prototype project of settig field weight with AHP method is developed ad the experimetal results are aalyzed. Fially, we coclude our work 200 2d Iteratioal Coferece o Iformatio ad Multimedia Techology (ICIMT 200) IPCSIT vol. 42 (202) (202) IACSIT Press, Sigapore DOI: 0.7763/IPCSIT.202.V42.0 Idex Weight Decisio Based o AHP for Iformatio Retrieval

More information

Euclidean Distance Based Feature Selection for Fault Detection Prediction Model in Semiconductor Manufacturing Process

Euclidean Distance Based Feature Selection for Fault Detection Prediction Model in Semiconductor Manufacturing Process Vol.133 (Iformatio Techology ad Computer Sciece 016), pp.85-89 http://dx.doi.org/10.1457/astl.016. Euclidea Distace Based Feature Selectio for Fault Detectio Predictio Model i Semicoductor Maufacturig

More information

Load balanced Parallel Prime Number Generator with Sieve of Eratosthenes on Cluster Computers *

Load balanced Parallel Prime Number Generator with Sieve of Eratosthenes on Cluster Computers * Load balaced Parallel Prime umber Geerator with Sieve of Eratosthees o luster omputers * Soowook Hwag*, Kyusik hug**, ad Dogseug Kim* *Departmet of Electrical Egieerig Korea Uiversity Seoul, -, Rep. of

More information

Chapter 3 Classification of FFT Processor Algorithms

Chapter 3 Classification of FFT Processor Algorithms Chapter Classificatio of FFT Processor Algorithms The computatioal complexity of the Discrete Fourier trasform (DFT) is very high. It requires () 2 complex multiplicatios ad () complex additios [5]. As

More information

Evaluation scheme for Tracking in AMI

Evaluation scheme for Tracking in AMI A M I C o m m u i c a t i o A U G M E N T E D M U L T I - P A R T Y I N T E R A C T I O N http://www.amiproject.org/ Evaluatio scheme for Trackig i AMI S. Schreiber a D. Gatica-Perez b AMI WP4 Trackig:

More information

Lecture Notes 6 Introduction to algorithm analysis CSS 501 Data Structures and Object-Oriented Programming

Lecture Notes 6 Introduction to algorithm analysis CSS 501 Data Structures and Object-Oriented Programming Lecture Notes 6 Itroductio to algorithm aalysis CSS 501 Data Structures ad Object-Orieted Programmig Readig for this lecture: Carrao, Chapter 10 To be covered i this lecture: Itroductio to algorithm aalysis

More information

Fuzzy Linear Regression Analysis

Fuzzy Linear Regression Analysis 12th IFAC Coferece o Programmable Devices ad Embedded Systems The Iteratioal Federatio of Automatic Cotrol September 25-27, 2013. Fuzzy Liear Regressio Aalysis Jaa Nowaková Miroslav Pokorý VŠB-Techical

More information

EMPIRICAL ANALYSIS OF FAULT PREDICATION TECHNIQUES FOR IMPROVING SOFTWARE PROCESS CONTROL

EMPIRICAL ANALYSIS OF FAULT PREDICATION TECHNIQUES FOR IMPROVING SOFTWARE PROCESS CONTROL Iteratioal Joural of Iformatio Techology ad Kowledge Maagemet July-December 2012, Volume 5, No. 2, pp. 371-375 EMPIRICAL ANALYSIS OF FAULT PREDICATION TECHNIQUES FOR IMPROVING SOFTWARE PROCESS CONTROL

More information

An Algorithm to Solve Multi-Objective Assignment. Problem Using Interactive Fuzzy. Goal Programming Approach

An Algorithm to Solve Multi-Objective Assignment. Problem Using Interactive Fuzzy. Goal Programming Approach It. J. Cotemp. Math. Scieces, Vol. 6, 0, o. 34, 65-66 A Algorm to Solve Multi-Objective Assigmet Problem Usig Iteractive Fuzzy Goal Programmig Approach P. K. De ad Bharti Yadav Departmet of Mathematics

More information

Dynamic Programming and Curve Fitting Based Road Boundary Detection

Dynamic Programming and Curve Fitting Based Road Boundary Detection Dyamic Programmig ad Curve Fittig Based Road Boudary Detectio SHYAM PRASAD ADHIKARI, HYONGSUK KIM, Divisio of Electroics ad Iformatio Egieerig Chobuk Natioal Uiversity 664-4 Ga Deokji-Dog Jeoju-City Jeobuk

More information

Mobile terminal 3D image reconstruction program development based on Android Lin Qinhua

Mobile terminal 3D image reconstruction program development based on Android Lin Qinhua Iteratioal Coferece o Automatio, Mechaical Cotrol ad Computatioal Egieerig (AMCCE 05) Mobile termial 3D image recostructio program developmet based o Adroid Li Qihua Sichua Iformatio Techology College

More information

Introduction. Nature-Inspired Computing. Terminology. Problem Types. Constraint Satisfaction Problems - CSP. Free Optimization Problem - FOP

Introduction. Nature-Inspired Computing. Terminology. Problem Types. Constraint Satisfaction Problems - CSP. Free Optimization Problem - FOP Nature-Ispired Computig Hadlig Costraits Dr. Şima Uyar September 2006 Itroductio may practical problems are costraied ot all combiatios of variable values represet valid solutios feasible solutios ifeasible

More information

Image Segmentation EEE 508

Image Segmentation EEE 508 Image Segmetatio Objective: to determie (etract) object boudaries. It is a process of partitioig a image ito distict regios by groupig together eighborig piels based o some predefied similarity criterio.

More information

Octahedral Graph Scaling

Octahedral Graph Scaling Octahedral Graph Scalig Peter Russell Jauary 1, 2015 Abstract There is presetly o strog iterpretatio for the otio of -vertex graph scalig. This paper presets a ew defiitio for the term i the cotext of

More information

Structuring Redundancy for Fault Tolerance. CSE 598D: Fault Tolerant Software

Structuring Redundancy for Fault Tolerance. CSE 598D: Fault Tolerant Software Structurig Redudacy for Fault Tolerace CSE 598D: Fault Tolerat Software What do we wat to achieve? Versios Damage Assessmet Versio 1 Error Detectio Iputs Versio 2 Voter Outputs State Restoratio Cotiued

More information

Project 2.5 Improved Euler Implementation

Project 2.5 Improved Euler Implementation Project 2.5 Improved Euler Implemetatio Figure 2.5.10 i the text lists TI-85 ad BASIC programs implemetig the improved Euler method to approximate the solutio of the iitial value problem dy dx = x+ y,

More information

Appendix D. Controller Implementation

Appendix D. Controller Implementation COMPUTER ORGANIZATION AND DESIGN The Hardware/Software Iterface 5 th Editio Appedix D Cotroller Implemetatio Cotroller Implemetatios Combiatioal logic (sigle-cycle); Fiite state machie (multi-cycle, pipelied);

More information

Adaptive Resource Allocation for Electric Environmental Pollution through the Control Network

Adaptive Resource Allocation for Electric Environmental Pollution through the Control Network Available olie at www.sciecedirect.com Eergy Procedia 6 (202) 60 64 202 Iteratioal Coferece o Future Eergy, Eviromet, ad Materials Adaptive Resource Allocatio for Electric Evirometal Pollutio through the

More information

6.854J / J Advanced Algorithms Fall 2008

6.854J / J Advanced Algorithms Fall 2008 MIT OpeCourseWare http://ocw.mit.edu 6.854J / 18.415J Advaced Algorithms Fall 2008 For iformatio about citig these materials or our Terms of Use, visit: http://ocw.mit.edu/terms. 18.415/6.854 Advaced Algorithms

More information

SOFTWARE usually does not work alone. It must have

SOFTWARE usually does not work alone. It must have Proceedigs of the 203 Federated Coferece o Computer Sciece ad Iformatio Systems pp. 343 348 A method for selectig eviromets for software compatibility testig Łukasz Pobereżik AGH Uiversity of Sciece ad

More information

Algorithms for Disk Covering Problems with the Most Points

Algorithms for Disk Covering Problems with the Most Points Algorithms for Disk Coverig Problems with the Most Poits Bi Xiao Departmet of Computig Hog Kog Polytechic Uiversity Hug Hom, Kowloo, Hog Kog csbxiao@comp.polyu.edu.hk Qigfeg Zhuge, Yi He, Zili Shao, Edwi

More information

. Written in factored form it is easy to see that the roots are 2, 2, i,

. Written in factored form it is easy to see that the roots are 2, 2, i, CMPS A Itroductio to Programmig Programmig Assigmet 4 I this assigmet you will write a java program that determies the real roots of a polyomial that lie withi a specified rage. Recall that the roots (or

More information

A Modified Multiband U Shaped and Microcontroller Shaped Fractal Antenna

A Modified Multiband U Shaped and Microcontroller Shaped Fractal Antenna al Joural o Recet ad Iovatio Treds i Computig ad Commuicatio ISSN: 221-8169 A Modified Multibad U Shaped ad Microcotroller Shaped Fractal Atea Shweta Goyal 1, Yogedra Kumar Katiyar 2 1 M.tech Scholar,

More information

FAST BIT-REVERSALS ON UNIPROCESSORS AND SHARED-MEMORY MULTIPROCESSORS

FAST BIT-REVERSALS ON UNIPROCESSORS AND SHARED-MEMORY MULTIPROCESSORS SIAM J. SCI. COMPUT. Vol. 22, No. 6, pp. 2113 2134 c 21 Society for Idustrial ad Applied Mathematics FAST BIT-REVERSALS ON UNIPROCESSORS AND SHARED-MEMORY MULTIPROCESSORS ZHAO ZHANG AND XIAODONG ZHANG

More information

Feature classification for multi-focus image fusion

Feature classification for multi-focus image fusion Iteratioal Joural of the Physical Scieces Vol. 6(0), pp. 4838-4847, 3 September, 0 Available olie at http://www.academicjourals.org/ijps DOI: 0.5897/IJPS.73 ISSN 99-950 0 Academic Jourals Full Legth Research

More information

VISUALSLX AN OPEN USER SHELL FOR HIGH-PERFORMANCE MODELING AND SIMULATION. Thomas Wiedemann

VISUALSLX AN OPEN USER SHELL FOR HIGH-PERFORMANCE MODELING AND SIMULATION. Thomas Wiedemann Proceedigs of the 2000 Witer Simulatio Coferece J. A. Joies, R. R. Barto, K. Kag, ad P. A. Fishwick, eds. VISUALSLX AN OPEN USER SHELL FOR HIGH-PERFORMANCE MODELING AND SIMULATION Thomas Wiedema Techical

More information

Elementary Educational Computer

Elementary Educational Computer Chapter 5 Elemetary Educatioal Computer. Geeral structure of the Elemetary Educatioal Computer (EEC) The EEC coforms to the 5 uits structure defied by vo Neuma's model (.) All uits are preseted i a simplified

More information

Analysis Metrics. Intro to Algorithm Analysis. Slides. 12. Alg Analysis. 12. Alg Analysis

Analysis Metrics. Intro to Algorithm Analysis. Slides. 12. Alg Analysis. 12. Alg Analysis Itro to Algorithm Aalysis Aalysis Metrics Slides. Table of Cotets. Aalysis Metrics 3. Exact Aalysis Rules 4. Simple Summatio 5. Summatio Formulas 6. Order of Magitude 7. Big-O otatio 8. Big-O Theorems

More information

LU Decomposition Method

LU Decomposition Method SOLUTION OF SIMULTANEOUS LINEAR EQUATIONS LU Decompositio Method Jamie Traha, Autar Kaw, Kevi Marti Uiversity of South Florida Uited States of America kaw@eg.usf.edu http://umericalmethods.eg.usf.edu Itroductio

More information

Performance Comparisons of PSO based Clustering

Performance Comparisons of PSO based Clustering Performace Comparisos of PSO based Clusterig Suresh Chadra Satapathy, 2 Guaidhi Pradha, 3 Sabyasachi Pattai, 4 JVR Murthy, 5 PVGD Prasad Reddy Ail Neeruoda Istitute of Techology ad Scieces, Sagivalas,Vishaapatam

More information

A Parallel DFA Minimization Algorithm

A Parallel DFA Minimization Algorithm A Parallel DFA Miimizatio Algorithm Ambuj Tewari, Utkarsh Srivastava, ad P. Gupta Departmet of Computer Sciece & Egieerig Idia Istitute of Techology Kapur Kapur 208 016,INDIA pg@iitk.ac.i Abstract. I this

More information

Lecture 1: Introduction and Strassen s Algorithm

Lecture 1: Introduction and Strassen s Algorithm 5-750: Graduate Algorithms Jauary 7, 08 Lecture : Itroductio ad Strasse s Algorithm Lecturer: Gary Miller Scribe: Robert Parker Itroductio Machie models I this class, we will primarily use the Radom Access

More information

Parallel Polygon Approximation Algorithm Targeted at Reconfigurable Multi-Ring Hardware

Parallel Polygon Approximation Algorithm Targeted at Reconfigurable Multi-Ring Hardware Parallel Polygo Approximatio Algorithm Targeted at Recofigurable Multi-Rig Hardware M. Arif Wai* ad Hamid R. Arabia** *Califoria State Uiversity Bakersfield, Califoria, USA **Uiversity of Georgia, Georgia,

More information

arxiv: v2 [cs.ds] 24 Mar 2018

arxiv: v2 [cs.ds] 24 Mar 2018 Similar Elemets ad Metric Labelig o Complete Graphs arxiv:1803.08037v [cs.ds] 4 Mar 018 Pedro F. Felzeszwalb Brow Uiversity Providece, RI, USA pff@brow.edu March 8, 018 We cosider a problem that ivolves

More information

Variance as a Stopping Criterion for Genetic Algorithms with Elitist Model

Variance as a Stopping Criterion for Genetic Algorithms with Elitist Model Fudameta Iformaticae 120 (2012) 145 164 145 DOI 10.3233/FI-2012-754 IOS Press Variace as a Stoppig Criterio for Geetic Algorithms with Elitist Model Diabadhu Bhadari, C. A. Murthy, Sakar K. Pal Ceter for

More information

GPUMP: a Multiple-Precision Integer Library for GPUs

GPUMP: a Multiple-Precision Integer Library for GPUs GPUMP: a Multiple-Precisio Iteger Library for GPUs Kaiyog Zhao ad Xiaowe Chu Departmet of Computer Sciece, Hog Kog Baptist Uiversity Hog Kog, P. R. Chia Email: {kyzhao, chxw}@comp.hkbu.edu.hk Abstract

More information

HADOOP: A NEW APPROACH FOR DOCUMENT CLUSTERING

HADOOP: A NEW APPROACH FOR DOCUMENT CLUSTERING Y.K. Patil* Iteratioal Joural of Advaced Research i ISSN: 2278-6244 IT ad Egieerig Impact Factor: 4.54 HADOOP: A NEW APPROACH FOR DOCUMENT CLUSTERING Prof. V.S. Nadedkar** Abstract: Documet clusterig is

More information

Mining from Quantitative Data with Linguistic Minimum Supports and Confidences

Mining from Quantitative Data with Linguistic Minimum Supports and Confidences Miig from Quatitative Data with Liguistic Miimum Supports ad Cofideces Tzug-Pei Hog, Mig-Jer Chiag ad Shyue-Liag Wag Departmet of Electrical Egieerig Natioal Uiversity of Kaohsiug Kaohsiug, 8, Taiwa, R.O.C.

More information

Outline. Research Definition. Motivation. Foundation of Reverse Engineering. Dynamic Analysis and Design Pattern Detection in Java Programs

Outline. Research Definition. Motivation. Foundation of Reverse Engineering. Dynamic Analysis and Design Pattern Detection in Java Programs Dyamic Aalysis ad Desig Patter Detectio i Java Programs Outlie Lei Hu Kamra Sartipi {hul4, sartipi}@mcmasterca Departmet of Computig ad Software McMaster Uiversity Caada Motivatio Research Problem Defiitio

More information

Using a Dynamic Interval Type-2 Fuzzy Interpolation Method to Improve Modeless Robots Calibrations

Using a Dynamic Interval Type-2 Fuzzy Interpolation Method to Improve Modeless Robots Calibrations Joural of Cotrol Sciece ad Egieerig 3 (25) 9-7 doi:.7265/2328-223/25.3. D DAVID PUBLISHING Usig a Dyamic Iterval Type-2 Fuzzy Iterpolatio Method to Improve Modeless Robots Calibratios Yig Bai ad Dali Wag

More information

Designing a learning system

Designing a learning system CS 75 Machie Learig Lecture Desigig a learig system Milos Hauskrecht milos@cs.pitt.edu 539 Seott Square, x-5 people.cs.pitt.edu/~milos/courses/cs75/ Admiistrivia No homework assigmet this week Please try

More information

Improvement of the Orthogonal Code Convolution Capabilities Using FPGA Implementation

Improvement of the Orthogonal Code Convolution Capabilities Using FPGA Implementation Improvemet of the Orthogoal Code Covolutio Capabilities Usig FPGA Implemetatio Naima Kaabouch, Member, IEEE, Apara Dhirde, Member, IEEE, Saleh Faruque, Member, IEEE Departmet of Electrical Egieerig, Uiversity

More information

Software development of components for complex signal analysis on the example of adaptive recursive estimation methods.

Software development of components for complex signal analysis on the example of adaptive recursive estimation methods. Software developmet of compoets for complex sigal aalysis o the example of adaptive recursive estimatio methods. SIMON BOYMANN, RALPH MASCHOTTA, SILKE LEHMANN, DUNJA STEUER Istitute of Biomedical Egieerig

More information

Lecture 5. Counting Sort / Radix Sort

Lecture 5. Counting Sort / Radix Sort Lecture 5. Coutig Sort / Radix Sort T. H. Corme, C. E. Leiserso ad R. L. Rivest Itroductio to Algorithms, 3rd Editio, MIT Press, 2009 Sugkyukwa Uiversity Hyuseug Choo choo@skku.edu Copyright 2000-2018

More information

A Polynomial Interval Shortest-Route Algorithm for Acyclic Network

A Polynomial Interval Shortest-Route Algorithm for Acyclic Network A Polyomial Iterval Shortest-Route Algorithm for Acyclic Network Hossai M Akter Key words: Iterval; iterval shortest-route problem; iterval algorithm; ucertaity Abstract A method ad algorithm is preseted

More information

Redundancy Allocation for Series Parallel Systems with Multiple Constraints and Sensitivity Analysis

Redundancy Allocation for Series Parallel Systems with Multiple Constraints and Sensitivity Analysis IOSR Joural of Egieerig Redudacy Allocatio for Series Parallel Systems with Multiple Costraits ad Sesitivity Aalysis S. V. Suresh Babu, D.Maheswar 2, G. Ragaath 3 Y.Viaya Kumar d G.Sakaraiah e (Mechaical

More information

An improved support vector machine based on particle swarm optimization in laser ultrasonic defect detection

An improved support vector machine based on particle swarm optimization in laser ultrasonic defect detection A improved support vector machie based o particle swarm optimizatio i laser ultrasoic defect detectio School of Sciece, North Uiversity of Chia, aiyua, Shaxi 35, Chia xut98@63.com,hhp9@63.com,xywag@6.com,43497@qq.com

More information

AN OPTIMIZATION NETWORK FOR MATRIX INVERSION

AN OPTIMIZATION NETWORK FOR MATRIX INVERSION 397 AN OPTIMIZATION NETWORK FOR MATRIX INVERSION Ju-Seog Jag, S~ Youg Lee, ad Sag-Yug Shi Korea Advaced Istitute of Sciece ad Techology, P.O. Box 150, Cheogryag, Seoul, Korea ABSTRACT Iverse matrix calculatio

More information

Probabilistic Fuzzy Time Series Method Based on Artificial Neural Network

Probabilistic Fuzzy Time Series Method Based on Artificial Neural Network America Joural of Itelliget Systems 206, 6(2): 42-47 DOI: 0.5923/j.ajis.2060602.02 Probabilistic Fuzzy Time Series Method Based o Artificial Neural Network Erol Egrioglu,*, Ere Bas, Cagdas Haka Aladag

More information

Morgan Kaufmann Publishers 26 February, COMPUTER ORGANIZATION AND DESIGN The Hardware/Software Interface. Chapter 5

Morgan Kaufmann Publishers 26 February, COMPUTER ORGANIZATION AND DESIGN The Hardware/Software Interface. Chapter 5 Morga Kaufma Publishers 26 February, 28 COMPUTER ORGANIZATION AND DESIGN The Hardware/Software Iterface 5 th Editio Chapter 5 Set-Associative Cache Architecture Performace Summary Whe CPU performace icreases:

More information

DETECTION OF LANDSLIDE BLOCK BOUNDARIES BY MEANS OF AN AFFINE COORDINATE TRANSFORMATION

DETECTION OF LANDSLIDE BLOCK BOUNDARIES BY MEANS OF AN AFFINE COORDINATE TRANSFORMATION Proceedigs, 11 th FIG Symposium o Deformatio Measuremets, Satorii, Greece, 2003. DETECTION OF LANDSLIDE BLOCK BOUNDARIES BY MEANS OF AN AFFINE COORDINATE TRANSFORMATION Michaela Haberler, Heribert Kahme

More information

CS 683: Advanced Design and Analysis of Algorithms

CS 683: Advanced Design and Analysis of Algorithms CS 683: Advaced Desig ad Aalysis of Algorithms Lecture 6, February 1, 2008 Lecturer: Joh Hopcroft Scribes: Shaomei Wu, Etha Feldma February 7, 2008 1 Threshold for k CNF Satisfiability I the previous lecture,

More information

The Counterchanged Crossed Cube Interconnection Network and Its Topology Properties

The Counterchanged Crossed Cube Interconnection Network and Its Topology Properties WSEAS TRANSACTIONS o COMMUNICATIONS Wag Xiyag The Couterchaged Crossed Cube Itercoectio Network ad Its Topology Properties WANG XINYANG School of Computer Sciece ad Egieerig South Chia Uiversity of Techology

More information

Harris Corner Detection Algorithm at Sub-pixel Level and Its Application Yuanfeng Han a, Peijiang Chen b * and Tian Meng c

Harris Corner Detection Algorithm at Sub-pixel Level and Its Application Yuanfeng Han a, Peijiang Chen b * and Tian Meng c Iteratioal Coferece o Computatioal Sciece ad Egieerig (ICCSE 015) Harris Corer Detectio Algorithm at Sub-pixel Level ad Its Applicatio Yuafeg Ha a, Peijiag Che b * ad Tia Meg c School of Automobile, Liyi

More information

1 Enterprise Modeler

1 Enterprise Modeler 1 Eterprise Modeler Itroductio I BaaERP, a Busiess Cotrol Model ad a Eterprise Structure Model for multi-site cofiguratios are itroduced. Eterprise Structure Model Busiess Cotrol Models Busiess Fuctio

More information

The Closest Line to a Data Set in the Plane. David Gurney Southeastern Louisiana University Hammond, Louisiana

The Closest Line to a Data Set in the Plane. David Gurney Southeastern Louisiana University Hammond, Louisiana The Closest Lie to a Data Set i the Plae David Gurey Southeaster Louisiaa Uiversity Hammod, Louisiaa ABSTRACT This paper looks at three differet measures of distace betwee a lie ad a data set i the plae:

More information

How do we evaluate algorithms?

How do we evaluate algorithms? F2 Readig referece: chapter 2 + slides Algorithm complexity Big O ad big Ω To calculate ruig time Aalysis of recursive Algorithms Next time: Litterature: slides mostly The first Algorithm desig methods:

More information

New Fuzzy Color Clustering Algorithm Based on hsl Similarity

New Fuzzy Color Clustering Algorithm Based on hsl Similarity IFSA-EUSFLAT 009 New Fuzzy Color Clusterig Algorithm Based o hsl Similarity Vasile Ptracu Departmet of Iformatics Techology Tarom Compay Bucharest Romaia Email: patrascu.v@gmail.com Abstract I this paper

More information

Evaluation of Different Fitness Functions for the Evolutionary Testing of an Autonomous Parking System

Evaluation of Different Fitness Functions for the Evolutionary Testing of an Autonomous Parking System Evaluatio of Differet Fitess Fuctios for the Evolutioary Testig of a Autoomous Parkig System Joachim Wegeer 1 ad Oliver Bühler 2 1 DaimlerChrysler AG, Research ad Techology, Alt-Moabit 96 a, D-10559 Berli,

More information

Memetic Algorithm: Hybridization of Hill Climbing with Selection Operator

Memetic Algorithm: Hybridization of Hill Climbing with Selection Operator Iteratioal Joural of Soft Computig ad Egieerig (IJSCE) ISSN: 2231-2307, Volume-3, Issue-2, May 2013 Memetic Algorithm: Hybridizatio of Hill Climbig with Selectio Operator Rakesh Kumar, Sajay Tyagi, Maju

More information

Unsupervised Discretization Using Kernel Density Estimation

Unsupervised Discretization Using Kernel Density Estimation Usupervised Discretizatio Usig Kerel Desity Estimatio Maregle Biba, Floriaa Esposito, Stefao Ferilli, Nicola Di Mauro, Teresa M.A Basile Departmet of Computer Sciece, Uiversity of Bari Via Oraboa 4, 7025

More information

New HSL Distance Based Colour Clustering Algorithm

New HSL Distance Based Colour Clustering Algorithm The 4th Midwest Artificial Itelligece ad Cogitive Scieces Coferece (MAICS 03 pp 85-9 New Albay Idiaa USA April 3-4 03 New HSL Distace Based Colour Clusterig Algorithm Vasile Patrascu Departemet of Iformatics

More information

EE123 Digital Signal Processing

EE123 Digital Signal Processing Last Time EE Digital Sigal Processig Lecture 7 Block Covolutio, Overlap ad Add, FFT Discrete Fourier Trasform Properties of the Liear covolutio through circular Today Liear covolutio with Overlap ad add

More information

A Generalized Set Theoretic Approach for Time and Space Complexity Analysis of Algorithms and Functions

A Generalized Set Theoretic Approach for Time and Space Complexity Analysis of Algorithms and Functions Proceedigs of the 10th WSEAS Iteratioal Coferece o APPLIED MATHEMATICS, Dallas, Texas, USA, November 1-3, 2006 316 A Geeralized Set Theoretic Approach for Time ad Space Complexity Aalysis of Algorithms

More information

Evolutionary Hybrid Genetic-Firefly Algorithm for Global Optimization

Evolutionary Hybrid Genetic-Firefly Algorithm for Global Optimization Iteratioal Joural of Computatioal Egieerig & Maagemet, Vol. 6 Issue, May ISSN (Olie): -789 www..org Evolutioary Hybrid Geetic-Firefly Algorithm for Global Optimizatio 7 Shaik Farook, P. Sagameswara Raju

More information

An Algorithm to Solve Fuzzy Trapezoidal Transshipment Problem

An Algorithm to Solve Fuzzy Trapezoidal Transshipment Problem Iteratioal Joural of Systems Sciece ad Applied Mathematics 206; (4): 58-62 http://www.sciecepublishiggroup.com/j/ssam doi: 0.648/j.ssam.206004.4 A Algorithm to Solve Fuzzy Trapezoidal Trasshipmet Problem

More information

An Efficient Algorithm for Graph Bisection of Triangularizations

An Efficient Algorithm for Graph Bisection of Triangularizations A Efficiet Algorithm for Graph Bisectio of Triagularizatios Gerold Jäger Departmet of Computer Sciece Washigto Uiversity Campus Box 1045 Oe Brookigs Drive St. Louis, Missouri 63130-4899, USA jaegerg@cse.wustl.edu

More information

Optimization of Multiple Input Single Output Fuzzy Membership Functions Using Clonal Selection Algorithm

Optimization of Multiple Input Single Output Fuzzy Membership Functions Using Clonal Selection Algorithm Optimizatio of Multiple Iput Sigle Output Fuzzy Membership Fuctios Usig Cloal Selectio Algorithm AYŞE MERVE ACILAR, AHMET ARSLAN Computer Egieerig Departmet Selcuk Uiversity Selcuk Uiversity, Eg.-Arch.

More information

Data Analysis. Concepts and Techniques. Chapter 2. Chapter 2: Getting to Know Your Data. Data Objects and Attribute Types

Data Analysis. Concepts and Techniques. Chapter 2. Chapter 2: Getting to Know Your Data. Data Objects and Attribute Types Data Aalysis Cocepts ad Techiques Chapter 2 1 Chapter 2: Gettig to Kow Your Data Data Objects ad Attribute Types Basic Statistical Descriptios of Data Data Visualizatio Measurig Data Similarity ad Dissimilarity

More information

Cache-Optimal Methods for Bit-Reversals

Cache-Optimal Methods for Bit-Reversals Proceedigs of the ACM/IEEE Supercomputig Coferece, November 1999, Portlad, Orego, U.S.A. Cache-Optimal Methods for Bit-Reversals Zhao Zhag ad Xiaodog Zhag Departmet of Computer Sciece College of William

More information

BAYESIAN WITH FULL CONDITIONAL POSTERIOR DISTRIBUTION APPROACH FOR SOLUTION OF COMPLEX MODELS. Pudji Ismartini

BAYESIAN WITH FULL CONDITIONAL POSTERIOR DISTRIBUTION APPROACH FOR SOLUTION OF COMPLEX MODELS. Pudji Ismartini Proceedig of Iteratioal Coferece O Research, Implemetatio Ad Educatio Of Mathematics Ad Scieces 014, Yogyakarta State Uiversity, 18-0 May 014 BAYESIAN WIH FULL CONDIIONAL POSERIOR DISRIBUION APPROACH FOR

More information

Algorithm Selection using Reinforcement Learning

Algorithm Selection using Reinforcement Learning Algorithm Selectio usig Reiforcemet Learig Michail G. Lagoudakis Departmet of Computer Sciece, Duke Uiversity, Durham, NC 2778, USA Michael L. Littma Shao Laboratory, AT&T Labs Research, Florham Park,

More information

Enhancing Efficiency of Software Fault Tolerance Techniques in Satellite Motion System

Enhancing Efficiency of Software Fault Tolerance Techniques in Satellite Motion System Joural of Iformatio Systems ad Telecommuicatio, Vol. 2, No. 3, July-September 2014 173 Ehacig Efficiecy of Software Fault Tolerace Techiques i Satellite Motio System Hoda Baki Departmet of Electrical ad

More information

A Novel Feature Extraction Algorithm for Haar Local Binary Pattern Texture Based on Human Vision System

A Novel Feature Extraction Algorithm for Haar Local Binary Pattern Texture Based on Human Vision System A Novel Feature Extractio Algorithm for Haar Local Biary Patter Texture Based o Huma Visio System Liu Tao 1,* 1 Departmet of Electroic Egieerig Shaaxi Eergy Istitute Xiayag, Shaaxi, Chia Abstract The locality

More information

Improving Information Retrieval System Security via an Optimal Maximal Coding Scheme

Improving Information Retrieval System Security via an Optimal Maximal Coding Scheme Improvig Iformatio Retrieval System Security via a Optimal Maximal Codig Scheme Dogyag Log Departmet of Computer Sciece, City Uiversity of Hog Kog, 8 Tat Chee Aveue Kowloo, Hog Kog SAR, PRC dylog@cs.cityu.edu.hk

More information

Chapter 11. Friends, Overloaded Operators, and Arrays in Classes. Copyright 2014 Pearson Addison-Wesley. All rights reserved.

Chapter 11. Friends, Overloaded Operators, and Arrays in Classes. Copyright 2014 Pearson Addison-Wesley. All rights reserved. Chapter 11 Frieds, Overloaded Operators, ad Arrays i Classes Copyright 2014 Pearso Addiso-Wesley. All rights reserved. Overview 11.1 Fried Fuctios 11.2 Overloadig Operators 11.3 Arrays ad Classes 11.4

More information