Discovering similar user navigation behavior in Web log data
|
|
- Julianna Copeland
- 6 years ago
- Views:
Transcription
1 Dscoverng smlar user navgaton behavor n Web log data Tawfq A. Al-asad 1 and Ahmed J. Obad 2 1 Department of Computer Scence, Drector, College of IT, Babylon Unversty, Iraq. 2 Department of Computer Scence, Faculty of Educaton, KUFA Unversty, Iraq. Abstarct Wth the growth of World Wde Web and large number Hosts are jon contnuously to the nternet, huge number of access events to Web stes pages were recorded by Servers n log fles, many users share, send, post and download lot of thngs from Web Stes, ths manner can be dffcult to many organzaton and Agents n order to montor and control that, the recorded nformaton and type of analyss used to extract useful knowldege and understandng t become a practcal challenges to many researchers. Log fles can provded many events nformaton regard to Clents actvtes, server actvtes and so on. Many organzaton employee many log fles analyss tools to predct, analyss and montor users behavor towards ste contents.in ths paper we proposed algorthms to analyss hdden nformaton contents n Log fles and dscoverng patterns by dentfed users along wth them navgaton behavors then clusterng smlar users based on dfferent nterestng log fle content for many Web stes that hosted n Web server. Fnd statstcs for every part n log fle command lne whch are not present n many log fles analyss tools are supported here and fnally dscoverng frequent Web stes-users and user's actvtes towards those Web stes Keywords: Web Usage Mnng ; pattern mnng; Clusterng; Log fle analyss; Web log data. INTRODUCTION Informaton on nternet and especally on Web stes ncreasng rapdly day by day, Web Stes play an mportant role n ths manner where authentcated users, users are always uploads, downloads, browsed many contents accordng to them needs and nterest. Web Server provde a way to browse Web Stes by assgnng an IP address or DNS to dentfy t n addton hosted n t, Server record every events n the form of log fle. The process of dscoverng hdden nformaton from Web log fle s called Web Mnng. The am of t s to obtan nformaton about navgatonal behavor and retreve useful nformaton from very large raw data, can be represented by several mllons of event records n log fle. Web log data contans dfferent knds of nformaton and ncludng web document, web structure and user profles. Web mnng classfed nto three categores depend on whch part of Web to be mned [1, 2]. Categores are Web Structure Mnng, Web Content Mnng and Web Usage Mnng, Fgure 1 llustrate the categores of Web Mnng algorthms. Web Structure Mnng s the task for dscoverng knowledge from the structure of hyperlnks wthn Web pages and gven useful nformaton for the relatonshp among Web pages [3, 4]. The clusterng process can play mportant role here by groupng the Web pages based on ther structure, pages can represented by nodes and ther lnks as edges among these nodes, clusterng process can be done here based on graph representaton and understandng structure of Web pages and ts related to other pages n other Web Ste Pages. Lnk structure n Web pages can be classfed nto two types: Frst, hyperlnks that connect dfferent parts n the same page (Intra). Second, hyperlnks that connect two or more dfferent pages (Inter). The other role can be appled here by dentfy trustworthy pages and ther hub pages for a gven subject. Trustworthy pages contan mportant nformaton and supported by several lnks referred to t that means these pages are hghly referenced. Hub pages contan many lnks to trustworthy pages that can gve a role for clusterng Pages based on trustworthy pages. Web Structure Mnng can be employed to effcently mprove nformaton retreval and document classfcaton tasks [5]. Web Content Mnng s the task of dscoverng dfferent knds of nformaton contents and mprovng effcent mechansms to organze and groupng (clusterng) multmeda content to the search engnes for accessng these contents by usng keywords, categores, related contents etc. Multmeda contents on Web pages are vared such as structured content (.e. XML documents), Sem-structured (.e. HTML pages), Unstructured content (.e. plant text), other related contents Images, Audos, Vdeos whch are added to those pages or lnked to other hosted Stes. Recently there are some challenges appear regard to that n the case of many Web stes were desgned by usng not only HTML language, other Languages and systems were nvted here such as Content Management Systems ( CMS) etc. and the plat texts here are encrypted and stored n an SQL data bases and users events were recorded as vsted artcles and n ths case need to combne web mnng algorthms n case to mnng clusterng and extracted useful nformaton from user behavors and contents related. Web CMS s responsble for storng, control and management data and other component n long-term uses. CMS consst of repostory used to store and preserve varous component and use varous databases to store t. Repostory n CMS contan two categores, the frst one comprses source fles as well as CMS confguraton fles, these fles contan nformaton about type of content, metadata, users and group of users along wth them access data, profles and preferences. The second repostory contan databases where content and fles wll be processed through CMS and nhert 8797
2 databases and Tables that are constructed for recall and process the content [6]. Many researches has been done n Web Content Mnng, ncludng text mnng and ts ssues such as: topc dscovery, assocaton pattern dscovery, Web pages classfcaton and Web document clusterng. Other mportant body of work by dscoverng knowledge from mages n the feld of mage processng. Other research ncludng Latent Semantc Indexng (LSI) whch tes to analyzng structure of elements n document collecton, another mportant role lookng for fnd the poston of words n document for solvng the document categorzaton problems and extractng patterns or rules. Topc detecton and trackng also addressed as Web Content Mnng [7, 8]. Web Usage Mnng s the task to dscoverng the nterestng patterns from Web Usage Data. Interestng patterns nclude nformaton about user access patterns along wth varous types of request have been made by sngle or many users. The am of Web Usage Mnng s to understand the browsng and navgaton through Web pages to enhance many thngs such as: the qualty of commercal servces, allotment Web portals [9] or mprove Web structure and Web Server performance [10]. Web Usage Mnng can be defned as the extracton of useful user patterns from Web server access logs fles based on data mnng technques. Sources of log fles nclude Web server, Clent server, proxy server and applcaton servers [11, 12]. By found more than one source place that store the navgaton patterns and users accesses that make the mnng process more dffcult. The best and relable result can be obtan from the log fle that has all three types of log fle. Web page accesses that were cached n proxy servers or n clent sde does not contan records on server sde. Proxy server provde addtonal nformaton however the requested page are mssng n the clent sde, that lead to problem for collectng nformaton from clent sde. Most of Web mnng algorthms work based on Server sde log data, commonly used mnng algorthms are assocaton rule mnng, sequence mnng, clusterng [13]. Web Data Web Mnng Web Structure Mnng Web Content Mnng Web Usage Mnng Lnks Structure Web Search content Pre-processng Internal structure URL mnng Search page content Result page content Pattern Dscovery Pattern Analyss Fgure 1. Web Mnng Categores The organzaton of the paper s as follow: secton 2 llustrate the related work, secton 3 dscuss the log fle types, format and parameters, secton 4 show the Web mnng phases and preprocessng steps for our log fle, secton 5 contan the proposed model and algorthm, secton 6 our analyss result and fnally concluson for our works. RELATED WORK In the feld of Web usage mnng there are several data mnng technques have been used n order to dscover nterestng knowledge based on lookng and focused approach. The nformaton ganed by these technques can be used n many areas such as reconstructng Web stes, predcton next vsted pages, group smlar users, recommendaton systems and so on. Clusterng s the data mnng process that group together smlar tems havng smlar propertes. The clusterng may nclude group of smlar users, pages, references stes etc. Dscoverng group of smlar users n user communtes have been dscussed n [20]. Whle n [23, 24, 25, and 26] authors used Assocaton rule mnng for dscoverng Web pages 8798
3 accessed drectly by other pages. Web Usage Mnng s presented n many approaches along wth applyng data mnng technques for dscoverng nformaton. In [27] where Assocaton rule used for dscoverng relaton among pages, also used for detect the assocaton among group of users wth partcular nterest. Frequent path traversal and patterns topology of paths used wth WAP tree for representng and savng effcent patterns, others such as n [28, 29 and 30] they used Web usage mnng and Meta data for dscoverng terrorst and attacks Web stes. Your paper's fgs must be wthout background fll color, no border fg and no border legend, no vertcal lne, no horzontal Server Log fle types Web server log fles are plan text fles and ndependent from the server, generally there are four types of server logs based on types of nformaton recorded whch llustrate n Table 1 : Access log fle Agent log fle Error log fle Referrer log fle SERVER LOG FILE ANALYSIS Table 1. Format Types of Web server log fles Log fle types Actons Format Extracted knowledge Access log fle 1. Records all users request processed by server. 2. Record nformaton about [Wed Oct 11 14:32: ] [error] [Clent ] clent dened by server confguraton: Users' profles. Frequent patterns. Bandwdth usage. users. /export/home/lve/ap/htdocs/test. Agent log fle 1. User browsers. 2. Browsers verson. "Mozlla/4.0 (compatble; MSIE 4.01; Wndows NT)" Agent verson. Operatng system used. Error log fle Referrer log fle Lst of errors for users request made by server. 1.Informaton about lnk. 2.Redrects vstor to Ste. [Wed Oct 11 14:32: ] [error] [clent ] clent dened by server confguraton: /export/home/lve/ap/htdocs/test " "/page.html" Types of errors. Generated errors IP address. Date and tme of error occurred. Browser used. Keywords. Redrect lnk content. Server Log Fle Format There are three types of log fle format as follow: Common log fle format Is used by most of the web servers. The format of ths log fle s standardzed and can be analyzed by web analyss program, the sample format of ths type s shown below user-dentfer frank [10/Oct/2000:13:55: ] "GET /apache_pb.gf HTTP/1.0" Combned log fle format Is same as common log fle format but there are addtonal nformaton present here, these nformaton are "referral part, user-agent part and cooke prt", the sample format of ths type as bellow user-dentfer frank [10/Oct/2000:13:55: ] "GET /apache_pb.gf HTTP/1.0" Multple access logs Is consder the combnaton of the prevous two types (common log and combned log) fle format, n ths type of log fle format multple drectores can be can be created for access logs, the sample format of ths type as shown below. LogFormat "%h %l %u %t \"%r\" %>s %b" common CustomLog logs/access_log common CustomLog logs/referer_log "%{Referer} -> %U" CustomLog logs/agent_log "%{User-agent}" Server Log Fle Parameters Log fles contan varous parameters and can be very useful for recognzed users browsng attrbutes, many attrbutes can be added or enabled depend on server confguraton and user prvacy agreement, some of cookes and prvate nformaton can be used but n general there are common parameters an be found n log fles. Below wll llustrate n TABLE 2, the lst of some parameters useful for analyss processes. Table 2. Parameters of log fle T Parameter name Descrpton 1 User name (IP address) Identfy users Who vsted Webste by ts IP address. 2 Tme stamp Date and tme when user browsed and spend tme. 3 Request Exact request lne by user 8799
4 4 Status code Code sent by server after each user request 5 Bytes Content length of document transferred 6 User agent The browser that user used to send request 7 Request type Method used by user to send request GET, POST There are several works have been done on log fles each work deal wth partcular ssue of mnng and task, ths paper focus on dentfyng users and then extract knowledge about user behavors to groupng smlar users based on them browsng actvtes, our contrbuton here n case to analyss log fle we select KUFA unversty Apache HTTP server verson 1.1 man web server log fle, as we menton above ths log fle ts standardze text fle format and we are applyng text mnng technques to tokenze and extract nterestng nformaton from that log fle. PHASES OF WEB USAGE MINING In order to extract knowledge from log fle, several problems exst when extract useful nformaton from that log fle and also there are many outler records need to be elmnate from t n ths case we are applyng general phases of Web Usage Mnng to analyss and understand the extracted and vald nformaton. The general phases of Web Usage Mnng as follow: 4 4XX 5 5XX 303 SEE OTHER 304 NOT MODIFIED 400 BAD REQUEST 401 AUTHORIZATION REQUIRED 402 PAYMENT REQUIRED 404 NOT FOUND 500 INTERNAL SERVER ERROR 501 METHOD NOT IMPLEMENTED 502 BAD GATEWAY 503 SERVICE UNAVAILABLE 504 GATEWAY TIME OUT Status code show the success and falures users request, records wth status code less than 200 and greater than 299 are consdered falure records and elmnated from log fle entres. Data cleanng also nclude elmnated records that browsed rrelevant paths such as CSS content, man ste paths, gf, cons and maps etc. by checked suffx part of URL. FIGURE 2 represent porton of KUFA unversty Man Web server (Lnux server) log fle format, n that server DNS were assgned to Host IP address to dentfy Web ste that browsed by several users, We are consder to elmnate the records that browsed Man page due ts common n many records because ts contan lnks to all web stes n our server. Phase 1: Preprocessng Preprocessng phase nclude some actvtes can be appled on log fle for cleanng, dentfyng users, vald URL path and also elmnate outlers from log fle, tasks on preprocessng phase as follow [13]: Data Cleanng log fle contan several records are rrelevant to our work lke redrect path to other Stes, entres belong to top/bottom frames and records contan server error message. Error message dentfed through the status code that has been sent by server when user request partcular content, server status code can be vary and vald status codes are show n table 3. T Code Syntax 1 1XX 2 2XX 3 3XX Table 3. HTTP server status codes Status code Descrpton 100 COUNTINUE 101 SWITCHING PROTOCOL 102 PROCESSING 200 OK 201 CREATED 202 ACCEPTED 203 NON-AUTHORITATIVE INFORMATION 301 MOVED PERMANENTLY 302 FOUND Fgure 2. Porton of Web Server Log fle format The result of ths step produce the vald entres n log fle, next step used to dentfyng unque users and dstngush users that belong to same IP address. The followng algorthm n Fgure 3 used for elmnated rrelevant entres n log fle data. Data Cleanng Algorthm Input: Web Server Log fle data Output: Log fle data Step1: Read log fle record from (Web Server Log Fle). Step2: IF (log Fle Record).URL == (gf, Css, Man.php, ndex.php ) AND (Status code < 200 Status code > 209) 8800
5 Remove from log fle. End IF. Step3: Repeat Step 1 and Step 2 untl EOF (Web Server Log fle). Step4: Stop and Save fle n Data base. END Fgure 3. Data Cleanng Algorthm Steps User Identfcaton Web Usage Mnng does not requred knowledge for user's dentfyng, there s a need to dstngush among dfferent user's behavor. Server logs record of multple sessons for user may vst Web ste frequently. By absent authentcaton mechansms n many Web Server some Web ste used Cookes n Clent-sde, Due to prvacy content ths feature may dsable by users, therefore IP address alone not suffcent to dentfy unque users n general by assgnng many sessons to map IP address [15]. In case of absent user authentcaton and clent-sde cookes the possble accurate user dentfyng method by combnaton IP addresses wth User agent and referrer [13]. The followng fgure (FIGURE 4) show userdentfcaton algorthm steps that used for dentfyng dfferent users from log fle browsng data User Identfcaton Algorthm Input: log fle data Output: Unque Users Table. Step1: Intalzaton Create Table nclude the followng feld: ([User ID, IP's address, Date, Tme, Request, Ste name, User Agent, Sze)]. Step2: Read record from Log fle data Step3: User's IP addresses of tow sequental records are compared. Step4: IF ((IP address) s not n Users Table) THEN Assgn User ID to IP address Add both to Users Table ELSE IF ((IP address) s n User's Table) THEN Check (User Agent f same) then Add t wth Same User ID ELSE Assgn (next User ID) to IP address Add both to Users Table Step5: Repeat Step2-5 untl EOF (log fle data) Step6: STOP, Store Result. END. Fgure 4. User-Identfcaton Algorthm steps Phase 2: Mnng Phase Many technques can be appled here after preprocessng phase to extract knowledge such as assocaton rule mnng, frequent pattern mnng, Classfcaton, Clusterng etc. Dscoverng and analyss users patterns We are focus on clusterng technque n case to extract knowledge about smlar User's behavors based on browsng Web stes characterstcs. Ths technque help n many aspects for understandng smlar user's nterested content and Web stes contents, frequent User's- Ste browsng content, Effects of Ste content to Users and other ndcators related to ths work. Table 4 llustrate nformaton ganed after preprocessng steps, based on ths result we buld approprate Model for applyng clusterng algorthm to group of smlar User's navgaton behavors. Table 4. Identfyng Unque user's navgaton User Id IP address Date Tme Request Ste Name Agent User /Oct/ :01:12 GET AR "Mozlla/5.0 (Phone; CPU Phone OS 8_4) User /Oct/ :01:15 GET journals "Mozlla/5.0 User /Oct/ :01:18 GET AR "Mozlla/5.0 (Wndows NT 6.1) User /Oct/ :01:26 GET journals "Mozlla/5.0 (Wndows NT 6.1) User /Oct/ :01:20 GET conf "Mozlla/5.0 (Phone; CPU Phone OS 8_4) User /Oct/ :01:15 GET journals "Mozlla/5.0 User /Oct/ :01:26 GET AR "Mozlla/5.0 User /Oct/ :45:36 GET Lbr "Mozlla/5.0 User /Oct/ :45:39 GET AR "Mozlla/5.0 User /Oct/ :45:42 GET Lbr "Mozlla/5.0 (Wndows NT 6.1) Usage Data pre-processng result s a set of Μ Web stes vews, W = {W 1, W 2, W 3 W m}, and a set of (Ν) user transactons, T = {t 1, t 2, t 3 t n} where each (t ) s a subset of W. For data mnng tasks such as assocaton rule mnng and clusterng the orderng of Web ste vews s not relevant, we represent each user transactons as a vector over Μ dmensonal space of Web stes vews. In most Web Usage mnng algorthm and collaboratve flterng applcatons 8801
6 weghts were used to construct profles of smlar users. Weghts may be user ratng, spend tme on that page and ether bnary representng the presence or absence of that user from page vew, product vew and Ste vew. In our stuaton we are deal wth ths cases by elmnated the records that vst man page due t's consder the gate for other Web stes lnks and consder the User spent tme for each transacton have been made by a partcular user. Spend tme threshold used here to dstngush the users that browsed Stes for vewng Ste component from others who search for partcular content. Vald user's transactons treated to buld Users-Web stes vst matrx, the followng Table (Table 5) represent the occurrence of users based on vald transactons to construct Users-Web ste vst matrx. Table 5. User-Web ste occurrences vst matrx User Id AR Journals Conf Lbr Art Busn Comm Educ Gelog User User User User User User User User User User User User Due to space of real Table we are show only small part, above Table show for example user1, user2, user4, user7 and user 12 are more nterested n AR Web ste whle user5, user9 and user11 more nterested for thess and readng books from Lbrary Web ste. User Web ste Vst matrx produce many vst fractons for ths purpose we consder the occurrences of all users to be wthn smlar scale. X 0 1 f x T f x T Then new Table 6 after applyng Equaton (1) and (2) represent a user's Web ste vst matrx, clusterng can be appled for the enhanced matrx to fnd groups of smlar users based on browsng and navgaton patterns. Gven the mappng of user transactons nto mult-dmensonal space as enhanced vectors of Web stes vst as n Table 6, standard Herarchal clusterng algorthm can effcent employed here to take the smlarty of groups of users members wth respect to many Web stes vst patterns n the manner to form each possble number of groups n that have smlar behavors. Many clusterng algorthm have been appled here some algorthms consder clck stream to cluster dynamc users behavors by usng Mxture Models, ths process can be too complex to be modeled by usng basc probablty dstrbuton because each user may show dfferent behavors correspond to dfferent tasks, dfferent task reflect dfferent dstrbuton perodcally n such applcaton such as dynamc Web stes. Mxture Markov Models were appled n [17, 18] to cluster users based on smlartes n navgaton behavors. PROPOSED MODEL In order to dscover smlar user navgaton behavor, log data need to be preprocesses, elmnate non-relevant data then applyng data mng technques on result data. When applyng data mnng technques on web data ths s called Web data mng, Web mnng nlcude several tasks based on problem found and nterestng result. Fgure 4 show our proposed model for dscoverng hdden nformaton n Log fle data for varos users actvtes. (2) Cluster analyss and groupng smlar users Many data mnng technques can be appled n ths manner to deal wth fractons of User-Vst matrc, for some data mnng algorthm dfferent range values lead to a tendency and mmoderate nfluence for varables on the fnal result n order to scale the effect of t [16]. Normalzaton work well n ths manner for small values close to (0.0) and hgher ones to (1.0). MIN-MAX Normalzaton one of smplest and most used by scalng the dfference by the range. The MIN-MAX formula s gven as n Equaton (1). x ( x x mn ) ( x x ) max mn (1) Then by applyng threshold the new value updated as n Equaton (2) Fgure 5. Proposed Dagram 8802
7 After applyng equaton (1, 2) for the result Table 6, users wth small number of vstng count were elmnate n each Web ste, users wth hgh vst count pck hgher values and were grouped nto users-web ste nterest. Table 7 show Users-Vst ntersect n each Web ste User Id Table 6. User-Web stes Intersecton matrx AR Journals Conf Lbr Art Busn Comm Educ Gelog User User User User User User User User User User User User above Table 6 show users wth 0 values are (nonnterest/non-vsted) Web stes by correspondng users, whle Web stes wth 1 values refer to users are more nterestng to vst and browsed contentf from those Web stes, from the above result we can nfer for example user1 and user4 are more nterest to Web ste 1and 2, whle user2 and user3 are nterest to Web ste 6 and 7. Incase to fnd the smlartes among users, many bnary smlartes measures can be appled here, n [19] lst smlarty and dstance measures were appled n bnary data. The goal of ths measures s to fnd smlartes among data ponts n our scenaro we are consder ths data as a dynamc because the behavors of users may change through tme and based on them nterest, n ths case we are applyng 2 scenaros as follow : The frst one s by usng Cluster Identfcaton Algorthm (CIA) whch can be vsualze groupng smlar users n our Result Matrx and dentfed smlar users by calculatng the cells ntersecton rato among them, ths process yeld blocks of smlar users that share smlar Web-Stes browsed and elmnated not browsed content Table 6 consder to applyng CIA algorthm. Second scenaro s by usng dstance measure, we are consder each user s a partcular case and ts browsed Webstes dffer from others, n order to fnd smlartes among users so we are arranged Webstes based navgaton orders for example the result n Table 4 and Table 5 are combned together to form a vectors for users, we use character-based coded to represent Web stes names to be smple for comparng, Users wth small hts occurrences were not consdered, user behavors can dscovered through contnuously vsted stes by users, relatve frequency were calculated here for each user, as n Equaton (3), m F ( U ) M T Where users = 1 N, F(U ) relatve frequency for user, m s hts count of user () n partcular Web ste J, M s total number of hts for user (), fnally T selected threshold. Mnmum values are dscarded that does not satsfyng the selected threshold, user's vectors result as follow: (3) Table 7. User-Web Stes Navgaton behavors User AR Journals Conf Lbr Art Busn Comm Educ Gelog Id User User User User User User User User User User User User Then fnd smlartes among users, smlar users are grouped together to form new cluster followng Table 8 show smlarty matrx among users: Table 8. User-Web Stes Navgaton behavors User User User User User User User User User User Use Id r 10 User User User User User User User User User9 0 0 User1 0 0 Jaccard and Bray-Curts for smlarty/dssmlarty measures were appled, from Table 7 many herarchal algorthms can be appled for clusterng smlar users such as Sngle Lnkage and Complete Lnkage the result of Table 7 after applyng clusterng algorthm has ben shown n Table 8. To fnd smlartes among users Equaton (4) used and result compare wth dstance measures used n Equaton (5). 8803
8 a S ( U A, U B) a b c (4) X j X k D ( U A, U B) ( X X ) (5) ID Cluster-name Users j Table 9. Users - Smlarty values 1 Cluster1 user1,user4,user7,user10,user12 2 Cluster2 user2,user3,user6,user9,user10,user11 3 Cluster3/4 user5 /user8 CONCLUSION Ths paper focuse on dscoverng the hdden nformaton from man server general log fle, man server contan combnaton for all Web stes access nformaton that hosted on t n text format, ths fle nclude navgaton actvtes for many Web stes n order to understand the behavors of users towards those stes not for sngle Web ste, the contrbuton of the paper s to extract nformaton from huge log fle and consder novel approaches to deal and analyss users patterns, then extracted useful nformaton for vald sessons after that clusterng approach has been appled to groupng smlar users navgatons behavors, ths can gve as ndcators frequent users nterest towards dfferent Web stes content, montor users actvtes for partcular Web ste, consume bandwdth for each user durng selected perod, montor Web stes vsts and browsed content and many others actvtes for future works. ACKNOWLEDGEMENT Ths work and data analyss result has been supported by Informaton Technology Research and Development Center (ITRDC), Unversty of KUFA, IRAQ and School of Informaton Technology (SIT, Babylon Unversty, IRAQ). REFERENCES [1] Kosala and Blockeel: Web mnng research: A survey, SIGKDD : SIGKDD Exploratons: Newsletter of the Specal Interest Group (SIG) on Knowledge Dscovery and Data Mnng, ACM, Vol. 2, [2] S. K. Madra, S. S. Bhowmck,W. K. Ng, and E.-P. Lm :"Research ssues n web data mnng" n Data Warehousng and Knowledge Dscovery [3] J. Hou and Y. Zhang: Effectvely fndng relevant web pages from lnkage Informaton. IEEE Trans. Knowledge Data Eng., Vol. 15, No. 4, pp , k [4] H. Han and R. Elmasr: Learnng rules for conceptual structure on the Web, J. Intell. Inf. Syst., Vol. 22, No. 3, pp , [5] Renáta Iváncsy, István Vajk: Frequent Pattern Mnng n Web Log Data, Acta Polytechnca Hungarca Vol. 3, No. 1, [6] Danel MICAN, Ncolae TOMAI, Robert Ioan COROŞ: Web Content Management Systems, a Collaboratve Envronment n the Informaton Socety, Informatca Economcă vol.13, no 2/2009. [7] Shqun Yn, Yuhu Qu, Chengwen Zhong, Jfu Zhou: Study of Web Informaton Extracton and Classfcaton Method, Wreless Communcatons, Networkng and Moble Computng, WCom [8] M. A. Bayr, I. H. Toroslu, A. Cosar: A Performance Comparson of Pattern Dscovery Methods on Web Log Data, AICCSA-06, the 4th ACS/IEEE Internatonal Conference on Computer Systems and Applcatons. [9] M. Ernak and M. Vazrganns: Web mnng for web personalzaton, ACM Trans. Inter. Tech., Vol. 3, No. 1, pp. 1-27, [10] J. Pe, J. Han, B. Mortazav-Asl, and H. Zhu: Mnng access patterns effcently from web logs, Proceedngs of the 4th Pacfc-Asa Conference on Knowledge Dscovery and Data Mnng, Current Issues and New Applcatons. London, UK: Sprnger-Verlag, 2000, pp [11] Kohav, R: Mnng e-commerce data: The good, the bad, and the ugly, Proceedngs of the 7th ACM SIGKDD Internatonal Conference on Knowledge Dscovery and Data Mnng, San Francsco, Calforna, 8-13, [12] Anthony Scme: Web mnng: Applcaton and technques, IDEA, chapter 19, pp [13] R. Cooley, B. Mobasher, and J. Srvastava: Data preparaton for mnng world wde web browsng patterns, Knowledge and Informaton Systems, Vol. 1, No. 1, pp. 5-32, [14] L.K. Joshla Grace, V. Maheswar, and Dhnaharan Nagamala: Web Log Data Analyss and Mnng, Proc CCSIT-2011, Sprnger CCIS, Vol 133, pp , Jan [15] Bng Lu: Web data mnng, explorng Hyperlnks, Contents and Usage Data, Second edton, Sprnger, pp , [16] Danel T. Larose, Chantal D. Larose: Data Mnng and predctve analyss, Second edton, WILEY, PP , [17] Cadez, I., D. Heckerman, C. Meek, P. Smyth, S. Whte: Model-based clusterng and vsualzaton of navgaton patterns on a web ste, Data Mnng and Knowledge Dscovery,7(4): p , [18] Ypma, A., T. Heskes: Automatc categorzaton of web pages and user clusterng wth mxtures of hdden Markov models, In Proceedngs of Mnng Web Data for Dscoverng Usage Patterns and Profles,WEBKDD-2002,
9 [19] Seung-Seok Cho, Sung-Hyuk Cha, Charles C. Tappert: A Survey of Bnary Smlarty and Dstance Measures Systemcs, Cybernetcs And Informatcs, Volume 8 - Number 1, [20] Palouras, G., C. Papatheodorou, V. Karkaletss, C. Spyropoulos: Dscoverng user communtes on the Internet usng unsupervsed machne Learnng technques Interactng wth Computers, 14(6): p , [21] Chen H., Fan H., Chau M., Zeng D.: MetaSpder: meta-searchng and categorzaton on the web Journal of the Amercan Socety for Informaton Scence and Technology, 52 (13), , [22] Chung W.: Vsualzng E-Busness stakeholders on the web: a methodology and expermental results Internatonal Journal of Electronc Busness, [23] J. Punn, M. Krshnamoorthy, M. Zak: Web usage mnng: Languages and algorthms, n Studes n Classfcaton, Data Analyss, and Knowledge Organzaton. Sprnger-Verlag, [24] P. Batsta, M. aro, J. Slva: Mnng web access logs of an on-lne newspaper, [25] O. R. Zaane, M. Xn, J. Han: Dscoverng web access patterns andtrends by applyng olap and data mnng technology on web logs, n ADL 98: Proceedngs of the Advances n Dgtal Lbrares Conference.Washngton, DC, USA: IEEE Computer Socety, pp. 1-19, [26] J. F. F. M. V. M. L Shen, Lng Cheng, T. Stenberg: Mnng the most nterestng web access assocatons, n WebNet 2000-World Conferenceon the WWW and Internet, pp , [27] M. Ernak, M. Vazrganns: Web mnng for web personalzaton, ACM Trans. Inter. Tech., Vol. 3, No. 1, pp. 1-27, [28] X. Ln, C. Lu, Y. Zhang, X. Zhou: Effcently computng frequent tree-lke topology patterns n a web envronment, n TOOLS 99: Proceedngs of the 31st Internatonal Conference on Technology of Object- Orented Language and Systems. Washngton, DC, USA: IEEE Computer Socety, p. 440, [29] X. A. Nanopoulos, Y. Manolopoulos: Fndng generalzed path patterns for web log data mnng, n ADBIS-DASFAA 00: Proceedngs of the East- European Conference on Advances n Databases and Informaton Systems Held Jontly wth Internatonal Conference on Database Systems for Advanced Applcatons. London, UK: Sprnger-Verlag, pp ,
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 informationParallelism 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 informationCluster Analysis of Electrical Behavior
Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School
More informationTerm Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task
Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto
More informationSubspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;
Subspace clusterng Clusterng Fundamental to all clusterng technques s the choce of dstance measure between data ponts; D q ( ) ( ) 2 x x = x x, j k = 1 k jk Squared Eucldean dstance Assumpton: All features
More informationUser Authentication Based On Behavioral Mouse Dynamics Biometrics
User Authentcaton Based On Behavoral Mouse Dynamcs Bometrcs Chee-Hyung Yoon Danel Donghyun Km Department of Computer Scence Department of Computer Scence Stanford Unversty Stanford Unversty Stanford, CA
More informationA Webpage Similarity Measure for Web Sessions Clustering Using Sequence Alignment
A Webpage Smlarty Measure for Web Sessons Clusterng Usng Sequence Algnment Mozhgan Azmpour-Kv School of Engneerng and Scence Sharf Unversty of Technology, Internatonal Campus Ksh Island, Iran mogan_az@ksh.sharf.edu
More informationLinkSelector: A Web Mining Approach to. Hyperlink Selection for Web Portals
nkselector: A Web Mnng Approach to Hyperlnk Selecton for Web Portals Xao Fang and Olva R. u Sheng Department of Management Informaton Systems Unversty of Arzona, AZ 8572 {xfang,sheng}@bpa.arzona.edu Submtted
More informationA Fast Content-Based Multimedia Retrieval Technique Using Compressed Data
A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,
More informationPreprocessing of Web Usage Data for Application in Prefetching to Reduce Web Latency
Internatonal Journal of Electrcal& Computer Scences IJECS-IJENS Vol:14 No:04 1 Preprocessng of Web Usage Data for Applcaton n Prefetchng to Reduce Web Latency G T Raju Professor, Department of CSE, RNS
More informationEffective Page Recommendation Algorithms Based on. Distributed Learning Automata and Weighted Association. Rules
Effectve Page Recommendaton Algorthms Based on Dstrbuted Learnng Automata and Weghted Assocaton Rules R. Forsat 1*, M. R. Meybod 2 1 Department of Computer Engneerng, Islamc Azad Unversty, Karaj Branch,
More informationFEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur
FEATURE EXTRACTION Dr. K.Vjayarekha Assocate Dean School of Electrcal and Electroncs Engneerng SASTRA Unversty, Thanjavur613 41 Jont Intatve of IITs and IISc Funded by MHRD Page 1 of 8 Table of Contents
More informationFINDING IMPORTANT NODES IN SOCIAL NETWORKS BASED ON MODIFIED PAGERANK
FINDING IMPORTANT NODES IN SOCIAL NETWORKS BASED ON MODIFIED PAGERANK L-qng Qu, Yong-quan Lang 2, Jng-Chen 3, 2 College of Informaton Scence and Technology, Shandong Unversty of Scence and Technology,
More informationQuery 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 informationA 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 informationModule Management Tool in Software Development Organizations
Journal of Computer Scence (5): 8-, 7 ISSN 59-66 7 Scence Publcatons Management Tool n Software Development Organzatons Ahmad A. Al-Rababah and Mohammad A. Al-Rababah Faculty of IT, Al-Ahlyyah Amman Unversty,
More informationLoad Balancing for Hex-Cell Interconnection Network
Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,
More informationA Binarization Algorithm specialized on Document Images and Photos
A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a
More informationLecture 5: Multilayer Perceptrons
Lecture 5: Multlayer Perceptrons Roger Grosse 1 Introducton So far, we ve only talked about lnear models: lnear regresson and lnear bnary classfers. We noted that there are functons that can t be represented
More informationCS 534: Computer Vision Model Fitting
CS 534: Computer Vson Model Fttng Sprng 004 Ahmed Elgammal Dept of Computer Scence CS 534 Model Fttng - 1 Outlnes Model fttng s mportant Least-squares fttng Maxmum lkelhood estmaton MAP estmaton Robust
More informationSLAM Summer School 2006 Practical 2: SLAM using Monocular Vision
SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,
More informationData Preprocessing Based on Partially Supervised Learning Na Liu1,2, a, Guanglai Gao1,b, Guiping Liu2,c
6th Internatonal Conference on Informaton Engneerng for Mechancs and Materals (ICIMM 2016) Data Preprocessng Based on Partally Supervsed Learnng Na Lu1,2, a, Guangla Gao1,b, Gupng Lu2,c 1 College of Computer
More informationEnhanced Watermarking Technique for Color Images using Visual Cryptography
Informaton Assurance and Securty Letters 1 (2010) 024-028 Enhanced Watermarkng Technque for Color Images usng Vsual Cryptography Enas F. Al rawashdeh 1, Rawan I.Zaghloul 2 1 Balqa Appled Unversty, MIS
More informationSteps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices
Steps for Computng the Dssmlarty, Entropy, Herfndahl-Hrschman and Accessblty (Gravty wth Competton) Indces I. Dssmlarty Index Measurement: The followng formula can be used to measure the evenness between
More informationProfessional 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 informationA Clustering Algorithm for Key Frame Extraction Based on Density Peak
Journal of Computer and Communcatons, 2018, 6, 118-128 http://www.scrp.org/ournal/cc ISSN Onlne: 2327-5227 ISSN Prnt: 2327-5219 A Clusterng Algorthm for Key Frame Extracton Based on Densty Peak Hong Zhao
More informationUSING GRAPHING SKILLS
Name: BOLOGY: Date: _ Class: USNG GRAPHNG SKLLS NTRODUCTON: Recorded data can be plotted on a graph. A graph s a pctoral representaton of nformaton recorded n a data table. t s used to show a relatonshp
More informationOutline. Type of Machine Learning. Examples of Application. Unsupervised Learning
Outlne Artfcal Intellgence and ts applcatons Lecture 8 Unsupervsed Learnng Professor Danel Yeung danyeung@eee.org Dr. Patrck Chan patrckchan@eee.org South Chna Unversty of Technology, Chna Introducton
More informationMachine Learning: Algorithms and Applications
14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of
More informationDetection of an Object by using Principal Component Analysis
Detecton of an Object by usng Prncpal Component Analyss 1. G. Nagaven, 2. Dr. T. Sreenvasulu Reddy 1. M.Tech, Department of EEE, SVUCE, Trupath, Inda. 2. Assoc. Professor, Department of ECE, SVUCE, Trupath,
More informationResearch on Categorization of Animation Effect Based on Data Mining
MATEC Web of Conferences 22, 0102 0 ( 2015) DOI: 10.1051/ matecconf/ 2015220102 0 C Owned by the authors, publshed by EDP Scences, 2015 Research on Categorzaton of Anmaton Effect Based on Data Mnng Na
More informationTN348: Openlab Module - Colocalization
TN348: Openlab Module - Colocalzaton Topc The Colocalzaton module provdes the faclty to vsualze and quantfy colocalzaton between pars of mages. The Colocalzaton wndow contans a prevew of the two mages
More informationUtilizing Content to Enhance a Usage-Based Method for Web Recommendation based on Q-Learning
Proceedngs of the Twenty-Frst Internatonal FLAIS Conference (2008) Utlzng Content to Enhance a Usage-Based Method for Web ecommendaton based on Q-Learnng Nma Taghpour Department of Computer Engneerng Amrkabr
More informationOn Some Entertaining Applications of the Concept of Set in Computer Science Course
On Some Entertanng Applcatons of the Concept of Set n Computer Scence Course Krasmr Yordzhev *, Hrstna Kostadnova ** * Assocate Professor Krasmr Yordzhev, Ph.D., Faculty of Mathematcs and Natural Scences,
More informationA PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION
1 THE PUBLISHING HOUSE PROCEEDINGS OF THE ROMANIAN ACADEMY, Seres A, OF THE ROMANIAN ACADEMY Volume 4, Number 2/2003, pp.000-000 A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION Tudor BARBU Insttute
More informationRelated-Mode Attacks on CTR Encryption Mode
Internatonal Journal of Network Securty, Vol.4, No.3, PP.282 287, May 2007 282 Related-Mode Attacks on CTR Encrypton Mode Dayn Wang, Dongda Ln, and Wenlng Wu (Correspondng author: Dayn Wang) Key Laboratory
More informationFuzzy C-Means Initialized by Fixed Threshold Clustering for Improving Image Retrieval
Fuzzy -Means Intalzed by Fxed Threshold lusterng for Improvng Image Retreval NAWARA HANSIRI, SIRIPORN SUPRATID,HOM KIMPAN 3 Faculty of Informaton Technology Rangst Unversty Muang-Ake, Paholyotn Road, Patumtan,
More informationMULTISPECTRAL 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 informationSequential search. Building Java Programs Chapter 13. Sequential search. Sequential search
Sequental search Buldng Java Programs Chapter 13 Searchng and Sortng sequental search: Locates a target value n an array/lst by examnng each element from start to fnsh. How many elements wll t need to
More informationAvailable 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 informationUnsupervised Learning and Clustering
Unsupervsed Learnng and Clusterng Supervsed vs. Unsupervsed Learnng Up to now we consdered supervsed learnng scenaro, where we are gven 1. samples 1,, n 2. class labels for all samples 1,, n Ths s also
More informationUnsupervised Learning
Pattern Recognton Lecture 8 Outlne Introducton Unsupervsed Learnng Parametrc VS Non-Parametrc Approach Mxture of Denstes Maxmum-Lkelhood Estmates Clusterng Prof. Danel Yeung School of Computer Scence and
More informationA Resources Virtualization Approach Supporting Uniform Access to Heterogeneous Grid Resources 1
A Resources Vrtualzaton Approach Supportng Unform Access to Heterogeneous Grd Resources 1 Cunhao Fang 1, Yaoxue Zhang 2, Song Cao 3 1 Tsnghua Natonal Labatory of Inforamaton Scence and Technology 2 Department
More informationPrivate Information Retrieval (PIR)
2 Levente Buttyán Problem formulaton Alce wants to obtan nformaton from a database, but she does not want the database to learn whch nformaton she wanted e.g., Alce s an nvestor queryng a stock-market
More informationClassifier Selection Based on Data Complexity Measures *
Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.
More informationThe Research of Support Vector Machine in Agricultural Data Classification
The Research of Support Vector Machne n Agrcultural Data Classfcaton Le Sh, Qguo Duan, Xnmng Ma, Me Weng College of Informaton and Management Scence, HeNan Agrcultural Unversty, Zhengzhou 45000 Chna Zhengzhou
More informationImprovement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration
Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,
More informationON SOME ENTERTAINING APPLICATIONS OF THE CONCEPT OF SET IN COMPUTER SCIENCE COURSE
Yordzhev K., Kostadnova H. Інформаційні технології в освіті ON SOME ENTERTAINING APPLICATIONS OF THE CONCEPT OF SET IN COMPUTER SCIENCE COURSE Yordzhev K., Kostadnova H. Some aspects of programmng educaton
More informationUser Tweets based Genre Prediction and Movie Recommendation using LSI and SVD
User Tweets based Genre Predcton and Move Recommendaton usng LSI and SVD Saksh Bansal, Chetna Gupta Department of CSE/IT Jaypee Insttute of Informaton Technology,sec-62 Noda, Inda sakshbansal76@gmal.com,
More informationEvaluation of an Enhanced Scheme for High-level Nested Network Mobility
IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.15 No.10, October 2015 1 Evaluaton of an Enhanced Scheme for Hgh-level Nested Network Moblty Mohammed Babker Al Mohammed, Asha Hassan.
More informationAn Optimal Algorithm for Prufer Codes *
J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,
More informationA Knowledge Management System for Organizing MEDLINE Database
A Knowledge Management System for Organzng MEDLINE Database Hyunk Km, Su-Shng Chen Computer and Informaton Scence Engneerng Department, Unversty of Florda, Ganesvlle, Florda 32611, USA Wth the exploson
More informationThe Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique
//00 :0 AM Outlne and Readng The Greedy Method The Greedy Method Technque (secton.) Fractonal Knapsack Problem (secton..) Task Schedulng (secton..) Mnmum Spannng Trees (secton.) Change Money Problem Greedy
More informationImpact of a New Attribute Extraction Algorithm on Web Page Classification
Impact of a New Attrbute Extracton Algorthm on Web Page Classfcaton Gösel Brc, Banu Dr, Yldz Techncal Unversty, Computer Engneerng Department Abstract Ths paper ntroduces a new algorthm for dmensonalty
More informationSimulation Based Analysis of FAST TCP using OMNET++
Smulaton Based Analyss of FAST TCP usng OMNET++ Umar ul Hassan 04030038@lums.edu.pk Md Term Report CS678 Topcs n Internet Research Sprng, 2006 Introducton Internet traffc s doublng roughly every 3 months
More informationKeywords - 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 informationFast 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 informationClassic Term Weighting Technique for Mining Web Content Outliers
Internatonal Conference on Computatonal Technques and Artfcal Intellgence (ICCTAI'2012) Penang, Malaysa Classc Term Weghtng Technque for Mnng Web Content Outlers W.R. Wan Zulkfel, N. Mustapha, and A. Mustapha
More informationUB at GeoCLEF Department of Geography Abstract
UB at GeoCLEF 2006 Mguel E. Ruz (1), Stuart Shapro (2), June Abbas (1), Slva B. Southwck (1) and Davd Mark (3) State Unversty of New York at Buffalo (1) Department of Lbrary and Informaton Studes (2) Department
More informationRanking Search Results by Web Quality Dimensions
Rankng Search Results by Web Qualty Dmensons Joshua C. C. Pun Department of Computer Scence HKUST Clear Water Bay, Kowloon Hong Kong punjcc@cs.ust.hk Frederck H. Lochovsky Department of Computer Scence
More informationLoad-Balanced Anycast Routing
Load-Balanced Anycast Routng Chng-Yu Ln, Jung-Hua Lo, and Sy-Yen Kuo Department of Electrcal Engneerng atonal Tawan Unversty, Tape, Tawan sykuo@cc.ee.ntu.edu.tw Abstract For fault-tolerance and load-balance
More informationTsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance
Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for
More informationConcurrent Apriori Data Mining Algorithms
Concurrent Apror Data Mnng Algorthms Vassl Halatchev Department of Electrcal Engneerng and Computer Scence York Unversty, Toronto October 8, 2015 Outlne Why t s mportant Introducton to Assocaton Rule Mnng
More informationDetermining the Optimal Bandwidth Based on Multi-criterion Fusion
Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn
More informationAudio Content Classification Method Research Based on Two-step Strategy
(IJACSA) Internatonal Journal of Advanced Computer Scence and Applcatons, Audo Content Classfcaton Method Research Based on Two-step Strategy Sume Lang Department of Computer Scence and Technology Chongqng
More informationA Web Site Classification Approach Based On Its Topological Structure
Internatonal Journal on Asan Language Processng 20 (2):75-86 75 A Web Ste Classfcaton Approach Based On Its Topologcal Structure J-bn Zhang,Zh-mng Xu,Kun-l Xu,Q-shu Pan School of Computer scence and Technology,Harbn
More informationSolving two-person zero-sum game by Matlab
Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by
More informationTECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS. Muradaliyev A.Z.
TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS Muradalyev AZ Azerbajan Scentfc-Research and Desgn-Prospectng Insttute of Energetc AZ1012, Ave HZardab-94 E-mal:aydn_murad@yahoocom Importance of
More informationEfficient Distributed File System (EDFS)
Effcent Dstrbuted Fle System (EDFS) (Sem-Centralzed) Debessay(Debsh) Fesehaye, Rahul Malk & Klara Naherstedt Unversty of Illnos-Urbana Champagn Contents Problem Statement, Related Work, EDFS Desgn Rate
More informationA New Approach For the Ranking of Fuzzy Sets With Different Heights
New pproach For the ankng of Fuzzy Sets Wth Dfferent Heghts Pushpnder Sngh School of Mathematcs Computer pplcatons Thapar Unversty, Patala-7 00 Inda pushpndersnl@gmalcom STCT ankng of fuzzy sets plays
More informationWishing you all a Total Quality New Year!
Total Qualty Management and Sx Sgma Post Graduate Program 214-15 Sesson 4 Vnay Kumar Kalakband Assstant Professor Operatons & Systems Area 1 Wshng you all a Total Qualty New Year! Hope you acheve Sx sgma
More informationMachine 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 informationA Deflected Grid-based Algorithm for Clustering Analysis
A Deflected Grd-based Algorthm for Clusterng Analyss NANCY P. LIN, CHUNG-I CHANG, HAO-EN CHUEH, HUNG-JEN CHEN, WEI-HUA HAO Department of Computer Scence and Informaton Engneerng Tamkang Unversty 5 Yng-chuan
More informationIP Camera Configuration Software Instruction Manual
IP Camera 9483 - Confguraton Software Instructon Manual VBD 612-4 (10.14) Dear Customer, Wth your purchase of ths IP Camera, you have chosen a qualty product manufactured by RADEMACHER. Thank you for the
More informationFeature 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 informationArabic Text Classification Using N-Gram Frequency Statistics A Comparative Study
Arabc Text Classfcaton Usng N-Gram Frequency Statstcs A Comparatve Study Lala Khresat Dept. of Computer Scence, Math and Physcs Farlegh Dcknson Unversty 285 Madson Ave, Madson NJ 07940 Khresat@fdu.edu
More informationThe 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 informationThe Effect of Similarity Measures on The Quality of Query Clusters
The effect of smlarty measures on the qualty of query clusters. Fu. L., Goh, D.H., Foo, S., & Na, J.C. (2004). Journal of Informaton Scence, 30(5) 396-407 The Effect of Smlarty Measures on The Qualty of
More informationOutline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like:
Self-Organzng Maps (SOM) Turgay İBRİKÇİ, PhD. Outlne Introducton Structures of SOM SOM Archtecture Neghborhoods SOM Algorthm Examples Summary 1 2 Unsupervsed Hebban Learnng US Hebban Learnng, Cntd 3 A
More informationMining Web Logs with PLSA Based Prediction Model to Improve Web Caching Performance
JOURAL OF COMPUTERS, VOL. 8, O. 5, MAY 2013 1351 Mnng Web Logs wth PLSA Based Predcton Model to Improve Web Cachng Performance Chub Huang Department of Automaton, USTC Key laboratory of network communcaton
More informationLecture #15 Lecture Notes
Lecture #15 Lecture Notes The ocean water column s very much a 3-D spatal entt and we need to represent that structure n an economcal way to deal wth t n calculatons. We wll dscuss one way to do so, emprcal
More informationAn 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 informationCompiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz
Compler Desgn Sprng 2014 Regster Allocaton Sample Exercses and Solutons Prof. Pedro C. Dnz USC / Informaton Scences Insttute 4676 Admralty Way, Sute 1001 Marna del Rey, Calforna 90292 pedro@s.edu Regster
More informationResearch and Application of Fingerprint Recognition Based on MATLAB
Send Orders for Reprnts to reprnts@benthamscence.ae The Open Automaton and Control Systems Journal, 205, 7, 07-07 Open Access Research and Applcaton of Fngerprnt Recognton Based on MATLAB Nng Lu* Department
More informationCourse Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms
Course Introducton Course Topcs Exams, abs, Proects A quc loo at a few algorthms 1 Advanced Data Structures and Algorthms Descrpton: We are gong to dscuss algorthm complexty analyss, algorthm desgn technques
More informationMachine Learning. Topic 6: Clustering
Machne Learnng Topc 6: lusterng lusterng Groupng data nto (hopefully useful) sets. Thngs on the left Thngs on the rght Applcatons of lusterng Hypothess Generaton lusters mght suggest natural groups. Hypothess
More informationEdge Detection in Noisy Images Using the Support Vector Machines
Edge Detecton n Nosy Images Usng the Support Vector Machnes Hlaro Gómez-Moreno, Saturnno Maldonado-Bascón, Francsco López-Ferreras Sgnal Theory and Communcatons Department. Unversty of Alcalá Crta. Madrd-Barcelona
More informationStudy of Data Stream Clustering Based on Bio-inspired Model
, pp.412-418 http://dx.do.org/10.14257/astl.2014.53.86 Study of Data Stream lusterng Based on Bo-nspred Model Yngme L, Mn L, Jngbo Shao, Gaoyang Wang ollege of omputer Scence and Informaton Engneerng,
More informationReal-time Motion Capture System Using One Video Camera Based on Color and Edge Distribution
Real-tme Moton Capture System Usng One Vdeo Camera Based on Color and Edge Dstrbuton YOSHIAKI AKAZAWA, YOSHIHIRO OKADA, AND KOICHI NIIJIMA Graduate School of Informaton Scence and Electrcal Engneerng,
More informationAccounting for the Use of Different Length Scale Factors in x, y and z Directions
1 Accountng for the Use of Dfferent Length Scale Factors n x, y and z Drectons Taha Soch (taha.soch@kcl.ac.uk) Imagng Scences & Bomedcal Engneerng, Kng s College London, The Rayne Insttute, St Thomas Hosptal,
More informationPRÉSENTATIONS DE PROJETS
PRÉSENTATIONS DE PROJETS Rex Onlne (V. Atanasu) What s Rex? Rex s an onlne browser for collectons of wrtten documents [1]. Asde ths core functon t has however many other applcatons that make t nterestng
More informationEnhancement 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 informationCS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 15
CS434a/541a: Pattern Recognton Prof. Olga Veksler Lecture 15 Today New Topc: Unsupervsed Learnng Supervsed vs. unsupervsed learnng Unsupervsed learnng Net Tme: parametrc unsupervsed learnng Today: nonparametrc
More informationVisual Thesaurus for Color Image Retrieval using Self-Organizing Maps
Vsual Thesaurus for Color Image Retreval usng Self-Organzng Maps Chrstopher C. Yang and Mlo K. Yp Department of System Engneerng and Engneerng Management The Chnese Unversty of Hong Kong, Hong Kong ABSTRACT
More information6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour
6.854 Advanced Algorthms Petar Maymounkov Problem Set 11 (November 23, 2005) Wth: Benjamn Rossman, Oren Wemann, and Pouya Kheradpour Problem 1. We reduce vertex cover to MAX-SAT wth weghts, such that the
More informationA 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 informationProblem Definitions and Evaluation Criteria for Computational Expensive Optimization
Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty
More informationQuantifying Performance Models
Quantfyng Performance Models Prof. Danel A. Menascé Department of Computer Scence George Mason Unversty www.cs.gmu.edu/faculty/menasce.html 1 Copyrght Notce Most of the fgures n ths set of sldes come from
More informationRelevance Feedback for Image Retrieval
Vashal D Dhale et al, / (IJCSIT Internatonal Journal of Computer Scence and Informaton Technologes, Vol 4 (2, 203, 39-323 Relevance Feedback for Image Retreval Vashal D Dhale, Dr A R Mahaan, Prof Uma Thakur
More informationUnsupervised Learning and Clustering
Unsupervsed Learnng and Clusterng Why consder unlabeled samples?. Collectng and labelng large set of samples s costly Gettng recorded speech s free, labelng s tme consumng 2. Classfer could be desgned
More information