The Impact of Feature Selection on Web Spam Detection
|
|
- Oscar Daniel
- 5 years ago
- Views:
Transcription
1 I.J. Itelliget Systems ad Applicatios, 2012, 9, Published Olie August 2012 i MECS ( -press.org/) DOI: /ijisa The Impact of Feature Selectio o Web Spam Detectio Jaber Karimpour Dept. of Computer Sciece, Uiversity of Tabriz, Tabriz, Ira karimpour@tabrizu.ac.ir Ali A. Noroozi Dept. of Computer Sciece, Uiversity of Tabriz, Tabriz, Ira aliasghar.oroozi@gmail.com Adeleh Abadi Dept. of Computer Sciece, Uiversity of Tabriz, Tabriz, Ira adeleh.abadi@gmail.com Abstract Search egie is oe of the most importat tools for maagig the massive amout of distributed web cotet. Web spammig tries to deceive search egies to rak some pages higher tha they deserve. May methods have bee proposed to combat web spammig ad to detect spam pages. Oe basic oe is usig classificatio, i.e., learig a classificatio model for classifyig web pages to spam or o-spam. This work tries to select the best feature set for classificatio of web spam usig imperialist competitive algorithm ad geetic algorithm. Imperialist competitive algorithm is a ovel optimizatio algorithm that is ispired by socio-political process of imperialis m i the real world. Experimets are carried out o W EBSPAM- UK2007 data set, which show feature selectio improves classificatio accuracy, ad imperialist competitive algorithm outperforms GA. Idex Terms Web Spam Detectio, Feature Selectio, Imperialistic Competitive Algorithm, Geetic Algorithm I. Itroductio With the explosive growth of iformatio o the web, it has become the most successful ad giat distributed computig applicatio today. Billios of web pages are shared by millios of orgaizatios, uiversities, researchers, etc. Web search provides great fuctioality for distributig, sharig, orgaizig, ad retrievig the growig amout of iformatio [1]. Search egies have become more ad more importat ad are used by millios of people to fid ecessary iformatio. It has become very importat for a web page, to be raked high i the importat search egies results. As a result may techiques are proposed to ifluece rakig ad improve the rak of a page. Some of these techiques are legal ad are called Search Egie Optimizatio (SEO) techiques, but some are ot legal or ethical ad try to deceive rakig algorithms. These spam pages try to rak pages higher tha they deserve [2]. Web spam refers to web cotet that get high rak i search egie results despite low iformatio value. Spammig ot oly misleads users, but also imposes time ad space cost to search egie crawlers ad idexers. That is why crawlers try to detect web spam pages to avoid processig ad idexig them. Cotet-based spammig methods basically tailor the cotets of the text fields i HTML pages to make spam pages more relevat to some queries. This kid of spammig is also called term spammig. There are two mai cotet spammig techiques, which simply create sythetic cotets cotaiig spam terms: repeatig some importat terms ad dumpig may urelated terms [3,4]. Lik spammig misuses lik structure of the web to spam pages. There are two mai kids of lik spammig. Out-lik spammig tries to boost the hub score of a page by addig out-liks i it poitig to some authoritative pages. Oe of the commo techiques of this kid of spammig is directory cloig, i.e., replicatig a large portio of a directory like Yahoo! i the spam page. I-lik spammig refers to persuadig other pages, especially authoritative oes, to poit to the spam page. I order to do this, a spammer might adopt these strategies: creatig a hoey pot, ifiltratig a web directory, postig liks o usergeerated cotet, participatig i lik exchage, buyig expired domais, ad creatig ow spam farm [2]. Hidig techiques are also used by spammers who wat to coceal or to hide the spammig seteces, terms, ad liks so that web users do ot see those [3]. Cotet hidig is used to make spam items ivisible. Oe simple method is to make the spam terms the same color as the page backgroud color. I cloakig, spam
2 62 The Impact of Feature Selectio o Web Spam Detectio web servers retur a HTML documet to the user ad a differet documet to a web crawler. I redirectig, a spammer ca hide the spammed page by automatically redirectig the browser to aother URL as soo as the page is loaded. I two latter techiques, the spammer ca preset the user with the iteded cotet ad the search egie with spam cotet [5]. Various methods have bee proposed to combat web spammig ad to detect spam pages. Oe importat ad basic type of methods is cosiderig web spam detectio as a biary classificatio problem [4]. I this kid of methods, some web pages are collected as traiig data ad labeled as spam or o-spam by a expert. The, a classifier model is leared from the traiig data. Oe ca use ay supervised learig algorithm to build this model. Further, the model is used to classify ay web page to spam or o-spam. The key issue is to desig features used i learig. Ntoulas et al. [4] propose some cotet-based features to detect cotet spam. Lik-based features are proposed for lik spam detectio [6,7]. Liu et al. [8] propose some user behavior features extracted from access logs of web server of a page. These features depict user behavior patters whe reachig a page (spam or o-spam). These patters are used to separate spam pages from o-spam oes, regardless of spammig techiques used. Erdelyi et al. [9] ivestigate the tradeoff betwee feature geeratio ad spam classificatio accuracy. They coclude that more features achieve better performace; however, the appropriate choice of the machie learig techiques for classificatio is probably more importat tha devisig ew complex features. Feature selectio is the process of fidig a optimal subset of features that cotribute sigificatly to the classificatio. Selectig a small subset of features ca decrease the cost ad the ruig time of a classificatio system. It may also icrease the classificatio accuracy because irrelevat or redudat features are removed [10]. Amog the may methods proposed for feature selectio, evolutioary optimizatio algorithms such as geetic algorithm (GA) have gaied a lot of attetio. Geetic algorithm has bee used as a efficiet feature selectio method i may applicatios [11,16]. I this paper, we icorporate geetic algorithm, ad imperialist competitive algorithm [12] to fid a optimal subset of features of the WEBSPAM-UK2007 data set [13,14]. The selected features are used for classificatio of the WEBSPAM-UK2007 data. The rest of the paper is orgaized as follows. Sectio 2 gives a brief itroductio of the imperialist competitive algorithm (ICA). Sectio 3 describes the feature selectio process by ICA ad GA. Experimetal results are discussed i sectio 4, ad fially, sectio 5 cocludes the paper. II. Imperialistic Competitive Algorithm The imperialist competitive algorithm is ispired by imperialis m i the real world [12]. Imperialism is the policy of extedig the power of a coutry beyod its boudaries ad weakeig other coutries to take cotrol of them. Fig. 1 Iitializatio of the empires: The more coloies a imperialist possesses, the bigger is its mark [12] This algorithm starts with a iitial society of radom geerated coutries. Some of the best coutries are selected to be imperialists ad others are selected to be coloies of these imperialists. The power of a empire which is the couterpart of fitess value i geetic algorithms, is the power of the imperialist coutry plus a percetage of mea power of its coloies. Figure 1 depicts the iitializatio of the empires. After assigig all coutries to imperialists, ad formig empires, coloies start movig towards the relevat imperialist (Assimilatio). The, some coutries radomly chage positio i the search space (Revolutio). After assimilatio ad revolutio, a coloy may get a better positio i the search space ad take cotrol the empire (substitutio for the imperialist). Fig. 2 Imperialistic competitio: The weakest coloy of the weakest empire is possessed by other empires [12] The, imperialistic competitio begis. All empires try to take cotrol of the weakest coloy of the weakest empire. This competitio reduces the power of weaker empires ad icreases the power of the powerful oes. Ay empire that caot compete with other empires ad icrease its power or at least prevet decreasig it, will
3 The Impact of Feature Selectio o Web Spam Detectio 63 gradually collapse. As a result, after some iteratios, the algorithm coverges ad oly oe imperialist remais ad all other coutries are coloies of it. Figure 2 depicts the imperialistic competitio. The more powerful a empire is, the more likely it will take cotrol of the weakest coloy of the weakest empire. The pseudo code of ICA is as follows: 1. Iitialize the empires 2. Assimilatio: Move the coloies toward their relevat imperialist 3. Revolutio: Radomly chage the characteristics of some coloies 4. Exchage the positio of a coloy ad Imperialist. If a coloy has more power tha that of imperialist, exchage the positios of that coloy ad the imperialist 5. Compute the total power of all empires 6. Imperialistic competitio: Give the weakest coloy from the weakest empire to the empire that has the most likelihood to possess it 7. Elimiate the powerless empires 8. If there is just oe empire, stop, else, go to 2 III. Feature Selectio WEBSPAM-UK2007 data set cotais 96 cotet based features. We use the imperialist competitive ad geetic algorithms to optimize the features that cotribute sigificatly to the classificatio. A. Feature Selectio Usig ICA I this sectio, the steps of feature selectio usig ICA are described. 1) Iitialize the empires I the geetic algorithm, each solutio to a optimizatio problem is a array, called chromosome. I ICA, this array is called coutry. I feature selectio, each coutry is a array of biary umbers. Whe coutry[i] is 1, the i th feature is selected for classificatio, ad whe it is 0, the i th feature is removed [15]. Figure 3 depicts the feature represetatio as a coutry. F F F F -1 Coutry F Feature subset = {F, F,..., F } Fig. 3 Feature represetatio as a coutry i ICA [15] The power of each coutry is calculated by F-score. F-score is a commoly used measure i machie learig ad iformatio retrieval [3,10]. The cofusio matrix of a give a classifier is cosidered as table 1. Table 1. Cofusio Matrix Classified spam Classified o-spam Actual spam A B Actual o-spam C D F-score is determied as follows F-score = 1 / (1 / Recall + 1 / Precisio) (1) Where Recall, ad Precisio are defied as follows Recall = C / (C + D) (2) Precisio = B / (B + D) (3) The algorithm starts by radomly iitializig a populatio of size N. N of the most powerful pop imp coutries are selected as imperialists ad form the emp ires. The remaiig coutries ( N col ) are assiged to empires based o the power of each empire. The ormalized power of each imperialist is defied by NP P N imp i 1 P i Where P is the power of coutry. The iitial umber of coloies of col empire will be (4) NC roud { NP * N } (5) To assig coloies to empires, is chose radomly ad assiged to coloies alog with the NC of the coloies imperialist will form imperialist. These empire. 2) Assimilatio I this phase, coloies move towards the relevat imperialist. Sice feature selectio is a discrete problem, we use followig operator for assimilatio [15] For each coloy Create a biary strig ad assig a radom geerated biary to each cell Copy the cells of the relevat imperialist, correspodig to the locatio of 1 s i the biary strig, to the same positios i the coloy 3) Revolutio The purpose of revolutio is preservig ad itroducig diversity. It allows the algorithm to avoid local miimum. Revolutio occurs accordig to a user defied revolutio probability. For each coloy, some cells are selected radomly ad their cotaiig biary is iverted ( 1 is iverted to 0, ad 0 is iverted to 1 ).
4 64 The Impact of Feature Selectio o Web Spam Detectio 4) Exchage the positios of a coloy ad imperialist After assimilatio ad revolutio, a coloy may gai more power tha that of imperialist. As a result, the best coloy of a empire ad its imperialist exchage positios. The, the algorithm will cotiue by the imperialist i a ew positio. 5) Compute the total power of empires The total power of a empire is maily affected by the power of its imperialist. Aother factor i computig the total power of a empire is the power of coloies of that empire. Of course, the mai power is by the power of the imperialist, ad the power of coloies has less impact. As a result, we defie the total power of empire is defied TP power ( imperialist ) mea{ power ( coloies of empire )} (6) Where is a positive factor which is cosidered to be less tha 1. Decreasig the value of icreases the role of the imperialist i determiig the total power of a empire ad icreasig it will icrease the role of the coloies. 6) Imperialistic Competitio I this importat phase of the algorithm, the empires compete to take cotrol of the weakest coloy of the weakest empire. Each empire has a likelihood of possessig the metioed coloy. The possessio probability of empire is obtaied by P emp TP i 1 N imp TP i As you ca otice, the most powerful empire does ot take possessio of the weakest coloy of the weakest empire, but it will be more likely to possess the metioed coloy. 7) Elimiate the powerless empires Imperialistic competitio causes some empires to lose power ad gradually collapse. Whe a empire loses all its coloies, we assume it is collapsed ad elimiate it. The imperialist of this powerless empire is possessed by other empires as a coloy. 8) Covergece As a result of imperialistic competitio ad elimiatio of powerless empires, the algorithm will coverge to the most powerful empire ad all the coutries will be uder the cotrol of this empire. The imperialist of this empire will determie the optimal subset of features selected for classificatio, because this imperialist is the most powerful of all coutries. B. Feature selectio usig GA I the geetic algorithm, each solutio to the feature selectio problem is a strig of biary umbers, called (7) chromosome. Whe chromosome[i] is 1, the i th feature is selected for classificatio, ad whe it is 0, the i th feature is ot selected [11,16]. The fitess fuctio is cosidered the accuracy of the classificatio model. I this research, we calculate the fitess value of each chromosome by F-score. F-score was described i the previous sectio. The algorithm starts by radomly iitializig a populatio of size N. The, crossover ad mutatio are doe. pop Crossover allows the geeratio of ew chromosomes by combiig curret best chromosomes. To do crossover, sigle poit crossover techique is used, i.e., oe crossover poit is selected, biary strig from begiig of chromosome to the crossover poit is copied from oe paret, the rest is copied from the secod paret. Figure 4 shows how childre are geerated from each pair of chromosomes by crossover. Mutatio is similar to revolutio i ICA. It maitais geetic diversity ad allows the algorithm to avoid local miimum. To do mutatio, i each chromosome, a radom cell is selected ad its cotaiig bit i iverted ( 1 is iverted to 0, ad 0 is iverted to 1 ). Mutatio ad crossover occur accordig to a previously defied mutatio ad crossover probability. Geetic algorithm iterates for some user defied umber of geeratios. Fig. 4 how childre are geerated from parets by crossover [17] IV. Experimetal Results I order to ivestigate the impact of feature selectio o web spam classificatio, W EBSPAM-UK2007 data are used. It is a publicly available web spam data collectio ad is based o a crawl of the.uk domai doe i May 2007 [13, 14]. It icludes 105 millio pages ad over 3 billio liks i hosts. The traiig set cotais 3849 hosts. This data set cotais cotet ad lik based features. I our experimets, we used oly cotet based features because they were eough to meet our purposes. The selected data set cotais 3849 data, with 208 spam ad 3641 o-spam pages. We partitioed this data set to two disjoit sets: traiig data set with 2449 data, ad test data set with 1000 data. After performig feature
5 The Impact of Feature Selectio o Web Spam Detectio 65 selectio usig the traiig set, the test set was used to evaluate the selected subset of features. The evaluatio of the overall process was based o weighted f-score which is a suitable measure for the spam classificatio problem. It was also used as the power fuctio i ICA ad fitess fuctio i GA. Bayesia Network, Decisio Tree (C4.5 algorithm), ad Support Vector Machie (SVM) were chose as learig algorithms to perform the classificatio ad calculate the weighted F-score. These algorithms a re powerful learig algorithms used i may web spam detectio researches [4, 5, 18]. Followig parameters were used for ICA Number of coutries = 100 Number of imperialists = 10 = 0.1 Revolutio rate = 0.01 Selected parameters for GA are as follows Iitial populatio = 100 Number of geeratios (iteratios) = 100 Crossover rate = 0.6 Mutatio rate = 0.01 Figure 5 depicts maximum ad mea power of all imperialists versus iteratio, usig Decisio Tress, SVM, ad Bayesia Network classifiers, i ICA. As show i this figure, by SVM ad Decisio Tree classifiers, the global maximum of the fuctio (maximum power) is foud i less tha 5 iteratios, while by Bayesia Network, it is foud i 12 th iteratio. Fig. 5 Mea ad maximum power of all imperialists versus iteratio, usig differet classifiers, i ICA Figures 6, ad 7 compare ICA power fuctio ad GA fitess fuctio versus iteratio. Figure 6 shows the power (fitess) of best aswer versus iteratio (geeratio), usig Bayesia Network classifier, i ICA ad GA. As you ca see, ICA coverges faster tha GA, ad has more power tha GA i all iteratios. Aother importat poit is that the iitial value of f-score which is the result of radom iitializatio of populatio i both algorithms, gets a higher icrease by ICA over iteratios. This poit shows that imperialistic competitio outperforms geetic evolutio i the problem of spam classificatio. Fig. 6 Power (fitess) of best aswer versus iteratio, usig Bayesia Network classifier
6 66 The Impact of Feature Selectio o Web Spam Detectio Fig. 7 Mea power (fitess) of all aswers versus iteratio, usig Bayesia Network classifier Figure 7 depicts mea power (fitess) of all aswers versus iteratio, usig Bayesia Network classifier, i ICA ad GA. As you ca see, ICA gets a higher icrease i mea power of all aswers. The optimal subset of features selected by ICA ad GA are used to trai a classificatio model. This model is evaluated by the test data set. Evaluatio results obtaied for Bayesia Network, Decisio Tree, ad SVM classifiers are show i table 2. These results idicate that feature selectio by both ICA ad GA techiques improves web spam classificatio. Furthermore, ICA based feature selectio outperforms GA based feature selectio i the problem of web spam detectio. Table 2 The impact of ICA ad GA based feature selectio o web spam classificatio, usig differet classifiers Bayesia Network Decisio Tress SVM Number Number of Number of of F-score F-score features features features F-score All features GA ICA V. Coclusio I this paper, we studied the impact of feature selectio o the problem of web spam classificatio. Feature selectio was performed by Imperialist Competitive Algorithm ad Geetic Algorithm. Experimetal results showed that selectig a optimal subset of features icreases classificatio accuracy, but ICA could fid better optimal aswers tha GA. I fact, we observed that reducig the umber of features decreases the classificatio cost ad icreases the classificatio accuracy. Other optimizatio methods, such as PSO ad at coloy ca be used for feature selectio ad compared with ICA ad GA i future works. Refereces [1] Caverlee J, Liu L, Webb S. A Parameterized Approach to Spam-Resiliet Lik Aalysis of the Web. IEEE Trasactios o Parallel ad Distributed Systems (TPDS), 2009, 20: [2] Gyogyi Z,Garcia-Molia H. Web spam taxoomy. I: First iteratioalworkshop o adversarial iformatio retrieval o the web (AIRWeb 05), Japa, [3] Liu B. Web Data Miig, Explorig Hyperliks, Cotets, ad Usage Data. Spriger, [4] Ntoulas A, Najork M, Maasse M, et al. Detectig Spam Web Pages through Cotet Aalysis. I Proc. of the 15th Itl. World Wide Web Coferece (WWW 06), [5] Wag W, Zeg G, Tag D. Usig evidece based cotet trust model for spam detectio. Expert Systems with Applicatios, (8): [6] Becchetti L, Castillo C, Doato D, et al. Lik-based characterizatio ad detectio of Web Spam. I Proc. Of 2d It. Workshop o Adversarial
7 The Impact of Feature Selectio o Web Spam Detectio 67 Iformatio Retrieval o the Web (AIRWeb 06), Seattle, WA, [7] Castillo C, Doato D, Giois A, et al. Kow your eighbors: Web spam detectio usig the web topology. I Proc. Of 30th Au. It. ACM SIGIR Cof. Research ad Developmet i Iformatio Retrieval (SIGIR 07), New York, [8] Liu Y, Ce R, Zhag M, et al. Idetifyig web spam with user behavior aalysis. I Proc. Of 4th It. Workshop o Adversarial Iformatio Retrieval o the Web (AIRWeb 08), Chia, [9] Erdelyi M, Garzo A, Beczur A A. Web spam classificatio: a few features worth more. I Proceedigs of the 2011 Joit WICOW/AIRWeb Workshop o Web Quality2011, Idia, [10] Ha J, Kaber M, Pei J. Data Miig, Cocepts ad Techiques. 3 rd ed, Morga Kaufma, [11] Vafaie H, De Jog K. Geetic algorithms as a tool for feature selectio i machie learig. I Proceedigs of Fourth Iteratioal Coferece o Tools with Artificial Itelligece (TAI '92), [12] Atashpaz-Gargari E, Lucas C. Imperialist competitive algorithm: A algorithm for optimizatio ispired by imperialistic competitio. IEEE Cogress o Evolutioary Computatio (CEC 2007), [13] Castillo C, Doato D, Becchetti L, et al. A referece collectio for webspam. SIGIR Forum, 2006, 40(2): [14] Yahoo Research. Web Spam Collectios. [cited 2011 May], Available from: ets/, 2007 [15] Mousavi Rad S J, Mollazade K, Akhlagia Tab F. Applicatio of Imperialist Competitive Algorithm for Feature Selectio: A Case Study o Bulk Rice Classificatio. Iteratioal Joural of Computer Applicatios, (16):41-48 [16] Yag J, Hoavar V. Feature subset selectio usig a geetic algorithm. Itelliget Systems ad their Applicatios, IEEE, (2): [17] Eibe A E, Smith J E. Itroductio to Evolutioary Computig, Spriger, [18] Araujo L, Martiez-Romo J. Web Spam Detectio: New Classificatio Features Based o Qualified Lik Aalysis ad Laguage Models. IEEE Trasactios o Iformatio Foresics ad Security, (3): KARIMPOUR Jaber ( ), male, Tabriz, Ira, Assistat Professor, his research directios iclude verificatio ad formal methods. NOROOZI Ali A. (1986-), male, Tabriz, Ira, Master of Sciece, his research directios iclude adversarial iformatio retrieval ad distributed systems. ABADI Adeleh ( ) female, Tabriz, Ira, Master of Sciece, his research directios iclude verificatio ad formal methods. How to cite this paper: Jaber Karimpour,Ali A. Noroozi,Adeleh Abadi,"The Impact of Feature Selectio o Web Spam Detectio", Iteratioal Joural of Itelliget Systems ad Applicatios(IJISA), vol.4, o.9, pp.61-67, DOI: /ijisa
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 informationAn 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 informationOnes 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 informationFuzzy 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 information6.854J / 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 informationDynamic 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 informationA 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 informationEuclidean 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 informationPattern 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 informationIntroduction. 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 informationMorgan 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 informationA 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 informationEvaluation of Support Vector Machine Kernels for Detecting Network Anomalies
Evaluatio of Support Vector Machie Kerels for Detectig Network Aomalies Prera Batta, Maider Sigh, Zhida Li, Qigye Dig, ad Ljiljaa Trajković Commuicatio Networks Laboratory http://www.esc.sfu.ca/~ljilja/cl/
More informationHeuristic 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 informationISSN (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 informationAnalysis of Documents Clustering Using Sampled Agglomerative Technique
Aalysis of Documets Clusterig Usig Sampled Agglomerative Techique Omar H. Karam, Ahmed M. Hamad, ad Sheri M. Moussa Abstract I this paper a clusterig algorithm for documets is proposed that adapts a samplig-based
More informationOptimization 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 informationSectio 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 informationA 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 informationHADOOP: 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 informationCIS 121 Data Structures and Algorithms with Java Spring Stacks, Queues, and Heaps Monday, February 18 / Tuesday, February 19
CIS Data Structures ad Algorithms with Java Sprig 09 Stacks, Queues, ad Heaps Moday, February 8 / Tuesday, February 9 Stacks ad Queues Recall the stack ad queue ADTs (abstract data types from lecture.
More informationWeb Text Feature Extraction with Particle Swarm Optimization
32 IJCSNS Iteratioal Joural of Computer Sciece ad Network Security, VOL.7 No.6, Jue 2007 Web Text Feature Extractio with Particle Swarm Optimizatio Sog Liagtu,, Zhag Xiaomig Istitute of Itelliget Machies,
More informationMałgorzata Sterna. Mateusz Cicheński, Mateusz Jarus, Michał Miszkiewicz, Jarosław Szymczak
Małgorzata Stera Mateusz Cicheński, Mateusz Jarus, Michał Miszkiewicz, Jarosław Szymczak Istitute of Computig Sciece Pozań Uiversity of Techology Pozań - Polad Scope of the Talk Problem defiitio MP Formulatio
More informationImage 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 informationVariance 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 informationarxiv: 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 informationCubic Polynomial Curves with a Shape Parameter
roceedigs of the th WSEAS Iteratioal Coferece o Robotics Cotrol ad Maufacturig Techology Hagzhou Chia April -8 00 (pp5-70) Cubic olyomial Curves with a Shape arameter MO GUOLIANG ZHAO YANAN Iformatio ad
More informationA Novel Hybrid Algorithm for Software Cost Estimation Based on Cuckoo Optimization and K-Nearest Neighbors Algorithms
Egieerig, Techology & Applied Sciece Research Vol. 6, No. 3, 26, 8-22 8 A Novel Hybrid Algorithm for Software Cost Estimatio Based o Optimizatio ad K-Nearest Neighbors Algorithms Elaz Eskadaria Miadoab
More informationECE4050 Data Structures and Algorithms. Lecture 6: Searching
ECE4050 Data Structures ad Algorithms Lecture 6: Searchig 1 Search Give: Distict keys k 1, k 2,, k ad collectio L of records of the form (k 1, I 1 ), (k 2, I 2 ),, (k, I ) where I j is the iformatio associated
More informationSearching a Russian Document Collection Using English, Chinese and Japanese Queries
Searchig a Russia Documet Collectio Usig Eglish, Chiese ad Japaese Queries Fredric C. Gey (gey@ucdata.berkeley.edu) UC Data Archive & Techical Assistace Uiversity of Califoria, Berkeley, CA 94720 USA ABSTRACT.
More informationAn 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 informationLecture 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 informationn n B. How many subsets of C are there of cardinality n. We are selecting elements for such a
4. [10] Usig a combiatorial argumet, prove that for 1: = 0 = Let A ad B be disjoit sets of cardiality each ad C = A B. How may subsets of C are there of cardiality. We are selectig elemets for such a subset
More informationAlgorithms 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 informationSOFTWARE 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 informationStructuring 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 informationHarris 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 informationProbabilistic 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 informationFREQUENCY ESTIMATION OF INTERNET PACKET STREAMS WITH LIMITED SPACE: UPPER AND LOWER BOUNDS
FREQUENCY ESTIMATION OF INTERNET PACKET STREAMS WITH LIMITED SPACE: UPPER AND LOWER BOUNDS Prosejit Bose Evagelos Kraakis Pat Mori Yihui Tag School of Computer Sciece, Carleto Uiversity {jit,kraakis,mori,y
More informationAdaptive 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 informationCIS 121 Data Structures and Algorithms with Java Spring Stacks and Queues Monday, February 12 / Tuesday, February 13
CIS Data Structures ad Algorithms with Java Sprig 08 Stacks ad Queues Moday, February / Tuesday, February Learig Goals Durig this lab, you will: Review stacks ad queues. Lear amortized ruig time aalysis
More informationPerformance 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 informationArithmetic Sequences
. Arithmetic Sequeces COMMON CORE Learig Stadards HSF-IF.A. HSF-BF.A.1a HSF-BF.A. HSF-LE.A. Essetial Questio How ca you use a arithmetic sequece to describe a patter? A arithmetic sequece is a ordered
More informationAccuracy Improvement in Camera Calibration
Accuracy Improvemet i Camera Calibratio FaJie L Qi Zag ad Reihard Klette CITR, Computer Sciece Departmet The Uiversity of Aucklad Tamaki Campus, Aucklad, New Zealad fli006, qza001@ec.aucklad.ac.z r.klette@aucklad.ac.z
More information15-859E: Advanced Algorithms CMU, Spring 2015 Lecture #2: Randomized MST and MST Verification January 14, 2015
15-859E: Advaced Algorithms CMU, Sprig 2015 Lecture #2: Radomized MST ad MST Verificatio Jauary 14, 2015 Lecturer: Aupam Gupta Scribe: Yu Zhao 1 Prelimiaries I this lecture we are talkig about two cotets:
More informationAnalysis 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 informationAdministrative UNSUPERVISED LEARNING. Unsupervised learning. Supervised learning 11/25/13. Final project. No office hours today
Admiistrative Fial project No office hours today UNSUPERVISED LEARNING David Kauchak CS 451 Fall 2013 Supervised learig Usupervised learig label label 1 label 3 model/ predictor label 4 label 5 Supervised
More informationChapter 1. Introduction to Computers and C++ Programming. Copyright 2015 Pearson Education, Ltd.. All rights reserved.
Chapter 1 Itroductio to Computers ad C++ Programmig Copyright 2015 Pearso Educatio, Ltd.. All rights reserved. Overview 1.1 Computer Systems 1.2 Programmig ad Problem Solvig 1.3 Itroductio to C++ 1.4 Testig
More informationImproving 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 informationOptimization 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 informationPruning and Summarizing the Discovered Time Series Association Rules from Mechanical Sensor Data Qing YANG1,a,*, Shao-Yu WANG1,b, Ting-Ting ZHANG2,c
Advaces i Egieerig Research (AER), volume 131 3rd Aual Iteratioal Coferece o Electroics, Electrical Egieerig ad Iformatio Sciece (EEEIS 2017) Pruig ad Summarizig the Discovered Time Series Associatio Rules
More informationMemetic 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 informationComputers and Scientific Thinking
Computers ad Scietific Thikig David Reed, Creighto Uiversity Chapter 15 JavaScript Strigs 1 Strigs as Objects so far, your iteractive Web pages have maipulated strigs i simple ways use text box to iput
More informationCreating Exact Bezier Representations of CST Shapes. David D. Marshall. California Polytechnic State University, San Luis Obispo, CA , USA
Creatig Exact Bezier Represetatios of CST Shapes David D. Marshall Califoria Polytechic State Uiversity, Sa Luis Obispo, CA 93407-035, USA The paper presets a method of expressig CST shapes pioeered by
More informationPython Programming: An Introduction to Computer Science
Pytho Programmig: A Itroductio to Computer Sciece Chapter 1 Computers ad Programs 1 Objectives To uderstad the respective roles of hardware ad software i a computig system. To lear what computer scietists
More informationNew 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 informationBayesian 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 informationA Hybrid Clustering Method Using Genetic Algorithm with New Variation Operators
Iteratioal Joural of Idustrial Egieerig & Productio Maagemet (2012) Jue 2012, Volume 23, Number 1 pp. 121-128 http:ijiepm.iust.ac.ir A Hybrid Clusterig Method Usig Geetic Algorithm with New Variatio Operators
More informationEvaluation 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 informationEFFECT OF QUERY FORMATION ON WEB SEARCH ENGINE RESULTS
Iteratioal Joural o Natural Laguage Computig (IJNLC) Vol. 2, No., February 203 EFFECT OF QUERY FORMATION ON WEB SEARCH ENGINE RESULTS Raj Kishor Bisht ad Ila Pat Bisht 2 Departmet of Computer Sciece &
More informationA Boolean Query Processing with a Result Cache in Mediator Systems
A Boolea Query Processig with a Result Cache i Mediator Systems Jae-heo Cheog ad Sag-goo Lee * Departmet of Computer Sciece Seoul Natioal Uiversity Sa 56-1 Shillim-dog Kwaak-gu, Seoul Korea {cjh, sglee}cygus.su.ac.kr
More informationShadow Document Methods of Results Merging
Shadow Documet Methods of Results Mergig Shegli Wu ad Fabio Crestai Departmet of Computer ad Iformatio Scieces Uiversity of Strathclyde, Glasgow, UK {s.wu,f.crestai}@cis.strath.ac.uk ABSTRACT I distributed
More informationBASED 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 informationA Note on Least-norm Solution of Global WireWarping
A Note o Least-orm Solutio of Global WireWarpig Charlie C. L. Wag Departmet of Mechaical ad Automatio Egieerig The Chiese Uiversity of Hog Kog Shati, N.T., Hog Kog E-mail: cwag@mae.cuhk.edu.hk Abstract
More informationCounting the Number of Minimum Roman Dominating Functions of a Graph
Coutig the Number of Miimum Roma Domiatig Fuctios of a Graph SHI ZHENG ad KOH KHEE MENG, Natioal Uiversity of Sigapore We provide two algorithms coutig the umber of miimum Roma domiatig fuctios of a graph
More informationInvestigating methods for improving Bagged k-nn classifiers
Ivestigatig methods for improvig Bagged k-nn classifiers Fuad M. Alkoot Telecommuicatio & Navigatio Istitute, P.A.A.E.T. P.O.Box 4575, Alsalmia, 22046 Kuwait Abstract- We experimet with baggig knn classifiers
More informationFeature 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 informationStone Images Retrieval Based on Color Histogram
Stoe Images Retrieval Based o Color Histogram Qiag Zhao, Jie Yag, Jigyi Yag, Hogxig Liu School of Iformatio Egieerig, Wuha Uiversity of Techology Wuha, Chia Abstract Stoe images color features are chose
More informationIMP: Superposer Integrated Morphometrics Package Superposition Tool
IMP: Superposer Itegrated Morphometrics Package Superpositio Tool Programmig by: David Lieber ( 03) Caisius College 200 Mai St. Buffalo, NY 4208 Cocept by: H. David Sheets, Dept. of Physics, Caisius College
More information. 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 informationA 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 informationChapter 9. Pointers and Dynamic Arrays. Copyright 2015 Pearson Education, Ltd.. All rights reserved.
Chapter 9 Poiters ad Dyamic Arrays Copyright 2015 Pearso Educatio, Ltd.. All rights reserved. Overview 9.1 Poiters 9.2 Dyamic Arrays Copyright 2015 Pearso Educatio, Ltd.. All rights reserved. Slide 9-3
More information9.1. Sequences and Series. Sequences. What you should learn. Why you should learn it. Definition of Sequence
_9.qxd // : AM Page Chapter 9 Sequeces, Series, ad Probability 9. Sequeces ad Series What you should lear Use sequece otatio to write the terms of sequeces. Use factorial otatio. Use summatio otatio to
More informationEvolutionary 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 informationAn 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 informationAnalysis 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 informationMorgan Kaufmann Publishers 26 February, COMPUTER ORGANIZATION AND DESIGN The Hardware/Software Interface. Chapter 5.
Morga Kaufma Publishers 26 February, 208 COMPUTER ORGANIZATION AND DESIGN The Hardware/Software Iterface 5 th Editio Chapter 5 Virtual Memory Review: The Memory Hierarchy Take advatage of the priciple
More informationDesigning 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 informationLecture 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 informationAssignment Problems with fuzzy costs using Ones Assignment Method
IOSR Joural of Mathematics (IOSR-JM) e-issn: 8-8, p-issn: 9-6. Volume, Issue Ver. V (Sep. - Oct.06), PP 8-89 www.iosrjourals.org Assigmet Problems with fuzzy costs usig Oes Assigmet Method S.Vimala, S.Krisha
More informationHomework 1 Solutions MA 522 Fall 2017
Homework 1 Solutios MA 5 Fall 017 1. Cosider the searchig problem: Iput A sequece of umbers A = [a 1,..., a ] ad a value v. Output A idex i such that v = A[i] or the special value NIL if v does ot appear
More informationA new algorithm to build feed forward neural networks.
A ew algorithm to build feed forward eural etworks. Amit Thombre Cetre of Excellece, Software Techologies ad Kowledge Maagemet, Tech Mahidra, Pue, Idia Abstract The paper presets a ew algorithm to build
More informationRecursive Procedures. How can you model the relationship between consecutive terms of a sequence?
6. Recursive Procedures I Sectio 6.1, you used fuctio otatio to write a explicit formula to determie the value of ay term i a Sometimes it is easier to calculate oe term i a sequece usig the previous terms.
More informationNeuro Fuzzy Model for Human Face Expression Recognition
IOSR Joural of Computer Egieerig (IOSRJCE) ISSN : 2278-0661 Volume 1, Issue 2 (May-Jue 2012), PP 01-06 Neuro Fuzzy Model for Huma Face Expressio Recogitio Mr. Mayur S. Burage 1, Prof. S. V. Dhopte 2 1
More informationOne advantage that SONAR has over any other music-sequencing product I ve worked
*gajedra* D:/Thomso_Learig_Projects/Garrigus_163132/z_productio/z_3B2_3D_files/Garrigus_163132_ch17.3d, 14/11/08/16:26:39, 16:26, page: 647 17 CAL 101 Oe advatage that SONAR has over ay other music-sequecig
More informationOur 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 informationProtected points in ordered trees
Applied Mathematics Letters 008 56 50 www.elsevier.com/locate/aml Protected poits i ordered trees Gi-Sag Cheo a, Louis W. Shapiro b, a Departmet of Mathematics, Sugkyukwa Uiversity, Suwo 440-746, Republic
More informationCMPT 125 Assignment 2 Solutions
CMPT 25 Assigmet 2 Solutios Questio (20 marks total) a) Let s cosider a iteger array of size 0. (0 marks, each part is 2 marks) it a[0]; I. How would you assig a poiter, called pa, to store the address
More informationEnsemble-based Feature Selection Criteria
Esemble-based Feature Selectio Criteria Terry Wideatt 1, Matthew Prior 1, Niv Effro 2, Natha Itrator 2 1 Cetre for Visio, Speech ad Sigal Proc (CVSSP), Uiversity of Surrey, Guildford, Surrey, Uited Kigdom
More informationImprovement 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 informationImage based Cats and Possums Identification for Intelligent Trapping Systems
Volume 159 No, February 017 Image based Cats ad Possums Idetificatio for Itelliget Trappig Systems T. A. S. Achala Perera School of Egieerig Aucklad Uiversity of Techology New Zealad Joh Collis School
More informationText Summarization using Neural Network Theory
Iteratioal Joural of Computer Systems (ISSN: 2394-065), Volume 03 Issue 07, July, 206 Available at http://www.ijcsolie.com/ Simra Kaur Jolly, Wg Cdr Ail Chopra 2 Departmet of CSE, Ligayas Uiversity, Faridabad
More informationDesigning a learning system
CS 75 Itro to Machie Learig Lecture Desigig a learig system Milos Hauskrecht milos@pitt.edu 539 Seott Square, -5 people.cs.pitt.edu/~milos/courses/cs75/ Admiistrivia No homework assigmet this week Please
More informationMapping Publishing and Mapping Adaptation in the Middleware of Railway Information Grid System
Mappig Publishig ad Mappig Adaptatio i the Middleware of Railway Iformatio Grid ystem You Gamei, Liao Huamig, u Yuzhog Istitute of Computig Techology, Chiese Academy of cieces, Beijig 00080 gameiu@ict.ac.c
More informationLecture 18. Optimization in n dimensions
Lecture 8 Optimizatio i dimesios Itroductio We ow cosider the problem of miimizig a sigle scalar fuctio of variables, f x, where x=[ x, x,, x ]T. The D case ca be visualized as fidig the lowest poit of
More informationData diverse software fault tolerance techniques
Data diverse software fault tolerace techiques Complemets desig diversity by compesatig for desig diversity s s limitatios Ivolves obtaiig a related set of poits i the program data space, executig the
More informationGraphs. Minimum Spanning Trees. Slides by Rose Hoberman (CMU)
Graphs Miimum Spaig Trees Slides by Rose Hoberma (CMU) Problem: Layig Telephoe Wire Cetral office 2 Wirig: Naïve Approach Cetral office Expesive! 3 Wirig: Better Approach Cetral office Miimize the total
More informationCS 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 informationLecture 17: Feature Subset Selection II
Lecture 17: Feature Subset Selectio II Expoetial search methods Brach ad Boud Approximate Mootoicity with Brach ad Boud Beam Search Radomized alorithms Radom Geeratio plus Sequetial Selectio Simulated
More informationLecture 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