A User Selection Method in Advertising System
|
|
- Nicholas Taylor
- 5 years ago
- Views:
Transcription
1 Int. J. Communcatons, etwork and System Scences, 2010, 3, do: /jcns Publshed Onlne January 2010 ( A User Selecton Method n Advertsng System Shy XIOG, Zhqng LI, Bo XIAO Pattern Recognton & Intellgent System Lab,Bejng Unversty of Posts and Telecommuncatons, Bejng, Chna Emal: xongshy@gmal.com, {lnzq, xaobo}@bupt.edu.cn Receved October 27, 2009; revsed ovember 28, 2009; accepted December 30, 2009 Abstract It s mportant for moble operators to recommend new servces. Tradtonal method s sendng advertsng messages to all moble users. But most of users who are not nterested n these servces treat the messages as Spam. Ths paper presents a method to fnd potental customers who are lkely to accept the servces. Ths method searchs the maxmum frequent temsets whch ndcate potental customers features from a large data set of users nformaton, then fnd potental customers from those maxmum frequent temsets by usng a bayesan network classfer. Expermental results demonstrate ths method can select users wth hgher accuracy. Keywords: User Selecton, Maxmum Frequent Itemsets, Bayesan etwork 1. Introducton Recent years, as the ncreasng number of moble operators and new servces, the competton between operators becomes more and more furous. The key of the competton s to wn more customers. How to ntroduce new servces to proper customers becomes one of the major problems. ow, advertsng system mostly uses these technologes: collaboratve flterng technology [1], content based recommendaton, knowledge based recommendaton, effcency based recommendaton, assocaton rules based recommendaton [2], etc. These methods are usually used to select and recommend products to specfed users. They analyse the user s hstorcal data and predct what products the user mostly need. For example, an onlne move system uses one of above technologes to serve ts user whose name s Mke. The system wll analyse Mke s hstory and recommend hm some moves n whch he may be nterested. So we can say these methods solve the problem to select products for users. The proplem we desre to solve n ths paper s a lttle dfferent. Here we have only one product, namely the new moble servce, and we want to know whch users want to buy ths servce. Those users are potental customers, and we need to pck out them from a large number of users. The problem s to select users for one product. As a result, those tradtonal methods wll not work well on ths problem. Ths paper presents a new method to select users. Ths method searchs the maxmum frequent temsets from a large data set, then fnd potental customers from those maxmum frequent temsets by usng a bayesan network classfer. 2. Analyss of Customer Features Usually, moble operators keep a large database of consumpton data. Ths data ncludes the tems that represent customers features, such as the telephone fees of every month, functon fees and nformaton fees, etc. Ths data also shows whether a customer accepts the new servce, those customers who accept new servce can be seen as postve samples, and others can be seen as negatve samples. As a result, the problem to select postve samples can be treated as a two-category classfcaton problem. Usually, two-category classfcaton problem should have data wth features shown n Fgure 1(a). However, the moble users data are qute dfferent (shown n Fgure 1(b). For example, some users who have the same features may belong to dfferent categores. What s more, number of negatve samples s much larger than the number of postve sample. Because of ths, routne classfcaton methods such as nave bayesan classfer s hard to classfy such data. 3. Frequently Itemsets 3.1. Fomal Model of Frequently Itemsets Let sample X has n dmensons(attrbutes) and the value may be contnuous. Dvde the value nto several seg- Copyrght 2010 ScRes.
2 S. Y. XIOG ET AL. 55 x2 O O x2 (a) (b) x1 x1 Fgure 1. Feature dstrbuton map. (a) Two-category classfcaton; (b) Moble users data. ments and gve each segment a unque seral code. Because each sample belongs to an segment, we can use the correspondng segment code to replace ths sample. In ths way, orgnal samples X are mapped to the segment codes Y : F : X Y. Dvde the orgnal samples uses the followng gudelnes: 1) The number of postve samples n each segment s unformly dstrbuted. 2) Each Attrbutes should have the same number of segments. Make each Attrbute dvded nto k segments so that the segments code should not bgger than k. If the value of a Attrubutes s a 0-1 type value, the number of segments code should be {0,1}, not {0,1,..., k -1}. The Attrbute segment s called tem. The Attrbute segment combnaton of mm ( n) attrbutes {,,..., } s called temsets. m s length of the temsets I. If a sample s attrbute segments n m dmensons are,,...,, we call ths sample satsfed the temsets. I If an temsets has postve sample number a whle has negatve sample number b, f a/( a b) s the success rate of temsets I. Gven a mnmum support threshold. If a >, we call I frequent temsets. j 3.2. Frequent Close Itemset Gven I {,,..., } be an temsets wth length m, add an tem of attrbute m 1 nto I, so I ' {,,...,, } m 1 has the length m 1. If I and I ' have the same number of postve samples, then f f. Ths can be proofed as below: Suppose both I and I ' have a postve samples. I has b negatve samples, whle I ' has b '. Snce I ' has one a a more tem m 1,we can get b b '. Thus a b, a b' th at s f f, and I ' has a hgher success rato. In case of f f, gnore the temsets I, and keep I '. When f f, I s called frequent close temsets(ab. FCI). There are some typcal algorthm to mne frequent close temsets such as Apror [3] and FP2Tree [4]. As a part of the user selecton method, samples n the FCI should be predcted usng Bayesan network classfer. Gven the number of attrbutes n and the number of segments k, then the number of temsets s: 1 C M ( M 1) 1 The number has an exponental growth, so t s mpossble to construct a Bayesan network classfer for all temsets. A compromsed way s colletng all of the samples to be one data set and construct a Bayesan network classfer based on ths data set. 4. Bayesan etwork Model for User Selecton 4.1. Bayesan etwork Bascs A Bayesan network s a graphcal model that encodes relatonshps among varables of nterest. A Bayesan network conssts of a set of nodes and a set of drected edges between nodes [5], whch shown n Fgure 2. In general, a Bayesan network s expressed as sgn B( GP, ), whch conssts of the followng two parts [6]. Fgure 2. Typcal graphc model of Bayesan networks. Copyrght 2010 ScRes.
3 56 S. Y. XIOG ET AL. 1) A drected acyclc graph G wth n nodes. The nodes of the graph represent random varables or events. The drected edges between nodes n the graph represent the causal relatonshps of the nodes. The mportant concept n Bayesan networks s the condtonal ndependence between varables. For any varable V, the parent varables of V s pa( V ), the s ndependent of the V varables set A( V ), whch s the set of the varables that are not chld varables of pa( V ), so the probablty of s calculated as: V log MDL( B D) B LL( B D) 2 where B s the number of parameters n the network. The second term s the negaton of the log lkelhood of B gven D : LL( B D) log( P ) B( u ) 1 Bayesan network structure learnng s P problem, now typcal method has K2 [8] developed by Cooper and Hers2kovts. pv ( AV ( ), pav ( )) pv ( pav ( )) 2) A condtonal probabltes table (ab. CPT) asso- 5. Model of Selecton Method cated wth each node P. The CPT s expressed as PV ( pav ( )) whch pctures the mutual relatonshp of The method searchs the maxmum frequent temsets each node and ts parent nodes. odes wth no parent from a large data set, then fnd potental customers from have a very smple probablty table, gvng the pror those maxmum frequent temsets by usng a bayesan probablty dstrbuton of the node. odes wth parents network classfer. In detal, we can get result by takng are much more complcated. These nodes have cond- followng steps: tonal probablty tables, whch gve a probablty dstr- 1) All users share the same attrbutes. For each buton for every combnaton of states of the varable s attrbute, users have dfferent values from each other. We parents. get all sample s values of each attrbute from the data set, The Bayesan network can represent all of the nodes and dvde these values nto k segments usng the method jont probablty due to the node relatonshp and the presented n Subsecton condtonal probablty table. Applyng the condtonal 2) Based on the data processng n step 1, set a ndependence nto the chan rule, we get the followng mnmum support number and mne all of the FCI. expresson: 3) Every FCI contan a certan number of samples, collect these samples and make them to be a new data set n pv (, V,... V) pv ( pav ( )) D. 1 2 n 4) Learn bayesan network from data set D whch s Bayesan etwork Structure Learnng Consder a fnte set U { V, V,... V of dscrete random 1 2 n } varables where each varable V may take on values from a fnte set, denoted by Val( V ). Formally, a Bayesan network for U s a par B G,, and defnes a unque jont probablty dstrbuton over U. The problem of learnng a Bayesan network can be nformally stated as: Gven a tranng set D { u, u,... u } of nstances of U, fnd a network 1 2 n B that best matches D. The common approach to ths pro- blem s to ntroduce a scorng functon that evaluates each network wth respect to the tranng data, and then to search for the best network accordng to ths functon. The two man scorng functons commonly used to learn Bayesan networks are the Bayesan scorng functon [8], and the functon based on the prncple of mnmal descrpton length (ab. MDL) [7]. We only ntroduce MDL scorng functon. The MDL scorng functon of a network B gven a tranng data set D, wrtten MDL( B D ), s gven by followng expreson: formed n step 3, usng K2 algorthm. 5) For samples wthout knowng s postve or s negatve, calculate ts postve probablty by bayesan network constructed n 4. Once we get the probablty, we can get result by comparng t wth gven threshold. If a user s predcted to be postve, then the recommendaton system wll send the advertsement message to ths user. On the other hand, user who are predcted to be negatve wll not receve advertsement messages. 6. Experment In experments we try the new method as well as method based on nave bayesan claasfer so that we can compare the results between both methods Classfy Usng ave Bayesan Classfer 1) Dscretze the values and tran nave bayesan classfer. 2) Gve a decson threshold and classfy the test data set usng the traned nave bayesan classfer. Results are shown n Table 1. Copyrght 2010 ScRes.
4 S. Y. XIOG ET AL. 57 Success rate umber of segments Fgure 3. Success rate for dfferent segments number Mnmum support number counter Fgure 4. Success rate for dfferent Mnmum support number. Success rate Fgure 5. Bayesan network learned by GEIE Steps of ew Method 1) We dvde the values of each attrbute nto k segments. Whle k s 10, 20, 30 respectvly, the results are shown n Fgure 3. It shows that the best result occurs whle k s 10. 2) Gven a mnmum support number and search FCI, then compare the expermental results when s 10, 20, 30, 40, 50 respectvly. The results are shown n Fgure 4. It shows that we get the hghest success rate whle s 50. We don t consder a mnmum support number larger than 50 because the coverage of potental customer wll decrease too much, see the data n Table 2. 3) Learn bayesan network from D usng a open source bayesan tool called GEIE ( whch uses K2 algorthm to learn Bayesan network structure. We get a Bayesan network shown n Fgure 5, where GPRS_ FLOW, MMS_FEE, etc are names of attrbutes. YD- SUCCESS s the target attrbute to be predcted. 4) Last step s predctng by Bayesan network classfer. Snce the number of negatve samples s much more large than number of postve samples, we set the threshold to be 0.1. That means f a sample has a probablty to be postve sample larger than 0.1, then t s determned to be postve Expermental Results In ths experment, we use a data set from a moble servce provder. We dvde the users nto two parts. One part wth the user number of s tranng data set. The other part wth users ncludng 577 postve users s test data set. For tradtonal method whch sends advertsng message to all users. It sends advertsements and gets 577 customers, the success rate s 4.7%. It sends advertsement to 100% of the users, and gets 100% of the potental customers. For method based on naïve Bayesan classfer whose results shown n Table 1, the results are almost the same as the results of tradtonal method. Obvously, naïve Bayesan classfer does not work well wth moble user s data set. For new method, the results are shown n Table 2. Take mnnum support number 50 for example, t sends 1621 advertsements and gets 303 customers, the success rate s 18.69%. It sends advertsement to 13.22% of all users, and gets 49.78% of the potental customers. The results show that ths user selecton method ncreases success rate effcently. Even the lowest success Copyrght 2010 ScRes.
5 58 S. Y. XIOG ET AL. Mn support umber Table 1. Results of naïve Bayesan classfer. Threshold Postve Correct Success rate % % % / Ta ble 2 Experment result. Pos tve Correct Success rate Cost Coverage of potental customer % 46.55% 92.73% Recomendaton system becomes more and more mpor- tan t to servce provders now. User selecton s one of the dffculty problems. Ths paper presents a method to select users, the method searchs the maxmum frequent temsets from a large data set, then fnd potental customers from those maxmum frequent temsets by usng a bayesan network classfer. The success rate s mproved dramatcally after usng the method. Ths method also cuts down advertsement cost for moble operators and avods makng large number of Spam messages. There are a lot of data sets whch have smlar features wth moble user s data set, so ths method can be used n many smlar advertsng recommendaton systems. It has a good unversalty. 8. Acknowledgement Th s work was supported by the natonal Hgh-tech Research and Development Plan of Chna under grant o.2007aa01z417 and the 111 Project of Chna under grant o. B References % 43.55% 84.95% [1] E. Rch, User modelng va stereotypes [D], Cogntve % 35.15% 71.71% Scence, Vol. 3, o. 4, pp , % 27.30% 63.86% [2] U. M. Fayyad, G. Patecsky-Shapro, and P. Smyth, % 13.22% 49.78% Advances n knowledge dscovery and data mnng [M], Calforna, AAAI/MIT Press, rate 9.85%, s dramat cally hgher than the success ra te [3] R. Agrawal and M. Srkant, Fast algorthms for mnng of tradtonal method and of naïve Bayesan classfer. It assocaton rules n large databases [R], IBM Almaden also saves advertsng cost and cuts down Spam remarkably. Research Center, Tech Rep: RJ9839, [4] J. W. Han, J. Pe, Y. W. Yn, et al., Mnng frequent patterns wthout canddate generaton: A frequent-pattern 7. Conclusons tree approach [J], Data Mnng and Knowledge Dscovery, Vol. 8, o. 1, pp , [5] F. V. Jensen, An ntroducton to Bayesan networks, Sprnger, ew York, [6] M. desjarns, Representng and reasonng wth probabl- Bayesan stc knowledge: A Bayesan approach, Uncertanty n Artfcal Intellgence, pp [7]. Fredman, D. Geger, and M. Goldszmdt, network classfers, Machne Learnng, pp , [8] G. Cooper and E. Herskovts, A Bayesan method for the nducton of probablstc networks from data [D], Machne Learnng, Vol. 9, o. 4, pp , Copyrght 2010 ScRes.
Cluster Analysis of Electrical Behavior
Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School
More 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 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 informationLearning the Kernel Parameters in Kernel Minimum Distance Classifier
Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department
More 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 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 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 informationSupport Vector Machines
Support Vector Machnes Decson surface s a hyperplane (lne n 2D) n feature space (smlar to the Perceptron) Arguably, the most mportant recent dscovery n machne learnng In a nutshell: map the data to a predetermned
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 informationImplementation Naïve Bayes Algorithm for Student Classification Based on Graduation Status
Internatonal Journal of Appled Busness and Informaton Systems ISSN: 2597-8993 Vol 1, No 2, September 2017, pp. 6-12 6 Implementaton Naïve Bayes Algorthm for Student Classfcaton Based on Graduaton Status
More informationAn Anti-Noise Text Categorization Method based on Support Vector Machines *
An Ant-Nose Text ategorzaton Method based on Support Vector Machnes * hen Ln, Huang Je and Gong Zheng-Hu School of omputer Scence, Natonal Unversty of Defense Technology, hangsha, 410073, hna chenln@nudt.edu.cn,
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 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 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 informationContent 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 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 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 informationAssociative Based Classification Algorithm For Diabetes Disease Prediction
Internatonal Journal of Engneerng Trends and Technology (IJETT) Volume-41 Number-3 - November 016 Assocatve Based Classfcaton Algorthm For Dabetes Dsease Predcton 1 N. Gnana Deepka, Y.surekha, 3 G.Laltha
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 informationParallel Implementation of Classification Algorithms Based on Cloud Computing Environment
TELKOMNIKA, Vol.10, No.5, September 2012, pp. 1087~1092 e-issn: 2087-278X accredted by DGHE (DIKTI), Decree No: 51/Dkt/Kep/2010 1087 Parallel Implementaton of Classfcaton Algorthms Based on Cloud Computng
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 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 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 Modified Median Filter for the Removal of Impulse Noise Based on the Support Vector Machines
A Modfed Medan Flter for the Removal of Impulse Nose Based on the Support Vector Machnes H. GOMEZ-MORENO, S. MALDONADO-BASCON, F. LOPEZ-FERRERAS, M. UTRILLA- MANSO AND P. GIL-JIMENEZ Departamento de Teoría
More informationBAYESIAN MULTI-SOURCE DOMAIN ADAPTATION
BAYESIAN MULTI-SOURCE DOMAIN ADAPTATION SHI-LIANG SUN, HONG-LEI SHI Department of Computer Scence and Technology, East Chna Normal Unversty 500 Dongchuan Road, Shangha 200241, P. R. Chna E-MAIL: slsun@cs.ecnu.edu.cn,
More informationSupport Vector Machines
/9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.
More informationEmpirical Distributions of Parameter Estimates. in Binary Logistic Regression Using Bootstrap
Int. Journal of Math. Analyss, Vol. 8, 4, no. 5, 7-7 HIKARI Ltd, www.m-hkar.com http://dx.do.org/.988/jma.4.494 Emprcal Dstrbutons of Parameter Estmates n Bnary Logstc Regresson Usng Bootstrap Anwar Ftranto*
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 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 informationTF 2 P-growth: An Efficient Algorithm for Mining Frequent Patterns without any Thresholds
TF 2 P-growth: An Effcent Algorthm for Mnng Frequent Patterns wthout any Thresholds Yu HIRATE, Ego IWAHASHI, and Hayato YAMANA Graduate School of Scence and Engneerng, Waseda Unversty {hrate, ego, yamana}@yama.nfo.waseda.ac.jp
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 information12/2/2009. Announcements. Parametric / Non-parametric. Case-Based Reasoning. Nearest-Neighbor on Images. Nearest-Neighbor Classification
Introducton to Artfcal Intellgence V22.0472-001 Fall 2009 Lecture 24: Nearest-Neghbors & Support Vector Machnes Rob Fergus Dept of Computer Scence, Courant Insttute, NYU Sldes from Danel Yeung, John DeNero
More informationMaximum Variance Combined with Adaptive Genetic Algorithm for Infrared Image Segmentation
Internatonal Conference on Logstcs Engneerng, Management and Computer Scence (LEMCS 5) Maxmum Varance Combned wth Adaptve Genetc Algorthm for Infrared Image Segmentaton Huxuan Fu College of Automaton Harbn
More 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 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 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 informationFast Feature Value Searching for Face Detection
Vol., No. 2 Computer and Informaton Scence Fast Feature Value Searchng for Face Detecton Yunyang Yan Department of Computer Engneerng Huayn Insttute of Technology Hua an 22300, Chna E-mal: areyyyke@63.com
More informationBioTechnology. An Indian Journal FULL PAPER. Trade Science Inc.
[Type text] [Type text] [Type text] ISSN : 0974-74 Volume 0 Issue BoTechnology 04 An Indan Journal FULL PAPER BTAIJ 0() 04 [684-689] Revew on Chna s sports ndustry fnancng market based on market -orented
More informationInvestigating the Performance of Naïve- Bayes Classifiers and K- Nearest Neighbor Classifiers
Journal of Convergence Informaton Technology Volume 5, Number 2, Aprl 2010 Investgatng the Performance of Naïve- Bayes Classfers and K- Nearest Neghbor Classfers Mohammed J. Islam *, Q. M. Jonathan Wu,
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 informationContext-Specific Bayesian Clustering for Gene Expression Data
Context-Specfc Bayesan Clusterng for Gene Expresson Data Yoseph Barash School of Computer Scence & Engneerng Hebrew Unversty, Jerusalem, 91904, Israel hoan@cs.huj.ac.l Nr Fredman School of Computer Scence
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 informationRecommended Items Rating Prediction based on RBF Neural Network Optimized by PSO Algorithm
Recommended Items Ratng Predcton based on RBF Neural Network Optmzed by PSO Algorthm Chengfang Tan, Cayn Wang, Yuln L and Xx Q Abstract In order to mtgate the data sparsty and cold-start problems of recommendaton
More informationDeep Classification in Large-scale Text Hierarchies
Deep Classfcaton n Large-scale Text Herarches Gu-Rong Xue Dkan Xng Qang Yang 2 Yong Yu Dept. of Computer Scence and Engneerng Shangha Jao-Tong Unversty {grxue, dkxng, yyu}@apex.sjtu.edu.cn 2 Hong Kong
More informationImproving anti-spam filtering, based on Naive Bayesian and neural networks in multi-agent filters
J. Appl. Envron. Bol. Sc., 5(7S)381-386, 2015 2015, TextRoad Publcaton ISSN: 2090-4274 Journal of Appled Envronmental and Bologcal Scences www.textroad.com Improvng ant-spam flterng, based on Nave Bayesan
More informationGeneralized Additive Bayesian Network Classifiers
Generalzed Addtve Bayesan Network Classfers Janguo L and Changshu Zhang and Tao Wang and Ymn Zhang Intel Chna Research Center, Bejng, Chna Department of Automaton, Tsnghua Unversty, Chna {janguo.l, tao.wang,
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 informationCSCI 5417 Information Retrieval Systems Jim Martin!
CSCI 5417 Informaton Retreval Systems Jm Martn! Lecture 11 9/29/2011 Today 9/29 Classfcaton Naïve Bayes classfcaton Ungram LM 1 Where we are... Bascs of ad hoc retreval Indexng Term weghtng/scorng Cosne
More informationDetermining Fuzzy Sets for Quantitative Attributes in Data Mining Problems
Determnng Fuzzy Sets for Quanttatve Attrbutes n Data Mnng Problems ATTILA GYENESEI Turku Centre for Computer Scence (TUCS) Unversty of Turku, Department of Computer Scence Lemmnkäsenkatu 4A, FIN-5 Turku
More informationModular PCA Face Recognition Based on Weighted Average
odern Appled Scence odular PCA Face Recognton Based on Weghted Average Chengmao Han (Correspondng author) Department of athematcs, Lny Normal Unversty Lny 76005, Chna E-mal: hanchengmao@163.com Abstract
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 informationLearning Distributed Bayesian Network Structure Using Majority-based Method
Learnng Dstrbuted Bayesan Network Structure Usng Majorty-based Method Sachn Shetty Rowan Unversty Glassboro, NJ 08028, USA Phone: 856-256-5379 Fax: 856-256-5241 Emal: shetty@rowan.edu Mn Song Old Domnon
More informationSpam Filtering Based on Support Vector Machines with Taguchi Method for Parameter Selection
E-mal Spam Flterng Based on Support Vector Machnes wth Taguch Method for Parameter Selecton We-Chh Hsu, Tsan-Yng Yu E-mal Spam Flterng Based on Support Vector Machnes wth Taguch Method for Parameter Selecton
More informationLecture 5: Probability Distributions. Random Variables
Lecture 5: Probablty Dstrbutons Random Varables Probablty Dstrbutons Dscrete Random Varables Contnuous Random Varables and ther Dstrbutons Dscrete Jont Dstrbutons Contnuous Jont Dstrbutons Independent
More informationAn Evolvable Clustering Based Algorithm to Learn Distance Function for Supervised Environment
IJCSI Internatonal Journal of Computer Scence Issues, Vol. 7, Issue 5, September 2010 ISSN (Onlne): 1694-0814 www.ijcsi.org 374 An Evolvable Clusterng Based Algorthm to Learn Dstance Functon for Supervsed
More informationAn Improved Image Segmentation Algorithm Based on the Otsu Method
3th ACIS Internatonal Conference on Software Engneerng, Artfcal Intellgence, Networkng arallel/dstrbuted Computng An Improved Image Segmentaton Algorthm Based on the Otsu Method Mengxng Huang, enjao Yu,
More informationSupport Vector Machines. CS534 - Machine Learning
Support Vector Machnes CS534 - Machne Learnng Perceptron Revsted: Lnear Separators Bnar classfcaton can be veed as the task of separatng classes n feature space: b > 0 b 0 b < 0 f() sgn( b) Lnear Separators
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 informationSolitary and Traveling Wave Solutions to a Model. of Long Range Diffusion Involving Flux with. Stability Analysis
Internatonal Mathematcal Forum, Vol. 6,, no. 7, 8 Soltary and Travelng Wave Solutons to a Model of Long Range ffuson Involvng Flux wth Stablty Analyss Manar A. Al-Qudah Math epartment, Rabgh Faculty of
More informationFAHP and Modified GRA Based Network Selection in Heterogeneous Wireless Networks
2017 2nd Internatonal Semnar on Appled Physcs, Optoelectroncs and Photoncs (APOP 2017) ISBN: 978-1-60595-522-3 FAHP and Modfed GRA Based Network Selecton n Heterogeneous Wreless Networks Xaohan DU, Zhqng
More informationVirtual Machine Migration based on Trust Measurement of Computer Node
Appled Mechancs and Materals Onlne: 2014-04-04 ISSN: 1662-7482, Vols. 536-537, pp 678-682 do:10.4028/www.scentfc.net/amm.536-537.678 2014 Trans Tech Publcatons, Swtzerland Vrtual Machne Mgraton based on
More informationAn Entropy-Based Approach to Integrated Information Needs Assessment
Dstrbuton Statement A: Approved for publc release; dstrbuton s unlmted. An Entropy-Based Approach to ntegrated nformaton Needs Assessment June 8, 2004 Wllam J. Farrell Lockheed Martn Advanced Technology
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 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 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 informationBiological Sequence Mining Using Plausible Neural Network and its Application to Exon/intron Boundaries Prediction
Bologcal Sequence Mnng Usng Plausble Neural Networ and ts Applcaton to Exon/ntron Boundares Predcton Kuochen L, Dar-en Chang, and Erc Roucha CECS, Unversty of Lousvlle, Lousvlle, KY 40292, USA Yuan Yan
More informationA Simple Methodology for Database Clustering. Hao Tang 12 Guangdong University of Technology, Guangdong, , China
for Database Clusterng Guangdong Unversty of Technology, Guangdong, 0503, Chna E-mal: 6085@qq.com Me Zhang Guangdong Unversty of Technology, Guangdong, 0503, Chna E-mal:64605455@qq.com Database clusterng
More informationYan et al. / J Zhejiang Univ-Sci C (Comput & Electron) in press 1. Improving Naive Bayes classifier by dividing its decision regions *
Yan et al. / J Zhejang Unv-Sc C (Comput & Electron) n press 1 Journal of Zhejang Unversty-SCIENCE C (Computers & Electroncs) ISSN 1869-1951 (Prnt); ISSN 1869-196X (Onlne) www.zju.edu.cn/jzus; www.sprngerlnk.com
More informationAnnouncements. Supervised Learning
Announcements See Chapter 5 of Duda, Hart, and Stork. Tutoral by Burge lnked to on web page. Supervsed Learnng Classfcaton wth labeled eamples. Images vectors n hgh-d space. Supervsed Learnng Labeled eamples
More informationA Heuristic for Mining Association Rules In Polynomial Time
A Heurstc for Mnng Assocaton Rules In Polynomal Tme E. YILMAZ General Electrc Card Servces, Inc. A unt of General Electrc Captal Corporaton 6 Summer Street, MS -39C, Stamford, CT, 697, U.S.A. egemen.ylmaz@gecaptal.com
More informationTESTING AND IMPROVING LOCAL ADAPTIVE IMPORTANCE SAMPLING IN LJF LOCAL-JT IN MULTIPLY SECTIONED BAYESIAN NETWORKS
TESTING AND IMPROVING LOCAL ADAPTIVE IMPORTANCE SAMPLING IN LJF LOCAL-JT IN MULTIPLY SECTIONED BAYESIAN NETWORKS Dan Wu 1 and Sona Bhatt 2 1 School of Computer Scence Unversty of Wndsor, Wndsor, Ontaro
More informationCase Mining from Large Databases
Case Mnng from Large Databases Qang Yang and Hong Cheng Department of Computer Scence, Hong Kong Unversty of Scence and Technology, Clearwater Bay, Kowloon Hong Kong {qyang, csch}@cs.ust.hk http://www.cs.ust.hk/~qyang
More informationNUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS
ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana
More informationUsing Neural Networks and Support Vector Machines in Data Mining
Usng eural etworks and Support Vector Machnes n Data Mnng RICHARD A. WASIOWSKI Computer Scence Department Calforna State Unversty Domnguez Hlls Carson, CA 90747 USA Abstract: - Multvarate data analyss
More informationFace Recognition Method Based on Within-class Clustering SVM
Face Recognton Method Based on Wthn-class Clusterng SVM Yan Wu, Xao Yao and Yng Xa Department of Computer Scence and Engneerng Tong Unversty Shangha, Chna Abstract - A face recognton method based on Wthn-class
More informationUnder-Sampling Approaches for Improving Prediction of the Minority Class in an Imbalanced Dataset
Under-Samplng Approaches for Improvng Predcton of the Mnorty Class n an Imbalanced Dataset Show-Jane Yen and Yue-Sh Lee Department of Computer Scence and Informaton Engneerng, Mng Chuan Unversty 5 The-Mng
More informationFrom Comparing Clusterings to Combining Clusterings
Proceedngs of the Twenty-Thrd AAAI Conference on Artfcal Intellgence (008 From Comparng Clusterngs to Combnng Clusterngs Zhwu Lu and Yuxn Peng and Janguo Xao Insttute of Computer Scence and Technology,
More informationProblem Set 3 Solutions
Introducton to Algorthms October 4, 2002 Massachusetts Insttute of Technology 6046J/18410J Professors Erk Demane and Shaf Goldwasser Handout 14 Problem Set 3 Solutons (Exercses were not to be turned n,
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 informationNetwork Intrusion Detection Based on PSO-SVM
TELKOMNIKA Indonesan Journal of Electrcal Engneerng Vol.1, No., February 014, pp. 150 ~ 1508 DOI: http://dx.do.org/10.11591/telkomnka.v1.386 150 Network Intruson Detecton Based on PSO-SVM Changsheng Xang*
More informationA Heuristic for Mining Association Rules In Polynomial Time*
Complete reference nformaton: Ylmaz, E., E. Trantaphyllou, J. Chen, and T.W. Lao, (3), A Heurstc for Mnng Assocaton Rules In Polynomal Tme, Computer and Mathematcal Modellng, No. 37, pp. 9-33. A Heurstc
More informationA Clustering Algorithm for Chinese Adjectives and Nouns 1
Clusterng lgorthm for Chnese dectves and ouns Yang Wen, Chunfa Yuan, Changnng Huang 2 State Key aboratory of Intellgent Technology and System Deptartment of Computer Scence & Technology, Tsnghua Unversty,
More informationExtraction of Fuzzy Rules from Trained Neural Network Using Evolutionary Algorithm *
Extracton of Fuzzy Rules from Traned Neural Network Usng Evolutonary Algorthm * Urszula Markowska-Kaczmar, Wojcech Trelak Wrocław Unversty of Technology, Poland kaczmar@c.pwr.wroc.pl, trelak@c.pwr.wroc.pl
More informationClassification Methods
1 Classfcaton Methods Ajun An York Unversty, Canada C INTRODUCTION Generally speakng, classfcaton s the acton of assgnng an object to a category accordng to the characterstcs of the object. In data mnng,
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 informationBIN XIA et al: AN IMPROVED K-MEANS ALGORITHM BASED ON CLOUD PLATFORM FOR DATA MINING
An Improved K-means Algorthm based on Cloud Platform for Data Mnng Bn Xa *, Yan Lu 2. School of nformaton and management scence, Henan Agrcultural Unversty, Zhengzhou, Henan 450002, P.R. Chna 2. College
More informationSHAPE RECOGNITION METHOD BASED ON THE k-nearest NEIGHBOR RULE
SHAPE RECOGNITION METHOD BASED ON THE k-nearest NEIGHBOR RULE Dorna Purcaru Faculty of Automaton, Computers and Electroncs Unersty of Craoa 13 Al. I. Cuza Street, Craoa RO-1100 ROMANIA E-mal: dpurcaru@electroncs.uc.ro
More informationClassification / Regression Support Vector Machines
Classfcaton / Regresson Support Vector Machnes Jeff Howbert Introducton to Machne Learnng Wnter 04 Topcs SVM classfers for lnearly separable classes SVM classfers for non-lnearly separable classes SVM
More informationAn Empirical Comparative Study of Online Handwriting Chinese Character Recognition:Simplified v.s.traditional
2013 12th Internatonal Conference on Document Analyss and Recognton An Emprcal Comparatve Study of Onlne Handwrtng Chnese Recognton:Smplfed v.s.tradtonal Yan Gao, Lanwen Jn +, Wexn Yang School of Electronc
More informationNon-Negative Matrix Factorization and Support Vector Data Description Based One Class Classification
IJCSI Internatonal Journal of Computer Scence Issues, Vol. 9, Issue 5, No, September 01 ISSN (Onlne): 1694-0814 www.ijcsi.org 36 Non-Negatve Matrx Factorzaton and Support Vector Data Descrpton Based One
More informationA METHOD FOR FACTOR SCREENING OF SIMULATION EXPERIMENTS BASED ON ASSOCIATION RULE MINING
A METHOD FOR FACTOR SCREENING OF SIMULATION EXPERIMENTS BASED ON ASSOCIATION RULE MINING Lngyun Lu (a), We L (b), Png Ma (c), Mng Yang (d) Control and Smulaton Center, Harbn Insttute of Technology, Harbn
More informationMULTISPECTRAL REMOTE SENSING IMAGE CLASSIFICATION WITH MULTIPLE FEATURES
MULISPECRAL REMOE SESIG IMAGE CLASSIFICAIO WIH MULIPLE FEAURES QIA YI, PIG GUO, Image Processng and Pattern Recognton Laboratory, Bejng ormal Unversty, Bejng 00875, Chna School of Computer Scence and echnology,
More informationOptimizing Naïve Bayes Algorithm for SMS Spam Filtering on Mobile Phone to Reduce the Consumption of Resources
Journal of Computers Vol. 28, No. 3, 2017, pp. 174-183 do:10.3966/199115592017062803014 Optmzng Naïve Bayes Algorthm for SMS Spam Flterng on Moble Phone to Reduce the Consumpton of Resources L-qun Bao
More informationAdaptive Transfer Learning
Adaptve Transfer Learnng Bn Cao, Snno Jaln Pan, Yu Zhang, Dt-Yan Yeung, Qang Yang Hong Kong Unversty of Scence and Technology Clear Water Bay, Kowloon, Hong Kong {caobn,snnopan,zhangyu,dyyeung,qyang}@cse.ust.hk
More informationA Powerful Feature Selection approach based on Mutual Information
6 IJCN Internatonal Journal of Computer cence and Network ecurty, VOL.8 No.4, Aprl 008 A Powerful Feature electon approach based on Mutual Informaton Al El Akad, Abdelall El Ouardgh, and Drss Aboutadne
More informationCOMPARITIVE ANALYSIS OF FUZZY DECISION TREE AND LOGISTIC REGRESSION METHODS FOR PAVEMENT TREATMENT PREDICTION
COMPARITIVE ANALYSIS OF FUZZY DECISION TREE AND LOGISTIC REGRESSION METHODS FOR PAVEMENT TREATMENT PREDICTION DEVINDER KAUR, HARICHARAN PULUGURTA Department of Electrcal and Computer Scences, Department
More informationA Topology-aware Random Walk
A Topology-aware Random Walk Inkwan Yu, Rchard Newman Dept. of CISE, Unversty of Florda, Ganesvlle, Florda, USA Abstract When a graph can be decomposed nto clusters of well connected subgraphs, t s possble
More informationA Post Randomization Framework for Privacy-Preserving Bayesian. Network Parameter Learning
A Post Randomzaton Framework for Prvacy-Preservng Bayesan Network Parameter Learnng JIANJIE MA K.SIVAKUMAR School Electrcal Engneerng and Computer Scence, Washngton State Unversty Pullman, WA. 9964-75
More informationThe Man-hour Estimation Models & Its Comparison of Interim Products Assembly for Shipbuilding
Internatonal Journal of Operatons Research Internatonal Journal of Operatons Research Vol., No., 9 4 (005) The Man-hour Estmaton Models & Its Comparson of Interm Products Assembly for Shpbuldng Bn Lu and
More informationSum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints
Australan Journal of Basc and Appled Scences, 2(4): 1204-1208, 2008 ISSN 1991-8178 Sum of Lnear and Fractonal Multobjectve Programmng Problem under Fuzzy Rules Constrants 1 2 Sanjay Jan and Kalash Lachhwan
More information