Recommended Items Rating Prediction based on RBF Neural Network Optimized by PSO Algorithm

Size: px
Start display at page:

Download "Recommended Items Rating Prediction based on RBF Neural Network Optimized by PSO Algorithm"

Transcription

1 Recommended Items Ratng Predcton based on RBF Neural Network Optmzed by PSO Algorthm Chengfang Tan, Cayn Wang, Yuln L and Xx Q Abstract In order to mtgate the data sparsty and cold-start problems of recommendaton tem ratngs and more accurately predct the ratngs of recommended tems, accordng to the characterstcs of the non-lnearty and randomness n recommendaton content changes, recommended tems ratngs predcton model based on RBF neural network optmzed by PSO algorthm s proposed and the parameters of RBF neural network are globally optmzed by usng the proposed PSO algorthm. Expermental results show that, compared wth the state-of-the-art models ncludng the tradtonal user-based collaboratng flterng method, the tradtonal tem-based collaboratng flterng method and RBF neural network model, the proposed RBF neural network optmzed by PSO algorthm can more accurately predct the recommendaton tem ratngs, whch has hgher predcton accuracy and much lower predcton error measured by Root Mean Square Error (RMSE), Precson@N and Recall@N. Keywords PSO algorthm, RBF neural network, recommended tem ratng predcton, data sparsty. I. INTRODUCTION he rapd development of Internet trggers the problem of Tnformaton overload. As an mportant means to solve the problem of nformaton overload, the recommendaton system s wdely used n the felds of electronc commerce, manly flterng out the unrelated tems to allevate the burden of users. At present, the recommendaton system can be roughly categorzed nto the content-based recommendaton system and the collaboratve flterng recommendaton system. However, Ths work was supported by Key Projects of Excellent Young Talent Support Program n Colleges and Unverstes of Anhu Provnce (gxyqzd216344),unversty Students Innovatve Undertakng Tranng Program of Anhu Provnce ( , ), Teachng Research Project of Suzhou Unversty (szxy216jy32), the Key Projects of Anhu Provnce Colleges and Unverstes Natural Scence Foundaton of Chna (KJ216A768) and Soft Scence Research Project of Anhu Scence and Technology Plan n 216 (167a2271). Chengfang Tan s wth School of Informaton Engneerng, Suzhou Unversty, Suzhou 234, Anhu, Chna (correspondng author; e-mal: @qq.com). Cayn Wang s wth School of Informaton Engneerng, Suzhou Unversty, Suzhou 234, Anhu, Chna. Yuln L s wth School of Informaton Engneerng, Suzhou Unversty, Suzhou 234, Anhu, Chna. Xx Q s wth School of Informaton Engneerng, Suzhou Unversty, Suzhou 234, Anhu, Chna.. one of the bggest problems exstng n the collaboratve flterng recommendaton s that, as most users are not wllng to provde tem ratng nformaton or new tems do not have enough ratng data, the clusterng effect from smlar users s poor [1,2]. That s to say, under the naccurate clusterng class of smlar users, the cross recommendaton among users wll serously affect the recommendaton qualty and even produce wrong recommendaton, ultmately, data sparsty and cold start problems are produced. Amng at the above mentoned problems of collaboratve flterng recommendaton systems, many scholars have put forward ther vews from dfferent angles, and have acheved good results [3]. For example, reference 4 fused the cosne smlarty and Pearson correlaton wth Jaccard smlarty coeffcents method for the accurate measurement of the smlarty between users. In reference 5, the user's socal relaton s ntroduced nto the recommendaton system to realze the better recommendaton. Reference 6 combned the trust relatonshp between users wth the collaboratve flterng recommendaton, utlzed the explct trust relatonshp and the transton of trust relaton to calculate the trust degree between users, and selected the nearest neghbor based on the strength of trust relatonshp. Reference 7 ntegrated the matrx factorzaton model wth the trust relatonshp between users, exploted the nfluence and the transtvty of trust relatonshp to map users and tems nto the same feature space, then the preference predcton of users was performed n the feature space. Accordng to reference 8, the personal nformaton of users could reflect the user's nterest, and put forward adoptng the user's personal nformaton to complement the user's preference. Reference 9 put forward the concept of the same scorng matrx from users, based on the smlarty between the users and the same scorng matrx, the not evaluated tems were performed the preference predcton Motvated by the above observatons, we propose a new recommended tems ratng predcton method based on RBF (Radal Bass Functon) neural network optmzed by PSO algorthm (short for PSO-RBF). PSO algorthm not only has a strong global search capablty, but also s easy to be acheved. The parameters of RBF neural network optmzed by PSO algorthm are the global optmal parameters, whch can overcome the problem of low relablty of RBF neural network learnng and guarantee that the recommended tems ratngs are more accurately predcted. ISSN:

2 II. THE PROPOSED RECOMMENDED ITEMS RATING PREDICTION BASED ON RBF NEURAL NETWORK OPTIMIZED BY PSO ALGORITHM In the followng sectons, we frst ntroduce partcle swarm optmzaton algorthm and RBF neural network descrbed, respectvely. Then based on RBF neural network optmze by PSO algorthm, a PSO- recommended tems ratng predcton model s presented n detal. Lastly, we reports expermental results. 1. Partcle swarm optmzaton algorthm Partcle Swarm Optmzaton algorthm [1] s an evolutonary algorthm based on swarm ntellgence proposed by Kennedy and Eberhart n In PSO algorthm, the feasble soluton of each optmzaton problem s regarded as a partcle n the search space. Assume that there are N partcles n moton n D dmensonal space, v = ( v,1, v,2 vd, ) denotes the velocty vector of partcle, X= ( x,1, x,2 xd, ) represents the space poston vector of partcle. Partcle constantly adjusts ts speed and space poston accordng to v and X untl t reaches the global optmal soluton. Pbest denotes the optmal soluton of partcle tself and Pbest = ( p,1, p,2 p, D ). The optmal soluton for the populaton at present s expressed by g best and () t t t t g = ( g, g g ). best,1,2, D v = ( v,1, v,2 vd, ) s the velocty of partcle n tme t, then the speed of n the t + 1 th teraton s expressed as follows: ( t + 1) t ( t) ( t) v = wv + c1rand( )( Pb X ) + c2rand( )( gb X ) (1) Where w s the nerta factor and ts value s greater than, c1 s cogntve learnng factor and c 2 s socal learnng factor, c1 and c 2 are both postve constants, rand( ) s a functon that generates the random number rangng from to 1. The velocty value of the partcle s n the range of[ vmn, v max ]. If the velocty of a partcle s less than vmn or greater than v max, then the velocty value of the partcle s set to the correspondng threshold value. In the t + 1 th teraton, the poston of partcle s denoted as shown n Equaton (2), X = X + v (2) ( t+ 1) t ( t+ 1) Durng actually utlzng PSO algorthm, f the number of teratons reaches the maxmum value or the partcle fnally fnds the best optmal locaton n accordance wth the mnmum ftness threshold, then teraton s termnated [11]. 2. RBF neural network In 1998, Broomhead et al. proposed a radal bass functon based neural network (short for RBF) [12]. Smlar wth other neural networks, RBF neural network has a good local approxmaton, fast convergence and so on. The structure of RBF neural network s shown n Fg. 1 as follows: x 1 x 2 x n h 1 h m W y1 Hdden Layer Input Layer Output Layer Fg. 1. Structure of RBF neural network From Fg. 1, t can be observed that RBF neural network ncludes nput layer, hdden layer and output layer. It s a feed forward neural network wth the structure of three layers. Input layer can acheve the functon of smply transferrng the nput data nto the hdden layer, hdden layer neurons are able to make the nput data produce non-lnear mappng. Subsequently, neurons n the output layer perform lnear weghted operatons towards the nonlnear data output from the hdden layer, and ultmately results are output n a lnear style. As shown n Fg. 1, X = [ x1, x2,, x ] T n s an n-dmensonal nput vector, H = [ h1, h2,, h ] T m s a radal bass functons used n the hdden layer, whch denotes the output of hdden layer unt and s usually represented by usng the Gauss functon. The equaton s presented as follows, X C h ( x) = exp 2 = 1, 2,, m (3) 2b In Equaton (3), C represents the center of the frst th bass functon, b denotes the wdth of the th bass functon, and b s a number greater than. s the Eucldean paradgm, m denotes the number of nodes n the hdden layer. W = [ w1, w2,, w ] T m s the weghted vector of RBF neural network. The output of RBF neural network s a lnear combnaton of the output of the hdden layer nodes and s denoted as, y( x) = Wh( x) (4) k k k = 1 q In Equaton (4), k = 1, 2,, q and q s the number of nodes n the output layer, W k represents the connected weghts between the th node from the hdden layer and the node from the output layer. From the descrpton on RBF neural network, t can be vsble that, n the RBF neural network k th, Gauss bass functon center vectorc, Gauss bass functon wdth vector b, the number of radal bass functon center m and the connected weghts W k between the output layer and the hdden layer need to be determned. After determnng the above mentoned C, b and m, the least squares method can be used to drectly compute the connected weghts W k between the output layer and the hdden layer. C, b and W k y 2 y q ISSN:

3 determne the performance of the whole RBF neural network together. 3. The proposed recommended tems ratng predcton model From the descrpton of RBF neural network, t can be concluded that the predcton performance of RBF neural network manly depends on Gauss bass functon center vector C, Gauss bass functon wdth vector b, and connected weghts W k between the output layer and the hdden layer. However, the tradtonal RBF neural network adopts the local nformaton based on the parameter space to set the relatve parameters, and t causes that the values of parameter C, b and w k are the local optmal solutons, not the global optmal solutons. Amng at the above mentoned dsadvantages from RBF neural network, durng the process of predctng the changes of recommendaton tems ratng, we make use of PSO algorthm to perform the global optmzaton towards three parameters C, b and w k n RBF neural network. The proposed recommended tems ratng predcton algorthm based on RBF neural network optmzed by PSO algorthm (short for PSO-RBF) s ntroduced n detal as follows: Step 1: Intalze the partcle swarm and neural network. The pre-processed recommendaton samples are fed nto RBF neural network for tranng, and three above mentoned parameters C, b,and W k are obtaned. Parameters C, b and w k are respectvely regarded as the ntal poston of the partcle, the velocty of the partcle and the poston vector of the partcle to form one by one partcle. Step 2. Compute and evaluate the ftness value of each partcle. Suppose v and vˆ respectvely denote the true value and the predctve value of the recommended tem ratng at a certan tme. Assume that σ s the number of tranng examples, Ftness functon adpt s defned to measure the merts of the selecton parameters and ts equaton s denoted as follows, 1 adpt = σ = 1 v ˆ v v From Equaton (5), t can be represented that the negatve mean relatve error between the true value v and the predctve value vˆ of the recommended tem ratng s regard as a measure of the ftness value of each partcle. Step 3. Update the search poston of each partcle accordng to the ftness value of each ndvdual, and calculate the ndvdual extreme value pbest and the populaton global extreme value gbest of each partcle. The detaled operatons are as follows, for each partcle, we compare the ftness value of each partcle wth the best poston pbest that ths partcle has ever experenced, f the current ftness value of ths partcle s better, and then the current ftness value of the partcle s regarded as the current best poston pbest. For each partcle, (5) we compare the ftness value of each partcle wth the populaton global extreme gbest that all partcles have ever experenced, f t s better, and then gbest s set to the current poston of the best partcle. Step 4. Utlzng the characterstcs of PSO algorthm, the three parameters C, b and w k n RBF neural network are contnuously performed globally optmzed. If the teraton reaches the maxmum number or the mean square error reaches the ntal settng value, then the partcle search s termnated and the optmal partcle poston s output; otherwse returns step 3, the teratve optmzaton s executed repeatedly, untl the optmal values of the parameters are obtaned n RBF neural network. Step 5. Input the measured data to predct the change of the recommended tems ratng. The decoded value of the best poston that the swarm has experenced s regarded as the structural parameter of RBF neural network, and the decoded value s taken nto RBF neural network. We utlze another group of the recommended tems ratng to test the traned recommended tems ratng predcton model and the predcton effcency of the recommended tems s obtaned. III. EXPERIMENTAL RESULTS AND VERIFICATION In ths secton, we conduct comparatve experments to evaluate the effectveness and effcency of the proposed PSO- recommended tems ratng predcton model. Specfcally, we am to answer the followng two questons: (1) Can the proposed PSO- recommendaton method mprove the recommendaton performance by accurately predctng the ratng of recommended tems? (2) Is the proposed PSO- recommended method able to mtgate the cold-start problem for recommendaton by accurately predctng the ratng of tems? We begn by ntroducng two real-world datasets, the expermental settngs and the evaluaton metrcs we choose, then we compare the PSO- recommendaton method wth the three state-of-the-art baselnes to answer the frst queston and we explore the capablty of the proposed PSO-RBF method n handlng the cold-start problem to answer the second queston. All the algorthms were mplemented usng Java and all experments were performed on a server runnng Wndows Server 28 wth four 3.GHz CPU cores and 12GB memory. In ths paper, we explot the same ordnary data structures for all the algorthms and do not consder any parallel computaton. 1. Expermental datasets We collect two real-world datasets for the comparatve experments, among whch, one s from the socal meda webste Epnons ( that provdes the onlne servces for users, users can perform many nformaton sharng operatons on Epnons, for example, revewng and ratng the tems, etc. From the orgnally collected datasets, we flter out users who rated few tems and also tems that receved less than 1 ratngs. The fnal adopted dataset from Epnons s shown n Table 1. ISSN:

4 Table 1. Statstcal nformaton on Epnos Number of users Another dataset s from Movelens, whch ncludes about 1 mllon move ratng nformaton that 6,4 users rated the 3,952 moves wth scores from 1 to 5. The statstcs of the MoveLens dataset s shown n Table 2. Table 2. Statstcal nformaton on MoveLens Number Number Number of users of moves of ratngs MoveLens 6,4 3,952 About 1M 2. Evaluaton metrcs and expermental settng (1) Evaluaton Metrcs Frstly, we adopt the Root Mean Squared Error (RMSE) to measure the predcton error. RMSE s defned as follows [13]: 1 2 RMSE = ( rˆ j rj ) (6) N, j Where r j denotes the ratng of recommendaton tem j by user, rˆj denotes the correspondng ratng predcted by the model, and N denotes the total number of the test set. The smaller the value of RMSE, the more precse a recommendaton s. Secondly, rankng a recommendaton lst s often more mportant than the ratng predcton, so we also evaluate the algorthms n terms of rankng. To measure the accuracy of rankng a recommendaton lst, we adopt two standard metrcs Precson@N (denoted by P@N) and Recall@N (denoted by R@N). Specfcally, we thnk that a tem ratng s successfully predcted f t s n the recommendaton lst. P@N defnes the rato of successfully predcted tem topcs to the N recommendatons, and R@N defnes the rato of successfully predcted tem topcs to the number of tem topcs to be predcted. Generally, P@N and R@N are computed by the followng equatons: N P@ N N Nt R@ N N N t (7) t t Number of tems Number of ratngs Epnons 4, , ,824 Where N s the number of recommended tems, N s the t number of tems contaned n both the ground truth and the top-n results, N denotes the number of tems contaned n t the ground truth but not n the top-n results. Note that P@N and R@N are the averages of precson values and recall values, respectvely. We wll demonstrate how the performance of recommendaton models vares wth dfferent values of N n the (8) followng subsecton E. (2) Expermental Settng To nvestgate the capablty of the proposed PSO-RBF based recommendaton method, for both two datasets Epnons and Movelens, we random select 4%, 6% and 8% as the tranng set respectvely and the correspondng remanng 6%, 4% and 2% as the testng set. The random selecton process s carred out 1 tmes ndependently, and the average RMSE are obtaned. Note that parameters of the proposed PSO-RBF based recommendaton method are determned va cross valdaton. 3. Baselne methods We compare the proposed PSO-RBF method wth the three baselne recommendaton methods. Specfcally, the three baselne methods are presented as follows, (1) : refers to the user-based collaboratve flterng method and s a memory-based recommendaton method. manly adopts the neghborhood nformaton of users n the user-tem ratng matrx for supportng the whole recommendaton process. In ths paper, we use the cosne smlarty to calculate user-user smlarty. (2) : refers to the tem-based collaboratve flterng method and s also a memory-based recommendaton method. manly adopts the neghborhood nformaton of tems n the user-tem ratng matrx for supportng the whole recommendaton process. In ths paper, we use the Pearson correlaton to calculate tem-tem smlarty. (3) RBF neural network based method: RBF neural network s one of the neural networks. About the detaled nformaton of RBF neural network, please refer to the above mentoned ntroducton on RBF neural network. 4. Performance comparsons n terms of RMSE To answer the frst queston referred to n the begnnng of ths secton, we compare the proposed PSO- method wth the above three baselne methods on two real-world datasets. The comparatve results are shown n Table 3 on RMSE. Table 3. Performance comparson on Epnons and Movelens n terms of RMSE Epnons MoveLens Tranng sze PSO-RB F based 4% % % % % % From the comparatve results n Table 3, we can conclude the followng observatons: In general, under three dfferent tranng szes, method always outperforms method, and compared to other three baselne methods, the proposed PSO- method always obtans better performance than other three baselne methods and s more robust to the data sparsty problem. These results ndcate that RBF neural network optmzed by PSO algorthm can further ISSN:

5 mprove the recommendaton tem ratng predcton performance. Wth the ncreasng sze of the tranng set, we have enough expermental data to tran every model and every model becomes more stable, hence, the performances of the PSO- method and other three baselne methods are all mproved on RMSE to some extent. Va aforementoned analyss, we can draw an answer to the frst queston that RBF neural network optmzed by PSO algorthm not only can sgnfcantly mprove the recommendaton performance but also can mtgate the data sparsty problem n the recommendaton systems. In summary, the proposed PSO-RBF recommendaton method s superor to the smple collaboratve flterng method, because tranng from RBF neural network optmzed by PSO algorthm fltered out the ratng values wth greater evaluaton bas. 5. Performance comparsons n terms of P@N and R@N In order to further verfy the effectveness of utlzng PSO algorthm to optmze RBF neural network for evaluatng recommendaton lsts, through comparatve experments, P@N and R@N are utlzed to sort the predcted tems accordng to the ratngs and the values of N are set to 1, 15 and 2, respectvely. The comparatve expermental results are presented n Fg. 2, Fg. 3, Fg. 4 and Fg. 5 as follows. Fg. 3. R@N performance comparson on Epnons dataset (N=1, 15 and 2 respectvely) P@N Recommendaton lst sze Fg. 4. P@N performance comparson on Movelens dataset (N=1, 15 and 2 respectvely) PSO- PSO-.5 PSO- R@N P@N.3.1 Recommendaton lst sze Fg. 2. P@N performance comparson on Epnons dataset (N=1, 15 and 2 respectvely) R@N PSO- Recommendaton lst sze Recommendaton lst sze Fg. 5. R@N Performance comparson on Movelens dataset (N=1, 15 and 2 respectvely) We start wth dscussng the results showed n Fg. 2, Fg. 3, Fg. 4 and Fg. 5. The changes of all recommendaton models wth varyng recommendaton lst sze on Epnons and Movelens are presented. Obvously, for all models, larger recommendaton lst sze decreases the precson but ncreases the recall. That s to say, the larger the recommendaton lst sze, the lower the precson s and the hgher the recall s. Ths s because larger recommendaton lst sze ncreases the probablty that a user s tem rankng s ht by one of the recommended tems. The maxmum recommendaton lst sze s set to 2, because larger sze may affects the effectve nteracton between a user and the tem. In addton, when the value of N s fxed, our proposed PSO- recommendaton approach obvously outperforms, and method n terms of recall and precson. We carefully compare the values of P@N and R@N, for example, on Epnons dataset, the value of P@1 from the PSO- method s the largest compared wth method, method and method n Fg. 2. On Movelens dataset, the value of R@1 from the PSO-RBF based method s also the largest compared wth method, ISSN:

6 method and method n Fg. 5. And other comparsons about the values of P@N and R@N are also smlar. Ths s because predctng the mssng ratngs of tems makes the PSO- recommendaton method be able to accurately calculate the smlarty between tems, so as to mprove the accuracy of the recommendaton tem rankng predcton. Hence, we can conclude that the dea of compensaton values for tems mssng ratng can mprove the accuracy of the recommendaton lsts, and mtgates the data sparsty and cold-start problems n the tradtonal recommendaton systems to some extent. 6. Ablty of copng wth cold-start tems To answer the second queston mentoned n the begnnng of Secton 4, we nvestgate the capablty of the proposed PSO-RBF method n handlng cold-start problems. In detal, smlar wth the above dstrbuton of the tranng set and the test set, we randomly select 8% tems from the tranng set and remove ther tem ratngs from the tranng set to the test set. In ths way, these 8% of tems do not have any tem ratngs and we consder these 8% of tems as cold-start tems. The results on those tranng sets wth cold-start tems are lsted n Table 4 for RMSE. Table 4. Performance comparson on Epnons and Movelens wth 8% cold-start tems n terms of RMSE Epnons MoveLens Tranng sze 4% 6% 8% 4% 6% 8% RBF-based (-.2381) (-.2295) (-.182) (-.1794) (-.1539) (-.14) (-.199) (-.1986) (-.1971) (-.28) (-.1726) ( (-.2197) (-.2326) (-.1989) (-.1949) (-.1615) (-.1323) PSO-RBF based (-.179) (-.962) (-.848) (-.1162) (-.121) (-.983) Note that numbers nsde parentheses n Table 4 denote the performance reductons compared to the performance wthout cold-start users n Table 3. We analyze Table 4 and compare the performance wth Table 3 wthout cold-start tems, t can be concluded that the performance of all methods degenerates when we ntroduce cold-start tems. For example, when tranng sze s 8%, RMSE for PSO- method ncreases up to from n Table 3 on Epnons, that s to say, RMSE decreases by = RMSE for PSO- method ncreases up to from n Table 3 on Movelens, that s to say, RMSE decreases by = We carefully observe Table 3 and Table 4, the performance degeneraton of the proposed PSO- recommendaton method s much smaller compared to the other three baselne methods. These results support that the proposed PSO- recommendaton method can mtgate cold-start problems for the recommendaton. In summary, the ntroducton of cold-start tems could degrade the recommendaton performance and the proposed framework s relatvely more robust to cold-start tems by predctng the mssng ratngs of tems. IV. CONCLUSION RBF neural network optmzed by PSO algorthm s a regularzed network and can approach any contnuous functon, whch has a strong predctve power. In ths paper, RBF neural network optmzed by PSO algorthm s desgned to accurately predct the recommended tems ratngs, whch s dfferent from calculatng the user smlarty wth the same tems ratngs by users n the tradtonal user based recommendaton method. Compared wth,, and RBF-based method, the predctve result of the proposed PSO- method s more accurate. The comparatve results also show that the PSO- model has better stablty and s relatvely more robust to cold-start tems by predctng the mssng ratngs of tems than the exstng recommended tem ratng predcton. Our next research work wll ncorporate the smlarty of tems, the smlarty of users and the socal relatonshp among users nto the PSO- recommendaton model, and construct the correlatons among the smlarty of tems, the smlarty of users and the socal relatonshp among users whch further mprove the accuracy of the PSO- recommendaton model. REFERENCES [1] M. Deshpande, G. Karyps, Item-based Top-n Recommendaton Algorthms, ACM Transactons on Informaton Systems, vol. 22, no. 1, pp , 24 [2] Y. Koren, R. Bell, C. Volnsky, Matrx Factorzaton Technques for Recommender Systems, Computer, vol. 42, no. 8, pp. 3-37, 29 [3] J. Km, D. Lee, K. Chung, Item Recommendaton based on Context-aware Model for Personalzed U-healthcare Servce, Multmeda Tools and Applcatons, vol.71, no. 2, pp , 214 [4] K. Cho, Y. Suh, A New Smlarty Functon for Selectng Neghbors for Each Target Item n Collaboratve Flterng, Knowledge-Based System, vol. 37, no.1, pp , 213 [5] C. Cheng, H. Yang, I. Kng, M. R. Lyu, Fused Matrx Factorzaton wth Geographcal and Socal Influence n Locaton-based Socal Networks, Proceedngs of the Twenty-Sxth AAAI Conference on Artfcal Intellgence, pp , 212 [6] Y. Du, L. Huang, M. Ho, A Collaboratve Flterng Recommendaton Method Fuson Trust Computng, Pattern Recognton and Artfcal Intellgent, vol. 5, no. 1, pp , 214 [7] M. Duan, Collaboratve Flterng Recommendaton Algorthm based on Trust Propagaton, Internatonal Journal of Securty and Its Applcaton, vol. 9, no. 7, pp ,215 [8] C. Laa, D. Lub, C. Lnb, Novel Personal and Group-based Trust Models n Collaboratve Flterng for Document Recommendaton, Informaton Scence, pp.31-49, 213 [9] S. Rendle, L. S. Theme, Parwse Interacton Tensor Factorzaton for Personalzed Tag Recommendaton, Proceedngs of the thrd ACM Internatonal Conference on Web Search and Data Mnng, pp.81-9, 21 [1] J. Kennedy, R. C. Eberhart, Partcle Swarm Optmzaton, Proceedngs of IEEE Internatonal Conference on Neural Networks, pp , 1995 [11] S. Solomon, P. Thulasraman, R. Thulasram, Collaboratve Mult-swarm PSO for Task Matchng Usng Graphcs Processng Unts, Proceedngs of the 13th Annual Conference on Genetc and Evolutonary Computaton, pp , 211 ISSN:

7 [12] H. G. Han, Q. L. Chen, J. F. Qao, An Effcent Self-organzng RBF Neural Network for Water Qualty Predcton, Neural Networks, vol. 24, no. 7, pp , 211 [13] S. H. Wang, J. L. Tang, H. Lu, Toward Dual Roles of Users n Recommender Systems, Proceedngs of the 24th ACM Internatonal Conference on Informaton and Knowledge Manage., pp , 215 Chengfang Tan was born on Feb. 16, She receved the Master degree n computer scence and technology from Nanjng Normal Unversty of Chna. Currently, she s a researcher (assstant professor) at Suzhou Unversty, Chna. Her major research nterests nclude nformaton processng and data mnng. She has publshed many papers n related felds. Cayn Wang was born on Sept. 16, He receved the Master degree n computer scence and technology from Hefe Polytechnc Unversty, Chna.. Hs major research nterests nclude artfcal ntellgence. He has publshed many papers n related felds. Yuln L s an undergraduate student from the School of Computer Scence and Technology, Suzhou Unversty. Her research nterests nclude machne learnng and network congeston control. Xx Q s an undergraduate student from the School of Computer Scence and Technology, Suzhou Unversty. Her research nterests nclude computer networks and securty, machne learnng, and ntellgent nformaton processng. ISSN:

A Clustering Algorithm Solution to the Collaborative Filtering

A Clustering Algorithm Solution to the Collaborative Filtering Internatonal Journal of Scence Vol.4 No.8 017 ISSN: 1813-4890 A Clusterng Algorthm Soluton to the Collaboratve Flterng Yongl Yang 1, a, Fe Xue, b, Yongquan Ca 1, c Zhenhu Nng 1, d,* Hafeng Lu 3, e 1 Faculty

More information

Determining the Optimal Bandwidth Based on Multi-criterion Fusion

Determining the Optimal Bandwidth Based on Multi-criterion Fusion Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn

More information

The Research of Support Vector Machine in Agricultural Data Classification

The Research of Support Vector Machine in Agricultural Data Classification The Research of Support Vector Machne n Agrcultural Data Classfcaton Le Sh, Qguo Duan, Xnmng Ma, Me Weng College of Informaton and Management Scence, HeNan Agrcultural Unversty, Zhengzhou 45000 Chna Zhengzhou

More information

Cluster Analysis of Electrical Behavior

Cluster Analysis of Electrical Behavior Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School

More information

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

Learning the Kernel Parameters in Kernel Minimum Distance Classifier Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department

More information

Clustering Algorithm Combining CPSO with K-Means Chunqin Gu 1, a, Qian Tao 2, b

Clustering Algorithm Combining CPSO with K-Means Chunqin Gu 1, a, Qian Tao 2, b Internatonal Conference on Advances n Mechancal Engneerng and Industral Informatcs (AMEII 05) Clusterng Algorthm Combnng CPSO wth K-Means Chunqn Gu, a, Qan Tao, b Department of Informaton Scence, Zhongka

More information

Machine Learning 9. week

Machine Learning 9. week Machne Learnng 9. week Mappng Concept Radal Bass Functons (RBF) RBF Networks 1 Mappng It s probably the best scenaro for the classfcaton of two dataset s to separate them lnearly. As you see n the below

More information

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto

More information

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth

More information

Chinese Word Segmentation based on the Improved Particle Swarm Optimization Neural Networks

Chinese Word Segmentation based on the Improved Particle Swarm Optimization Neural Networks Chnese Word Segmentaton based on the Improved Partcle Swarm Optmzaton Neural Networks Ja He Computatonal Intellgence Laboratory School of Computer Scence and Engneerng, UESTC Chengdu, Chna Department of

More information

Complexity Analysis of Problem-Dimension Using PSO

Complexity Analysis of Problem-Dimension Using PSO Proceedngs of the 7th WSEAS Internatonal Conference on Evolutonary Computng, Cavtat, Croata, June -4, 6 (pp45-5) Complexty Analyss of Problem-Dmenson Usng PSO BUTHAINAH S. AL-KAZEMI AND SAMI J. HABIB,

More information

Journal of Chemical and Pharmaceutical Research, 2014, 6(6): Research Article. A selective ensemble classification method on microarray data

Journal of Chemical and Pharmaceutical Research, 2014, 6(6): Research Article. A selective ensemble classification method on microarray data Avalable onlne www.ocpr.com Journal of Chemcal and Pharmaceutcal Research, 2014, 6(6):2860-2866 Research Artcle ISSN : 0975-7384 CODEN(USA) : JCPRC5 A selectve ensemble classfcaton method on mcroarray

More information

Classifier Selection Based on Data Complexity Measures *

Classifier Selection Based on Data Complexity Measures * Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.

More information

Network Intrusion Detection Based on PSO-SVM

Network Intrusion Detection Based on PSO-SVM TELKOMNIKA Indonesan Journal of Electrcal Engneerng Vol.1, No., February 014, pp. 150 ~ 1508 DOI: http://dx.do.org/10.11591/telkomnka.v1.386 150 Network Intruson Detecton Based on PSO-SVM Changsheng Xang*

More information

Meta-heuristics for Multidimensional Knapsack Problems

Meta-heuristics for Multidimensional Knapsack Problems 2012 4th Internatonal Conference on Computer Research and Development IPCSIT vol.39 (2012) (2012) IACSIT Press, Sngapore Meta-heurstcs for Multdmensonal Knapsack Problems Zhbao Man + Computer Scence Department,

More information

An Improved Particle Swarm Optimization for Feature Selection

An Improved Particle Swarm Optimization for Feature Selection Journal of Bonc Engneerng 8 (20)?????? An Improved Partcle Swarm Optmzaton for Feature Selecton Yuannng Lu,2, Gang Wang,2, Hulng Chen,2, Hao Dong,2, Xaodong Zhu,2, Sujng Wang,2 Abstract. College of Computer

More information

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,

More information

Classifier Swarms for Human Detection in Infrared Imagery

Classifier Swarms for Human Detection in Infrared Imagery Classfer Swarms for Human Detecton n Infrared Imagery Yur Owechko, Swarup Medasan, and Narayan Srnvasa HRL Laboratores, LLC 3011 Malbu Canyon Road, Malbu, CA 90265 {owechko, smedasan, nsrnvasa}@hrl.com

More information

Available online at Available online at Advanced in Control Engineering and Information Science

Available online at   Available online at   Advanced in Control Engineering and Information Science Avalable onlne at wwwscencedrectcom Avalable onlne at wwwscencedrectcom Proceda Proceda Engneerng Engneerng 00 (2011) 15000 000 (2011) 1642 1646 Proceda Engneerng wwwelsevercom/locate/proceda Advanced

More information

BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET

BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET 1 BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET TZU-CHENG CHUANG School of Electrcal and Computer Engneerng, Purdue Unversty, West Lafayette, Indana 47907 SAUL B. GELFAND School

More information

Classifying Acoustic Transient Signals Using Artificial Intelligence

Classifying Acoustic Transient Signals Using Artificial Intelligence Classfyng Acoustc Transent Sgnals Usng Artfcal Intellgence Steve Sutton, Unversty of North Carolna At Wlmngton (suttons@charter.net) Greg Huff, Unversty of North Carolna At Wlmngton (jgh7476@uncwl.edu)

More information

A Binarization Algorithm specialized on Document Images and Photos

A Binarization Algorithm specialized on Document Images and Photos A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a

More information

Collaboratively Regularized Nearest Points for Set Based Recognition

Collaboratively Regularized Nearest Points for Set Based Recognition Academc Center for Computng and Meda Studes, Kyoto Unversty Collaboratvely Regularzed Nearest Ponts for Set Based Recognton Yang Wu, Mchhko Mnoh, Masayuk Mukunok Kyoto Unversty 9/1/013 BMVC 013 @ Brstol,

More information

K-means Optimization Clustering Algorithm Based on Hybrid PSO/GA Optimization and CS validity index

K-means Optimization Clustering Algorithm Based on Hybrid PSO/GA Optimization and CS validity index Orgnal Artcle Prnt ISSN: 3-6379 Onlne ISSN: 3-595X DOI: 0.7354/jss/07/33 K-means Optmzaton Clusterng Algorthm Based on Hybrd PSO/GA Optmzaton and CS valdty ndex K Jahanbn *, F Rahmanan, H Rezae 3, Y Farhang

More information

Research of Neural Network Classifier Based on FCM and PSO for Breast Cancer Classification

Research of Neural Network Classifier Based on FCM and PSO for Breast Cancer Classification Research of Neural Network Classfer Based on FCM and PSO for Breast Cancer Classfcaton Le Zhang 1, Ln Wang 1, Xujewen Wang 2, Keke Lu 2, and Ajth Abraham 3 1 Shandong Provncal Key Laboratory of Network

More information

A Notable Swarm Approach to Evolve Neural Network for Classification in Data Mining

A Notable Swarm Approach to Evolve Neural Network for Classification in Data Mining A Notable Swarm Approach to Evolve Neural Network for Classfcaton n Data Mnng Satchdananda Dehur 1, Bjan Bhar Mshra 2 and Sung-Bae Cho 1 1 Soft Computng Laboratory, Department of Computer Scence, Yonse

More information

An Image Fusion Approach Based on Segmentation Region

An Image Fusion Approach Based on Segmentation Region Rong Wang, L-Qun Gao, Shu Yang, Yu-Hua Cha, and Yan-Chun Lu An Image Fuson Approach Based On Segmentaton Regon An Image Fuson Approach Based on Segmentaton Regon Rong Wang, L-Qun Gao, Shu Yang 3, Yu-Hua

More information

Using Fuzzy Logic to Enhance the Large Size Remote Sensing Images

Using Fuzzy Logic to Enhance the Large Size Remote Sensing Images Internatonal Journal of Informaton and Electroncs Engneerng Vol. 5 No. 6 November 015 Usng Fuzzy Logc to Enhance the Large Sze Remote Sensng Images Trung Nguyen Tu Huy Ngo Hoang and Thoa Vu Van Abstract

More information

Simulation: Solving Dynamic Models ABE 5646 Week 11 Chapter 2, Spring 2010

Simulation: Solving Dynamic Models ABE 5646 Week 11 Chapter 2, Spring 2010 Smulaton: Solvng Dynamc Models ABE 5646 Week Chapter 2, Sprng 200 Week Descrpton Readng Materal Mar 5- Mar 9 Evaluatng [Crop] Models Comparng a model wth data - Graphcal, errors - Measures of agreement

More information

Feature Selection as an Improving Step for Decision Tree Construction

Feature Selection as an Improving Step for Decision Tree Construction 2009 Internatonal Conference on Machne Learnng and Computng IPCSIT vol.3 (2011) (2011) IACSIT Press, Sngapore Feature Selecton as an Improvng Step for Decson Tree Constructon Mahd Esmael 1, Fazekas Gabor

More information

Maximum Variance Combined with Adaptive Genetic Algorithm for Infrared Image Segmentation

Maximum Variance Combined with Adaptive Genetic Algorithm for Infrared Image Segmentation Internatonal Conference on Logstcs Engneerng, Management and Computer Scence (LEMCS 5) Maxmum Varance Combned wth Adaptve Genetc Algorthm for Infrared Image Segmentaton Huxuan Fu College of Automaton Harbn

More information

An Item-Targeted User Similarity Method for Data Service Recommendation

An Item-Targeted User Similarity Method for Data Service Recommendation 01 IEEE 16th Internatonal Enterprse Dstrbuted Object Computng Conference Workshops An Item-Targeted User Smlarty Method for Data Serce Recommendaton Cheng Zhang, Xaofang Zhao Insttute of Computng Technology

More information

Parallelism for Nested Loops with Non-uniform and Flow Dependences

Parallelism for Nested Loops with Non-uniform and Flow Dependences Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr

More information

THE PATH PLANNING ALGORITHM AND SIMULATION FOR MOBILE ROBOT

THE PATH PLANNING ALGORITHM AND SIMULATION FOR MOBILE ROBOT Journal of Theoretcal and Appled Informaton Technology 30 th Aprl 013. Vol. 50 No.3 005-013 JATIT & LLS. All rghts reserved. ISSN: 199-8645 www.jatt.org E-ISSN: 1817-3195 THE PATH PLANNING ALGORITHM AND

More information

Using Particle Swarm Optimization for Enhancing the Hierarchical Cell Relay Routing Protocol

Using Particle Swarm Optimization for Enhancing the Hierarchical Cell Relay Routing Protocol 2012 Thrd Internatonal Conference on Networkng and Computng Usng Partcle Swarm Optmzaton for Enhancng the Herarchcal Cell Relay Routng Protocol Hung-Y Ch Department of Electrcal Engneerng Natonal Sun Yat-Sen

More information

Query Clustering Using a Hybrid Query Similarity Measure

Query Clustering Using a Hybrid Query Similarity Measure Query clusterng usng a hybrd query smlarty measure Fu. L., Goh, D.H., & Foo, S. (2004). WSEAS Transacton on Computers, 3(3), 700-705. Query Clusterng Usng a Hybrd Query Smlarty Measure Ln Fu, Don Hoe-Lan

More information

Programming in Fortran 90 : 2017/2018

Programming in Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Exercse 1 : Evaluaton of functon dependng on nput Wrte a program who evaluate the functon f (x,y) for any two user specfed values

More information

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty

More information

Face Recognition University at Buffalo CSE666 Lecture Slides Resources:

Face Recognition University at Buffalo CSE666 Lecture Slides Resources: Face Recognton Unversty at Buffalo CSE666 Lecture Sldes Resources: http://www.face-rec.org/algorthms/ Overvew of face recognton algorthms Correlaton - Pxel based correspondence between two face mages Structural

More information

Image Emotional Semantic Retrieval Based on ELM

Image Emotional Semantic Retrieval Based on ELM Internatonal Conference on Logstcs Engneerng, Management and Computer Scence (LEMCS 2014) Image Emotonal Semantc Retreval Based on ELM Pele Zhang, Mn Yao, Shenzhang La College of computer scence & Technology

More information

CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION

CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION 24 CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION The present chapter proposes an IPSO approach for multprocessor task schedulng problem wth two classfcatons, namely, statc ndependent tasks and

More information

A Unified Framework for Semantics and Feature Based Relevance Feedback in Image Retrieval Systems

A Unified Framework for Semantics and Feature Based Relevance Feedback in Image Retrieval Systems A Unfed Framework for Semantcs and Feature Based Relevance Feedback n Image Retreval Systems Ye Lu *, Chunhu Hu 2, Xngquan Zhu 3*, HongJang Zhang 2, Qang Yang * School of Computng Scence Smon Fraser Unversty

More information

Your Neighbors Affect Your Ratings: On Geographical Neighborhood Influence to Rating Prediction

Your Neighbors Affect Your Ratings: On Geographical Neighborhood Influence to Rating Prediction Your Neghbors Affect Your Ratngs: On Geographcal Neghborhood Influence to Ratng Predcton Longke Hu lhu003@e.ntu.edu.sg Axn Sun axsun@ntu.edu.sg Yong Lu luy0054@e.ntu.edu.sg School of Computer Engneerng,

More information

The Codesign Challenge

The Codesign Challenge ECE 4530 Codesgn Challenge Fall 2007 Hardware/Software Codesgn The Codesgn Challenge Objectves In the codesgn challenge, your task s to accelerate a gven software reference mplementaton as fast as possble.

More information

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points; Subspace clusterng Clusterng Fundamental to all clusterng technques s the choce of dstance measure between data ponts; D q ( ) ( ) 2 x x = x x, j k = 1 k jk Squared Eucldean dstance Assumpton: All features

More information

FINDING IMPORTANT NODES IN SOCIAL NETWORKS BASED ON MODIFIED PAGERANK

FINDING IMPORTANT NODES IN SOCIAL NETWORKS BASED ON MODIFIED PAGERANK FINDING IMPORTANT NODES IN SOCIAL NETWORKS BASED ON MODIFIED PAGERANK L-qng Qu, Yong-quan Lang 2, Jng-Chen 3, 2 College of Informaton Scence and Technology, Shandong Unversty of Scence and Technology,

More information

Keywords - Wep page classification; bag of words model; topic model; hierarchical classification; Support Vector Machines

Keywords - Wep page classification; bag of words model; topic model; hierarchical classification; Support Vector Machines (IJCSIS) Internatonal Journal of Computer Scence and Informaton Securty, Herarchcal Web Page Classfcaton Based on a Topc Model and Neghborng Pages Integraton Wongkot Srura Phayung Meesad Choochart Haruechayasak

More information

BIN XIA et al: AN IMPROVED K-MEANS ALGORITHM BASED ON CLOUD PLATFORM FOR DATA MINING

BIN XIA et al: AN IMPROVED K-MEANS ALGORITHM BASED ON CLOUD PLATFORM FOR DATA MINING An Improved K-means Algorthm based on Cloud Platform for Data Mnng Bn Xa *, Yan Lu 2. School of nformaton and management scence, Henan Agrcultural Unversty, Zhengzhou, Henan 450002, P.R. Chna 2. College

More information

Optimal Design of Nonlinear Fuzzy Model by Means of Independent Fuzzy Scatter Partition

Optimal Design of Nonlinear Fuzzy Model by Means of Independent Fuzzy Scatter Partition Optmal Desgn of onlnear Fuzzy Model by Means of Independent Fuzzy Scatter Partton Keon-Jun Park, Hyung-Kl Kang and Yong-Kab Km *, Department of Informaton and Communcaton Engneerng, Wonkwang Unversty,

More information

EVALUATION OF THE PERFORMANCES OF ARTIFICIAL BEE COLONY AND INVASIVE WEED OPTIMIZATION ALGORITHMS ON THE MODIFIED BENCHMARK FUNCTIONS

EVALUATION OF THE PERFORMANCES OF ARTIFICIAL BEE COLONY AND INVASIVE WEED OPTIMIZATION ALGORITHMS ON THE MODIFIED BENCHMARK FUNCTIONS Academc Research Internatonal ISS-L: 3-9553, ISS: 3-9944 Vol., o. 3, May 0 EVALUATIO OF THE PERFORMACES OF ARTIFICIAL BEE COLOY AD IVASIVE WEED OPTIMIZATIO ALGORITHMS O THE MODIFIED BECHMARK FUCTIOS Dlay

More information

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for

More information

Performance Evaluation of Information Retrieval Systems

Performance Evaluation of Information Retrieval Systems Why System Evaluaton? Performance Evaluaton of Informaton Retreval Systems Many sldes n ths secton are adapted from Prof. Joydeep Ghosh (UT ECE) who n turn adapted them from Prof. Dk Lee (Unv. of Scence

More information

An Optimal Algorithm for Prufer Codes *

An Optimal Algorithm for Prufer Codes * J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,

More information

Data Mining For Multi-Criteria Energy Predictions

Data Mining For Multi-Criteria Energy Predictions Data Mnng For Mult-Crtera Energy Predctons Kashf Gll and Denns Moon Abstract We present a data mnng technque for mult-crtera predctons of wnd energy. A mult-crtera (MC) evolutonary computng method has

More information

Parallel Artificial Bee Colony Algorithm for the Traveling Salesman Problem

Parallel Artificial Bee Colony Algorithm for the Traveling Salesman Problem Parallel Artfcal Bee Colony Algorthm for the Travelng Salesman Problem Kun Xu, Mngyan Jang, Dongfeng Yuan The School of Informaton Scence and Engneerng Shandong Unversty, Jnan, 250100, Chna E-mal: xukun_sdu@163.com,

More information

Parameters Optimization of SVM Based on Improved FOA and Its Application in Fault Diagnosis

Parameters Optimization of SVM Based on Improved FOA and Its Application in Fault Diagnosis Parameters Optmzaton of SVM Based on Improved FOA and Its Applcaton n Fault Dagnoss Qantu Zhang1*, Lqng Fang1, Sca Su, Yan Lv1 1 Frst Department, Mechancal Engneerng College, Shjazhuang, Hebe Provnce,

More information

S1 Note. Basis functions.

S1 Note. Basis functions. S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type

More information

A Novel Adaptive Descriptor Algorithm for Ternary Pattern Textures

A Novel Adaptive Descriptor Algorithm for Ternary Pattern Textures A Novel Adaptve Descrptor Algorthm for Ternary Pattern Textures Fahuan Hu 1,2, Guopng Lu 1 *, Zengwen Dong 1 1.School of Mechancal & Electrcal Engneerng, Nanchang Unversty, Nanchang, 330031, Chna; 2. School

More information

Virtual Machine Migration based on Trust Measurement of Computer Node

Virtual Machine Migration based on Trust Measurement of Computer Node Appled Mechancs and Materals Onlne: 2014-04-04 ISSN: 1662-7482, Vols. 536-537, pp 678-682 do:10.4028/www.scentfc.net/amm.536-537.678 2014 Trans Tech Publcatons, Swtzerland Vrtual Machne Mgraton based on

More information

Video Proxy System for a Large-scale VOD System (DINA)

Video Proxy System for a Large-scale VOD System (DINA) Vdeo Proxy System for a Large-scale VOD System (DINA) KWUN-CHUNG CHAN #, KWOK-WAI CHEUNG *# #Department of Informaton Engneerng *Centre of Innovaton and Technology The Chnese Unversty of Hong Kong SHATIN,

More information

Data Mining: Model Evaluation

Data Mining: Model Evaluation Data Mnng: Model Evaluaton Aprl 16, 2013 1 Issues: Evaluatng Classfcaton Methods Accurac classfer accurac: predctng class label predctor accurac: guessng value of predcted attrbutes Speed tme to construct

More information

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION Paulo Quntlano 1 & Antono Santa-Rosa 1 Federal Polce Department, Brasla, Brazl. E-mals: quntlano.pqs@dpf.gov.br and

More information

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng

More information

High-Boost Mesh Filtering for 3-D Shape Enhancement

High-Boost Mesh Filtering for 3-D Shape Enhancement Hgh-Boost Mesh Flterng for 3-D Shape Enhancement Hrokazu Yagou Λ Alexander Belyaev y Damng We z Λ y z ; ; Shape Modelng Laboratory, Unversty of Azu, Azu-Wakamatsu 965-8580 Japan y Computer Graphcs Group,

More information

BioTechnology. An Indian Journal FULL PAPER. Trade Science Inc.

BioTechnology. An Indian Journal FULL PAPER. Trade Science Inc. [Type text] [Type text] [Type text] ISSN : 0974-74 Volume 0 Issue BoTechnology 04 An Indan Journal FULL PAPER BTAIJ 0() 04 [684-689] Revew on Chna s sports ndustry fnancng market based on market -orented

More information

IN recent years, recommender systems, which help users discover

IN recent years, recommender systems, which help users discover Heterogeneous Informaton Network Embeddng for Recommendaton Chuan Sh, Member, IEEE, Bnbn Hu, Wayne Xn Zhao Member, IEEE and Phlp S. Yu, Fellow, IEEE 1 arxv:1711.10730v1 [cs.si] 29 Nov 2017 Abstract Due

More information

A Self-adaptive Similarity-based Fitness Approximation for Evolutionary Optimization

A Self-adaptive Similarity-based Fitness Approximation for Evolutionary Optimization A Self-adaptve Smlarty-based Ftness Approxmaton for Evolutonary Optmzaton Je Tan Dvson of Industral and System Engneerng, Tayuan Unversty of Scence and Technology, Tayuan, 34 Chna College of Informaton

More information

Support Vector Machines

Support Vector Machines /9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.

More information

Research of Dynamic Access to Cloud Database Based on Improved Pheromone Algorithm

Research of Dynamic Access to Cloud Database Based on Improved Pheromone Algorithm , pp.197-202 http://dx.do.org/10.14257/dta.2016.9.5.20 Research of Dynamc Access to Cloud Database Based on Improved Pheromone Algorthm Yongqang L 1 and Jn Pan 2 1 (Software Technology Vocatonal College,

More information

Quality Improvement Algorithm for Tetrahedral Mesh Based on Optimal Delaunay Triangulation

Quality Improvement Algorithm for Tetrahedral Mesh Based on Optimal Delaunay Triangulation Intellgent Informaton Management, 013, 5, 191-195 Publshed Onlne November 013 (http://www.scrp.org/journal/m) http://dx.do.org/10.36/m.013.5601 Qualty Improvement Algorthm for Tetrahedral Mesh Based on

More information

Application of Improved Fish Swarm Algorithm in Cloud Computing Resource Scheduling

Application of Improved Fish Swarm Algorithm in Cloud Computing Resource Scheduling , pp.40-45 http://dx.do.org/10.14257/astl.2017.143.08 Applcaton of Improved Fsh Swarm Algorthm n Cloud Computng Resource Schedulng Yu Lu, Fangtao Lu School of Informaton Engneerng, Chongqng Vocatonal Insttute

More information

Optimizing SVR using Local Best PSO for Software Effort Estimation

Optimizing SVR using Local Best PSO for Software Effort Estimation Journal of Informaton Technology and Computer Scence Volume 1, Number 1, 2016, pp. 28 37 Journal Homepage: www.jtecs.ub.ac.d Optmzng SVR usng Local Best PSO for Software Effort Estmaton Dnda Novtasar 1,

More information

Signed Distance-based Deep Memory Recommender

Signed Distance-based Deep Memory Recommender Sgned Dstance-based Deep Memory Recommender ABSTRACT Personalzed recommendaton algorthms learn a user s preference for an tem, by measurng a dstance/smlarty between them. However, some of exstng recommendaton

More information

Fuzzy Rough Neural Network and Its Application to Feature Selection

Fuzzy Rough Neural Network and Its Application to Feature Selection 70 Internatonal Journal of Fuzzy Systems, Vol. 3, No. 4, December 0 Fuzzy Rough Neural Network and Its Applcaton to Feature Selecton Junyang Zhao and Zhl Zhang Abstract For the sake of measurng fuzzy uncertanty

More information

Outline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1

Outline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1 4/14/011 Outlne Dscrmnatve classfers for mage recognton Wednesday, Aprl 13 Krsten Grauman UT-Austn Last tme: wndow-based generc obect detecton basc ppelne face detecton wth boostng as case study Today:

More information

A Time-driven Data Placement Strategy for a Scientific Workflow Combining Edge Computing and Cloud Computing

A Time-driven Data Placement Strategy for a Scientific Workflow Combining Edge Computing and Cloud Computing > REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 1 A Tme-drven Data Placement Strategy for a Scentfc Workflow Combnng Edge Computng and Cloud Computng Bng Ln, Fangnng

More information

SHAPE RECOGNITION METHOD BASED ON THE k-nearest NEIGHBOR RULE

SHAPE RECOGNITION METHOD BASED ON THE k-nearest NEIGHBOR RULE SHAPE RECOGNITION METHOD BASED ON THE k-nearest NEIGHBOR RULE Dorna Purcaru Faculty of Automaton, Computers and Electroncs Unersty of Craoa 13 Al. I. Cuza Street, Craoa RO-1100 ROMANIA E-mal: dpurcaru@electroncs.uc.ro

More information

Professional competences training path for an e-commerce major, based on the ISM method

Professional competences training path for an e-commerce major, based on the ISM method World Transactons on Engneerng and Technology Educaton Vol.14, No.4, 2016 2016 WIETE Professonal competences tranng path for an e-commerce maor, based on the ISM method Ru Wang, Pn Peng, L-gang Lu & Lng

More information

The Shortest Path of Touring Lines given in the Plane

The Shortest Path of Touring Lines given in the Plane Send Orders for Reprnts to reprnts@benthamscence.ae 262 The Open Cybernetcs & Systemcs Journal, 2015, 9, 262-267 The Shortest Path of Tourng Lnes gven n the Plane Open Access Ljuan Wang 1,2, Dandan He

More information

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 1. SSDH: Semi-supervised Deep Hashing for Large Scale Image Retrieval

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 1. SSDH: Semi-supervised Deep Hashing for Large Scale Image Retrieval IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY SSDH: Sem-supervsed Deep Hashng for Large Scale Image Retreval Jan Zhang, and Yuxn Peng arxv:607.08477v2 [cs.cv] 8 Jun 207 Abstract Hashng

More information

Fast Computation of Shortest Path for Visiting Segments in the Plane

Fast Computation of Shortest Path for Visiting Segments in the Plane Send Orders for Reprnts to reprnts@benthamscence.ae 4 The Open Cybernetcs & Systemcs Journal, 04, 8, 4-9 Open Access Fast Computaton of Shortest Path for Vstng Segments n the Plane Ljuan Wang,, Bo Jang

More information

Learning-Based Top-N Selection Query Evaluation over Relational Databases

Learning-Based Top-N Selection Query Evaluation over Relational Databases Learnng-Based Top-N Selecton Query Evaluaton over Relatonal Databases Lang Zhu *, Wey Meng ** * School of Mathematcs and Computer Scence, Hebe Unversty, Baodng, Hebe 071002, Chna, zhu@mal.hbu.edu.cn **

More information

Feature-Based Matrix Factorization

Feature-Based Matrix Factorization Feature-Based Matrx Factorzaton arxv:1109.2271v3 [cs.ai] 29 Dec 2011 Tanq Chen, Zhao Zheng, Quxa Lu, Wenan Zhang, Yong Yu {tqchen,zhengzhao,luquxa,wnzhang,yyu}@apex.stu.edu.cn Apex Data & Knowledge Management

More information

Performance Assessment and Fault Diagnosis for Hydraulic Pump Based on WPT and SOM

Performance Assessment and Fault Diagnosis for Hydraulic Pump Based on WPT and SOM Performance Assessment and Fault Dagnoss for Hydraulc Pump Based on WPT and SOM Be Jkun, Lu Chen and Wang Zl PERFORMANCE ASSESSMENT AND FAULT DIAGNOSIS FOR HYDRAULIC PUMP BASED ON WPT AND SOM. Be Jkun,

More information

A Clustering Algorithm for Chinese Adjectives and Nouns 1

A Clustering Algorithm for Chinese Adjectives and Nouns 1 Clusterng lgorthm for Chnese dectves and ouns Yang Wen, Chunfa Yuan, Changnng Huang 2 State Key aboratory of Intellgent Technology and System Deptartment of Computer Scence & Technology, Tsnghua Unversty,

More information

Recommender System based on Higherorder Logic Data Representation

Recommender System based on Higherorder Logic Data Representation Recommender System based on Hgherorder Logc Data Representaton Lnna L Bngru Yang Zhuo Chen School of Informaton Engneerng, Unversty Of Scence and Technology Bejng Abstract Collaboratve Flterng help users

More information

Enhancement of Infrequent Purchased Product Recommendation Using Data Mining Techniques

Enhancement of Infrequent Purchased Product Recommendation Using Data Mining Techniques Enhancement of Infrequent Purchased Product Recommendaton Usng Data Mnng Technques Noraswalza Abdullah, Yue Xu, Shlomo Geva, and Mark Loo Dscplne of Computer Scence Faculty of Scence and Technology Queensland

More information

Optimizing Document Scoring for Query Retrieval

Optimizing Document Scoring for Query Retrieval Optmzng Document Scorng for Query Retreval Brent Ellwen baellwe@cs.stanford.edu Abstract The goal of ths project was to automate the process of tunng a document query engne. Specfcally, I used machne learnng

More information

Classifier Ensemble Design using Artificial Bee Colony based Feature Selection

Classifier Ensemble Design using Artificial Bee Colony based Feature Selection IJCSI Internatonal Journal of Computer Scence Issues, Vol. 9, Issue 3, No 2, May 2012 ISSN (Onlne): 1694-0814 www.ijcsi.org 522 Classfer Ensemble Desgn usng Artfcal Bee Colony based Feature Selecton Shunmugaprya

More information

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana

More information

The Study of Remote Sensing Image Classification Based on Support Vector Machine

The Study of Remote Sensing Image Classification Based on Support Vector Machine Sensors & Transducers 03 by IFSA http://www.sensorsportal.com The Study of Remote Sensng Image Classfcaton Based on Support Vector Machne, ZHANG Jan-Hua Key Research Insttute of Yellow Rver Cvlzaton and

More information

Comparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments

Comparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments Comparson of Heurstcs for Schedulng Independent Tasks on Heterogeneous Dstrbuted Envronments Hesam Izakan¹, Ath Abraham², Senor Member, IEEE, Václav Snášel³ ¹ Islamc Azad Unversty, Ramsar Branch, Ramsar,

More information

A New Hybrid Method Based on Improved Particle Swarm Optimization, Ant Colony Algorithm and HMM for Web Information Extraction

A New Hybrid Method Based on Improved Particle Swarm Optimization, Ant Colony Algorithm and HMM for Web Information Extraction A New Hybrd Method Based on Improved Partcle Swarm Optmzaton, Ant Colony Algorthm and HMM for Web Informaton Extracton Rong LI, Hong-bn WANG Department of Computer, Xnzhou Teachers Unversty, Xnzhou, Chna

More information

An Indian Journal FULL PAPER ABSTRACT KEYWORDS. Trade Science Inc.

An Indian Journal FULL PAPER ABSTRACT KEYWORDS. Trade Science Inc. [Type text] [Type text] [Type text] ISSN : 97-735 Volume Issue 9 BoTechnology An Indan Journal FULL PAPER BTAIJ, (9), [333-3] Matlab mult-dmensonal model-based - 3 Chnese football assocaton super league

More information

Corner-Based Image Alignment using Pyramid Structure with Gradient Vector Similarity

Corner-Based Image Alignment using Pyramid Structure with Gradient Vector Similarity Journal of Sgnal and Informaton Processng, 013, 4, 114-119 do:10.436/jsp.013.43b00 Publshed Onlne August 013 (http://www.scrp.org/journal/jsp) Corner-Based Image Algnment usng Pyramd Structure wth Gradent

More information

Learning a Class-Specific Dictionary for Facial Expression Recognition

Learning a Class-Specific Dictionary for Facial Expression Recognition BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 16, No 4 Sofa 016 Prnt ISSN: 1311-970; Onlne ISSN: 1314-4081 DOI: 10.1515/cat-016-0067 Learnng a Class-Specfc Dctonary for

More information

Wishing you all a Total Quality New Year!

Wishing you all a Total Quality New Year! Total Qualty Management and Sx Sgma Post Graduate Program 214-15 Sesson 4 Vnay Kumar Kalakband Assstant Professor Operatons & Systems Area 1 Wshng you all a Total Qualty New Year! Hope you acheve Sx sgma

More information

Outline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like:

Outline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like: Self-Organzng Maps (SOM) Turgay İBRİKÇİ, PhD. Outlne Introducton Structures of SOM SOM Archtecture Neghborhoods SOM Algorthm Examples Summary 1 2 Unsupervsed Hebban Learnng US Hebban Learnng, Cntd 3 A

More information

A New Feature of Uniformity of Image Texture Directions Coinciding with the Human Eyes Perception 1

A New Feature of Uniformity of Image Texture Directions Coinciding with the Human Eyes Perception 1 A New Feature of Unformty of Image Texture Drectons Concdng wth the Human Eyes Percepton Xng-Jan He, De-Shuang Huang, Yue Zhang, Tat-Mng Lo 2, and Mchael R. Lyu 3 Intellgent Computng Lab, Insttute of Intellgent

More information

Outline. Type of Machine Learning. Examples of Application. Unsupervised Learning

Outline. Type of Machine Learning. Examples of Application. Unsupervised Learning Outlne Artfcal Intellgence and ts applcatons Lecture 8 Unsupervsed Learnng Professor Danel Yeung danyeung@eee.org Dr. Patrck Chan patrckchan@eee.org South Chna Unversty of Technology, Chna Introducton

More information