An Inductive Fuzzy Classification Approach applied to Individual Marketing
|
|
- Kristina Blair
- 8 years ago
- Views:
Transcription
1 An Inductve Fuzzy Classfcaton Approach appled to Indvdual Marketng Mchael Kaufmann, Andreas Meer Abstract A data mnng methodology for an nductve fuzzy classfcaton s ntroduced. The nducton step s based on dervng fuzzy restrctons from data whose membershp functons are nferred from normalzed lkelhood ratos of target class membershp. A case study s presented where ths data mnng process s appled n predctve analytcs for an ndvdual marketng campagn n the onlne channel of a fnancal servce provder. Index Terms Data Mnng, Fuzzy Classfcaton, Fuzzy Logc, Predcton, Indvdual Marketng, Fuzzy Target Groups I. INTRODUCTION AND MOTIVATION Fuzzy Logc proposed by L. A. Zadeh s based on ntutve reasonng and takes nto account human subectvty and mprecson. Unlke classcal data mnng technques such as cluster or regresson analyss, fuzzy logc enables the use of classfcaton va fuzzy propostons comnbed by fuzzy logc operators. Fuzzy classfcaton creates more subtle and smooth dstnctons between equvalence classes than crsp classfcaton. [] and [2] presented the mplementaton of a classfcaton query language n order to flter data usng lngustc terms. They proposed the nvestgaton of a desgn methodology for the applcaton of fuzzy classfcaton n data mnng. In ths paper, a data mnng methodology usng nductve fuzzy classfcaton s ntroduced n order to address ths ssue. [3] emphaszes the nductve aspect of data mnng, namely learnng, and suggests a motvaton: Predcton. Ths explans how pattern models are to be dscovered and for what they can be used. If appled correctly, data models not only descrbe the past, but also help to forecast the future. In ths sense, successful data mnng s a practcal demonstraton of nductve logc: Learnng generalzatons from specfc examples. Hence, ths paper presents a methodology for the probablstc nducton of a multvarate fuzzy classfcaton based on lkelhood ratos. M. Kaufmann (correspondng author) s PhD-Student at the Research Center for Fuzzy Marketng Methods ( at the Unversty of Frbourg Swtzerland (e-mal: Mchael.Kaufmann@unfr.ch). A. Meer s Professor of Informaton Systems Resarch. He s the head of the Research Center for Fuzzy Marketng Methods at the Unversty of Frbourg Swtzerland. The paper s structured as follows: Secton II ntroduces nductve fuzzy classfcaton theoretcally. Secton III proposes a pragmatc methodology for an nductve fuzzy classfcaton. Secton IV presents a case study of a database marketng campagn where the target group has been selected usng the proposed method. Secton V summarzes the results and shows ssues for further research. II. THEORY OF INDUCTIVE FUZZY CLASSIFICATION Ths secton summarzes the theoretcal background of classfcaton, nducton and fuzzy restrctons n order to develop a theory of Inductve Fuzzy Classfcaton. A. Classfcaton Classfcaton s the process of groupng elements satsfyng the same characterstc constrants nto a set. In a class logc, a class C {e P(e)} s defned as a set C of elements e satsfyng predcate P. In the multvarate case, P refers to multple element attrbutes as a formula combned by logcal operators. B. Inductve Classfcaton Inductve classfcaton, a form of supervsed learnng ([3], [4]), s the process of learnng from examples to decde whether an element e belongs to a gven class y based on ts attrbutes. A tranng data set d s a (n+) m matrx wth n columns,, n, and a column Y ndcatng the class membershp. The columns, {,, n} are called dependent varables or attrbutes, and Y s called the target varable or the class label. The rows represent data elements e, {,, m}. The matrx content conssts of attrbute values x representng the value of the -th attrbute for the -th element, and of labels y representng the class membershp for the -th element. In case of a bnary classfcaton, for each row ndex the label y s equal to f and only f the element e s n class y. The learnng process nduces a model M from the tranng set d based on the dstrbuton of the target varable on the dependent varables, mappng from the Cartesan product of the dependent varable s domans nto the set {0,}. The model can be used to predct the class membershp of data elements. For an element e d n a new dataset d wth unknown class membershp, a predcton Y M(x,, x n ) can be computed by applyng the model to the element s attrbute values. When elements are grouped nto an
2 2 equvalence class y {e Y }, ths knd of classfcaton s nductve because the classfcaton constrants predcate has been learned from the data. C. Fuzzy Classfcaton Fuzzy classfcaton s the process of groupng elements nto a fuzzy set whose membershp functon s defned by the truth value of a fuzzy constrant predcate. A fuzzy class C s a fuzzy set [5] defned by a fuzzy proposton [6] e s R, where e s an element of a unverse of dscourse U and R s a fuzzy restrcton of U. The degree of membershp C (e) v(e s R) s defned by the truth value v of the correspondng fuzzy proposton.in the multvarate case, P s a fuzzy restrcton of U defned by a combnaton of fuzzy restrctons of attrbute domans combned by fuzzy logc operators. D. Inductve Fuzzy Classfcaton Consequently, Inductve Fuzzy Classfcaton (IFC) s the process of groupng elements nto a fuzzy set whose membershp functon s derved by supervsed learnng from data. In ths approach, the predctve model s represented by a multvarate fuzzy classfcaton. The fuzzy membershp functons of each attrbute are nduced from data usng a normalzaton of the lkelhood rato functon. The multvarate aggregaton of the unvarate fuzzy attrbute doman restrctons provdes the predcton of the classfer, whch can be defuzzfed usng an alpha-cut. The nducton step s not learnng of fuzzy rules themselves, as t s n many cases (e.g. [7], [8]) but nducng membershp functons to fuzzy restrctons of multple attrbute domans. The focus s not on the combnaton of fuzzfed nputs, but on the defnton of multvarate fuzzy set membershps for nducton, as presented n [9] and [0]. In fact, by usng lkelhood ratos for nducton, ths s a Bayes-lke approach, whose usefulness has been shown by []. III. A METHODOLOGY FOR INDUCTIVE FUZZY CLASSIFICATION As proposed by [], a desgn methodology s needed n order to apply fuzzy classfcaton to data mnng. In the followng secton, a conceptual framework for nductve fuzzy classfcaton s proposed. The am s to derve a multvarate fuzzy class nductvely n order to obtan a predctve classfcaton. The resultng classfer s able to perform a fuzzy classfcaton whch s nductve by learnng the fuzzy constrant predcate P from the lkelhood of class membershp. The resultng fuzzy class s based on a formula of fuzzy propostons on the dependent varables, whose truth values and correspondng fuzzy set membershp functons are derved from normalzed lkelhood ratos. Fgure : Inductve fuzzy classfcaton process As shown n Fgure, the process for dervng a fuzzy class by nducton conssts of preparng the data, selectng the relevant attrbutes, nducng membershp functons to the fuzzy restrctons on the attrbute domans from the tranng set, transformng the attrbute values nto unvarate fuzzy set membershp values, classfyng data elements by aggregatng the fuzzfed attrbutes nto a multvarate fuzzy classfcaton of data elements, and an evaluaton of the predctve performance of the resultng classfer. A. Inductve Fuzzy Classfcaton Process Overvew The dea of the process s to develop a well performng classfer whch accurately predcts the membershp of data elements e n the target class y. The classfer wll perform a fuzzy classfcaton by assgnng data elements to a fuzzy class y havng the followng property: The hgher the degree of membershp of a data element n the fuzzy class y, the greater the lkelhood of class membershp n the crsp target class y. In order to accomplsh ths, a tranng set s prepared form source data, and the relevant attrbutes are selected usng an nterestngness measure. For every attrbute, a fuzzy restrcton y on ts doman s defned based on the prncple of maxmum lkelhood of target class membershp. Each y s nduced from the data such that the degree of membershp of an attrbute value x n y s proportonal to the condtonal relatve frequency of Y. Then, n the unvarate classfcaton step, each relevant varable s fuzzfed by transformng t nto a degree of membershp n ts assocated fuzzy restrcton y. The multvarate fuzzy classfcaton step conssts of aggregatng the fuzzfed attrbutes nto one multvarate fuzzy class of data elements y. Ths fuzzy equvalence class then corresponds to an Inductve Fuzzy Classfcaton whch can be used for predcton. The last step of the process s model evaluaton by analyzng the predcton performance of the classfer. Ths s done by comparng the forecasts wth the real class membershps n a test set. In the followng subsectons, every step of the IFC process s descrbed n detal.
3 3 B. Data Preparaton In order to analyze the data, a tranng set s composed by combnng data from varous sources nto a sngle coherent matrx. All possbly relevant element attrbutes are merged nto one table. The class label for the target varable has to be defned, calculated, and added to the tranng set. The class label s restrcted to a bnary varable where ndcates membershp n the target class. C. Attrbute Selecton Only a few varables n the tranng set are actually relevant for predctng the class label. In order to select the relevant attrbutes, a varety of measures s avalable, for example mutual nformaton or ch squared ndependence. The resultng lst s a descendng rankng of dependent varables by ther relevance. Only the best n varables should be taken nto account, where the n-th best varable should stll show a sgnfcant dependence wth the target varable. D. Inducton of Membershp Functons Intutvely, the am s to assgn to every data element a membershp degree n the fuzzy data element class y. In order to accomplsh ths, y s derved as an aggregaton of fuzzy restrctons: For every attrbute, a fuzzy restrcton y s defned on ts doman whch represents the lkelhood of Y. The hgher the lkelhood of Y s n relaton to Y0, the greater should be the degree of membershp of a value x n y. The lkelhood of class membershp s transformed nto a membershp functon usng the followng method: For each value x dom( ), the fuzzy restrcton y s defned as the (sampled) condtonal probablty for Y, dvded by the sum of the condtonal probabltes of x for Y and Y0 respectvely. Thus, the nduced membershp degree of x n y s defned by x Y ) ( x) : () y x Y ) + x Y 0) Accordng to the defnton of lkelhood L(y x)x y) and lkelhood rato LR y (x)l(y x)/l(not(y) x), equaton () corresponds to a normalzaton of the lkelhood rato (NLR): y ( x) : + + LR ( x) Y x Y 0) L(0 x) + x Y ) L( x) : NLR ( x) x Y ) x Y ) + x Y 0) Y For categorcal varables, the calculaton of the membershp functon s straght forward: The degree of membershp for each value s defned by the correspondng NLR. However, for contnuous varables, the membershp functon could be contnuous too. In that case, a soluton s to calculate the NLR for quantles of dom( ), and to approxmate a contnuous functon. For examples of membershp functon nducton for categorcal or contnuous attrbutes, see the case study n secton IV. E. Unvarate Fuzzy Classfcaton of Attrbute Values Once the membershp functons have been nduced as descrbed n the prevous secton, the relevant attrbutes are fuzzfed. In order to transform the attrbutes nto degrees of membershp, each varable s transformed by applyng the membershp functon nduced by formula (). Ths step smply transforms every value x nto the membershp degree of the value n the correspondng fuzzy restrcton y. F. Multvarate Fuzzy Classfcaton of Data Elements In order to obtan a fuzzy membershp degree n the fuzzfed target class y for data elements e, multple attrbute membershp degrees are aggregated usng fuzzy logc operators. For smplcty, assume the followng equaton: y ( e ) y ( x ) That s, for every fuzzy restrcton on an attrbute doman there s a correspondng fuzzy set of data elements wth the same name. Consequently, the multvarate fuzzy class y s defned as an aggregaton usng a functon F : e ) : F( ( e ), K, ( e )) y' ( y There are dfferent possbltes for multvarate fuzzy aggregaton. In lterature, the focus s often on generatng a fuzzy rule base by searchng a functon as a combnaton of fuzzy logc operators. A smpler canddate for the aggregaton functon s the gamma operator [2] whch s composed of a multplcaton of an algebrac dsuncton and an algebrac conuncton weghted by an exponent parameter γ [0,]: Γ γ ( y : ( e ), K, ( e )) U y : ( γ ( e ) I y ( e ) ( x )) y yn γ γ ( yn ( x )) Ths aggregaton operator compensates between conuncton and dsuncton to a degree specfed by the γ-parameter. Thus, by combnng the nductvely fuzzfed attrbutes nto a multvarate fuzzy class of data elements defned by the membershp functon y :U [0,], a multvarate model s obtaned from the tranng set. Ths model can be used for Inductve Fuzzy Classfcaton of unlabeled data for predcton. Usng an alpha cut for an α [0,] leads to a bnary classfer. y γ (3)
4 4 Table : Frequences, condtonal probabltes and NLRs for the categorcal attrbute customer segment. : customer segment Y Y0 Y) Y0) NLR Y() Bass ' ' SMB 249 2' Gold 2'666 54' Premum 5'432 0' Total 29' '339 In the followng sectons, the applcaton of the data mnng process steps presented n secton III to the target group selecton s descrbed. Fgure 2: Analytcs appled to ndvdual marketng n the onlne channel G. Model Evaluaton In order to evaluate the predctve performance, the classfer s appled to a hold-out test set, and the predctons are compared wth the actual class label. There are dfferent metrcs to evaluate the predctve performance of the nduced model, for example the mutual nformaton between the predctons Y and the class labels Y, or alternatvely ther correlaton. Usng such a measure, the model parameters γ [0,] and α [0,] can be optmzed to yeld a maxmal predctve performance. IV. CASE STUDY: MEMBERSHIP FUNCTION INDUCTION FOR FUZZY TARGET GROUPS PostFnance Inc. ( s a proftable busness unt of Swss Post. Its actvtes contrbute sgnfcantly to the fnancal servces market n Swtzerland. PostFnance s an analytc enterprse feedng busness processes wth nformaton ganed from predctve analytcs. The marketng process uses data mnng scores on product affnty for ndvdual customers. Target groups for onlne marketng campagns are selected usng nductve classfcaton on customer databases, as llustrated by Fgure 2. For every customer, the ndvdual advertsement message s mapped accordng to the target group selected by predctve scorng. Inductve Fuzzy Classfcaton was appled n an onlne marketng campagn of PostFnance promotng nvestment funds. The am of the nductve target group selecton was to forecast the customers wth an enhanced probablty of buyng nvestments funds. The resultng fuzzy classfer yelded a fuzzy target group for an ndvdual marketng campagn, where every customer s part of the target group only to a certan degree. A. Data Preparaton In order to prepare a test set for model nducton, a sample customer data set was selected from the customer data warehouse. Ths data set contaned an anonymous customer number assocated wth customer specfc attrbutes. As target varable, the class label was set to f the customer had nvestment funds n hs product portfolo, and to 0 else. B. Attrbute Selecton Mutual nformaton of the dependent varables wth the target varable was chosen as a useful rankng method. The followng attrbutes were selected as relevant: Customer segment (0.047 bt) Number of products (0.036 bt) Overall balance (0.035 bt) Loyalty (0.02 bt) Customer group (0.06 bt) Balance on prvate account (0.04 bt) Age (0.03 bt) C. Inducton of Membershp Functons Usng the method presented n secton III, for each of the relevant attrbutes the fuzzy restrcton correspondng to the lkelhood of havng nvestment funds was nduced. In the followng secton, the nducton processes for a categorcal and a contnuous attrbute are descrbed n detal. As frst example, n the doman of the categorcal attrbute customer segment ( ), there are four values Bass, SMB, Gold and Premum. For customers who have not yet bought nvestment funds, the am s to defne a degree of membershp n a fuzzy restrcton on the customer segment doman n order to classfy them for ther lkelhood to buy that product n the future. The frequences and condtonal probabltes are presented n Table. The frst column ndcates the customer segment. Column Y contans the number of customers of each segment who have bought nvestment funds. Column Y0 contans the number of customers of each segment who have not bought that product.
5 5 Fuzzy Restrcton NLR Bass SMB Gold Premum Customer Segment Fgure 3: Membershp functon resultng from nducton for the categorcal attrbute customer segment Fuzzy Restrcton NLR F(x) 0.7/(+EP( *LN(x+))) '000 00'000 50'000 Overall Balance Fgure 4: Membershp functon resultng from nducton for the contnuous attrbute overall balance Accordng to formula (), the membershp degree of segment Bass n the fuzzy restrcton y was nduced as follows: y (" Bass") NLR " Bass" Y ) " Bass" Y ) + " Bass" Y 0) Y (" Bass") The other degrees of the membershp were nduced analogcally. The resultng values are shown by Table n column NLR. The correspondng membershp functon s llustrated by Fgure 3. As second example, for the contnuous attrbute overall balance the membershp functon was nduced n the followng way: Frst, the NLR for decles of the attrbute s doman was calculated analogcally to the prevous example, represented by grey squares n Fgure 4. Then, a functon A / ( + exb C ln(x + )))+D was ftted by optmzng the parameters A to D usng the method of [3]. The resultng membershp functon for the customer overall balance s shown as a lne n Fgure 4. D. Unvarate Fuzzy Classfcaton of Attrbute Values Every relevant attrbute of the orgnal data set was transformed to a fuzzy membershp degree usng SQL (Structured Query Language) drectly n the database. Table 2: Optmzaton of the gamma parameter for multvarate fuzzy aggregaton I(Y;Y') I(Y;Y') LR(Y') Gamma Gamma Gamma Categorcal varables were transformed usng a case dfferentaton statement. For example, the attrbute customer segment was transformed usng the followng SQL command: select case when customer_segment 'Bass' then 0.3 when customer_segment 'SMB' then 0.52 when customer_segment 'Gold' then 0.74 when customer_segment 'Premum' then 0.86 end as fuzzy_customer_segment from ads_funds Contnuous varables were transformed usng a functon expresson. For example, the attrbute overall balance was fuzzfed usng the followng SQL expresson: select 0.7/(+EP(7.-0.8*LN(overall_balance + )))+0.09 as fuzzy_overall_balance from ads_funds E. Multvarate Fuzzy Classfcaton of Data Elements The ndvdual fuzzy attrbute doman restrctons were then aggregated to a multvarate fuzzy class of customers. Ths was done usng an SQL statement mplementng the gamma operator defned n equaton (3). In order to defne the gamma parameter, dfferent performance measures were calculated. As shown n Table 2, a gamma of, correspondng to a algebrac dsuncton or full compensaton, was most successful. Thus, the multvarate fuzzy classfcaton was performed usng the followng SQL statement: select (-fuzzy_number_of_products) * (-fuzzy_customer_segment) * (-fuzzy_overall_balance) * (-fuzzy_loyalty) * (-fuzzy_age) * (-fuzzy_balance_on_prvate_account) * (-fuzzy_customer_group) ) as multvarate_fuzzy_classfcaton from ads_funds As a result, a fuzzy class of customers was calculated, whose degree of membershp ndcates the product affnty for nvestment funds and represents a product affnty score. Ths can be used n ndvdual marketng for defnng target groups for dfferent products.
6 6 Table 3: Resultng product sellng ratos per target group Test group Y Y0 Sold products : Fuzzy classfcaton % 2: Crsp classfcaton % 3: Random selecton % F. Model Evaluaton In order to test the resultng fuzzy classfer, a plot marketng campagn was performed usng the resultng fuzzy target group. A target group of 5000 customers wth the hghest membershp degree was selected from the fuzzy class usng an alpha cut (test group ). As a comparson, 5000 other customers were selected usng a crsp classfcaton (test group 2), usng the followng crsp constrants: Customer segment n {Gold, Premum, SMB} Customer_group 50 Plus Loyalty > 4 Number_of_products > Age between 35 and 75 Balance on prvate account > 3000 Overall balance > Thrd, 5000 customers were selected randomly (test group 3). To each of those 5000 customers, an onlne advertsement for nvestment funds was shown. After the marketng campagn, the product sellng rato wthn three months was measured. The results are shown n Table 3. The product sellng rato for the target group selected by nductve fuzzy classfcaton was the most effectve. V. CONCLUSION A methodology for an nductve fuzzy classfcaton was ntroduced. The nducton process conssts of generatng fuzzy restrctons on attrbute domans from normalzed lkelhood ratos. An ndvdual marketng campagn usng fuzzy nductve target group selecton showed that fuzzy classfcaton led to a hgher product sellng rato than crsp classfcaton or random selecton. In the case study, Inductve Fuzzy Classfcaton s predctons of product affnty were more accurate than those of the crsp classfcaton rules on exactly the same attrbutes. The concluson s that n comparson to crsp classfcaton, fuzzy classfcaton has an advantageous feature that leads to better predctve performance. Further research could encompass the followng ssues: Reference mplementaton of an IFC data mnng tool supportng and automatng the IFC process steps. An nterpreter for an nductve fuzzy classfcaton language (IFCL) s n development by a Master s Student Proect [4] Evaluaton of the IFC data mnng methodology usng dfferent benchmark data sets n comparson to conventonal predcton technques. Evaluaton of the mechansm how fuzzy classfcaton provdes more accurate predcton results than crsp classfcaton. Development of a conceptual framework for busness applcatons of fuzzy predctve analytcs n the context of nformaton systems research. ACKNOWLEDGMENT The authors thank Davd Wyder from PostFnance for supportng research wth a plot marketng campagn. The case study was conducted by the FMsquare reserach center for fuzzy marketng methods ( nvestgatng busness applcatons of fuzzy logc such as fuzzy data warehouses, fuzzy predctve analytcs or fuzzy customer performance measurement. REFERENCES [] A. Meer, G. Schndler, N. Werro. Fuzzy classfcaton on relatonal databases. In Galndo M. (Ed.): Handbook of Research on Fuzzy Informaton Processng n Databases. Volume II. Informaton Scence Reference, pp , [2] Ncolas Werro. Fuzzy classfcaton for onlne customers. PhD Thess, Unversty of Frbourg, [3] Ian H. Wtten and Ebe Frank. Data mnng, practcal machne learnng tools and technques, second edton. Morgan Kaufmann Publshers, [4] Eyke Hüllermeer. Fuzzy methods n machne learnng and data mnng: Status and prospects. Fuzzy Sets and Systems 56(3), , [5] L. A. Zadeh. Fuzzy sets. Informaton and Control, 8, , 965. [6] L. A. Zadeh. Calculus of fuzzy restrctons. In: L.A.Zadeh et al., Eds, Fuzzy sets and ther applcatons to cogntve and decson processes. NewYork: Academc Press, 975. [7] W. Danhu, T.S. Dllon, E.J. Chang. A data mnng approach for fuzzy classfcaton rule generaton. IFSA World Congress and 20th NAFIPS Internatonal Conference, 200. Jont 9th, pp vol.5, July 200. [8] Y. Hu, R. Chen and G. Tzeng. Fndng fuzzy classfcaton rules usng data mnng technques. Pattern Recogn. Lett. 24, -3, , [9] C.J. Km and B.D. Russell. Automatc generaton of membershp functon and fuzzy rule usng nductve reasonng. Thrd Internatonal Conference on Industral Fuzzy Control and Intellgent Systems, pp , 993. [0] T. Hong and S. Wang. Determnng approprate membershp functons to smplfy fuzzy nducton. Intell. Data Anal. 4, (Jan. 2000), [] S. Roychowdhury. Bayes-lke Classfer wth Fuzzy Lkelhood. Fuzzy Systems, 2006 IEEE Internatonal Conference on, pp , [2] H.-J. Zmmermann. Fuzzy set theory and ts applcatons. London: Kluwer Academc Publshers, 992. [3] E.G. John. Smplfed curve fttng usng spreadsheet add-ns. Int. J. Engng Ed. 4, , 998. [4] Phlppe Mayer. Implementaton of an nductve fuzzy classfcaton language nterpreter. Master s Dssertaton, Unversty of Frbourg, Swtzerland (n Progress).
benefit is 2, paid if the policyholder dies within the year, and probability of death within the year is ).
REVIEW OF RISK MANAGEMENT CONCEPTS LOSS DISTRIBUTIONS AND INSURANCE Loss and nsurance: When someone s subject to the rsk of ncurrng a fnancal loss, the loss s generally modeled usng a random varable or
More informationThe Development of Web Log Mining Based on Improve-K-Means Clustering Analysis
The Development of Web Log Mnng Based on Improve-K-Means Clusterng Analyss TngZhong Wang * College of Informaton Technology, Luoyang Normal Unversty, Luoyang, 471022, Chna wangtngzhong2@sna.cn Abstract.
More informationTHE APPLICATION OF DATA MINING TECHNIQUES AND MULTIPLE CLASSIFIERS TO MARKETING DECISION
Internatonal Journal of Electronc Busness Management, Vol. 3, No. 4, pp. 30-30 (2005) 30 THE APPLICATION OF DATA MINING TECHNIQUES AND MULTIPLE CLASSIFIERS TO MARKETING DECISION Yu-Mn Chang *, Yu-Cheh
More informationPerformance Management and Evaluation Research to University Students
631 A publcaton of CHEMICAL ENGINEERING TRANSACTIONS VOL. 46, 2015 Guest Edtors: Peyu Ren, Yancang L, Hupng Song Copyrght 2015, AIDIC Servz S.r.l., ISBN 978-88-95608-37-2; ISSN 2283-9216 The Italan Assocaton
More informationWhat is Candidate Sampling
What s Canddate Samplng Say we have a multclass or mult label problem where each tranng example ( x, T ) conssts of a context x a small (mult)set of target classes T out of a large unverse L of possble
More informationModule 2 LOSSLESS IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur
Module LOSSLESS IMAGE COMPRESSION SYSTEMS Lesson 3 Lossless Compresson: Huffman Codng Instructonal Objectves At the end of ths lesson, the students should be able to:. Defne and measure source entropy..
More informationBayesian Network Based Causal Relationship Identification and Funding Success Prediction in P2P Lending
Proceedngs of 2012 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 25 (2012) (2012) IACSIT Press, Sngapore Bayesan Network Based Causal Relatonshp Identfcaton and Fundng Success
More informationNEURO-FUZZY INFERENCE SYSTEM FOR E-COMMERCE WEBSITE EVALUATION
NEURO-FUZZY INFERENE SYSTEM FOR E-OMMERE WEBSITE EVALUATION Huan Lu, School of Software, Harbn Unversty of Scence and Technology, Harbn, hna Faculty of Appled Mathematcs and omputer Scence, Belarusan State
More informationThe OC Curve of Attribute Acceptance Plans
The OC Curve of Attrbute Acceptance Plans The Operatng Characterstc (OC) curve descrbes the probablty of acceptng a lot as a functon of the lot s qualty. Fgure 1 shows a typcal OC Curve. 10 8 6 4 1 3 4
More informationForecasting the Demand of Emergency Supplies: Based on the CBR Theory and BP Neural Network
700 Proceedngs of the 8th Internatonal Conference on Innovaton & Management Forecastng the Demand of Emergency Supples: Based on the CBR Theory and BP Neural Network Fu Deqang, Lu Yun, L Changbng School
More informationAn Alternative Way to Measure Private Equity Performance
An Alternatve Way to Measure Prvate Equty Performance Peter Todd Parlux Investment Technology LLC Summary Internal Rate of Return (IRR) s probably the most common way to measure the performance of prvate
More informationImproved SVM in Cloud Computing Information Mining
Internatonal Journal of Grd Dstrbuton Computng Vol.8, No.1 (015), pp.33-40 http://dx.do.org/10.1457/jgdc.015.8.1.04 Improved n Cloud Computng Informaton Mnng Lvshuhong (ZhengDe polytechnc college JangSu
More informationERP Software Selection Using The Rough Set And TPOSIS Methods
ERP Software Selecton Usng The Rough Set And TPOSIS Methods Under Fuzzy Envronment Informaton Management Department, Hunan Unversty of Fnance and Economcs, No. 139, Fengln 2nd Road, Changsha, 410205, Chna
More informationFuzzy TOPSIS Method in the Selection of Investment Boards by Incorporating Operational Risks
, July 6-8, 2011, London, U.K. Fuzzy TOPSIS Method n the Selecton of Investment Boards by Incorporatng Operatonal Rsks Elssa Nada Mad, and Abu Osman Md Tap Abstract Mult Crtera Decson Makng (MCDM) nvolves
More informationImproved Mining of Software Complexity Data on Evolutionary Filtered Training Sets
Improved Mnng of Software Complexty Data on Evolutonary Fltered Tranng Sets VILI PODGORELEC Insttute of Informatcs, FERI Unversty of Marbor Smetanova ulca 17, SI-2000 Marbor SLOVENIA vl.podgorelec@un-mb.s
More informationA Probabilistic Theory of Coherence
A Probablstc Theory of Coherence BRANDEN FITELSON. The Coherence Measure C Let E be a set of n propostons E,..., E n. We seek a probablstc measure C(E) of the degree of coherence of E. Intutvely, we want
More informationHow To Understand The Results Of The German Meris Cloud And Water Vapour Product
Ttel: Project: Doc. No.: MERIS level 3 cloud and water vapour products MAPP MAPP-ATBD-ClWVL3 Issue: 1 Revson: 0 Date: 9.12.1998 Functon Name Organsaton Sgnature Date Author: Bennartz FUB Preusker FUB Schüller
More informationTHE DISTRIBUTION OF LOAN PORTFOLIO VALUE * Oldrich Alfons Vasicek
HE DISRIBUION OF LOAN PORFOLIO VALUE * Oldrch Alfons Vascek he amount of captal necessary to support a portfolo of debt securtes depends on the probablty dstrbuton of the portfolo loss. Consder a portfolo
More informationMultiple-Period Attribution: Residuals and Compounding
Multple-Perod Attrbuton: Resduals and Compoundng Our revewer gave these authors full marks for dealng wth an ssue that performance measurers and vendors often regard as propretary nformaton. In 1994, Dens
More informationExtending Probabilistic Dynamic Epistemic Logic
Extendng Probablstc Dynamc Epstemc Logc Joshua Sack May 29, 2008 Probablty Space Defnton A probablty space s a tuple (S, A, µ), where 1 S s a set called the sample space. 2 A P(S) s a σ-algebra: a set
More informationPSYCHOLOGICAL RESEARCH (PYC 304-C) Lecture 12
14 The Ch-squared dstrbuton PSYCHOLOGICAL RESEARCH (PYC 304-C) Lecture 1 If a normal varable X, havng mean µ and varance σ, s standardsed, the new varable Z has a mean 0 and varance 1. When ths standardsed
More informationInstitute of Informatics, Faculty of Business and Management, Brno University of Technology,Czech Republic
Lagrange Multplers as Quanttatve Indcators n Economcs Ivan Mezník Insttute of Informatcs, Faculty of Busness and Management, Brno Unversty of TechnologCzech Republc Abstract The quanttatve role of Lagrange
More informationSingle and multiple stage classifiers implementing logistic discrimination
Sngle and multple stage classfers mplementng logstc dscrmnaton Hélo Radke Bttencourt 1 Dens Alter de Olvera Moraes 2 Vctor Haertel 2 1 Pontfíca Unversdade Católca do Ro Grande do Sul - PUCRS Av. Ipranga,
More informationFeature selection for intrusion detection. Slobodan Petrović NISlab, Gjøvik University College
Feature selecton for ntruson detecton Slobodan Petrovć NISlab, Gjøvk Unversty College Contents The feature selecton problem Intruson detecton Traffc features relevant for IDS The CFS measure The mrmr measure
More informationSearching for Interacting Features for Spam Filtering
Searchng for Interactng Features for Spam Flterng Chuanlang Chen 1, Yun-Chao Gong 2, Rongfang Be 1,, and X. Z. Gao 3 1 Department of Computer Scence, Bejng Normal Unversty, Bejng 100875, Chna 2 Software
More informationMining Feature Importance: Applying Evolutionary Algorithms within a Web-based Educational System
Mnng Feature Importance: Applyng Evolutonary Algorthms wthn a Web-based Educatonal System Behrouz MINAEI-BIDGOLI 1, and Gerd KORTEMEYER 2, and Wllam F. PUNCH 1 1 Genetc Algorthms Research and Applcatons
More informationStatistical Methods to Develop Rating Models
Statstcal Methods to Develop Ratng Models [Evelyn Hayden and Danel Porath, Österrechsche Natonalbank and Unversty of Appled Scences at Manz] Source: The Basel II Rsk Parameters Estmaton, Valdaton, and
More informationLecture 2: Single Layer Perceptrons Kevin Swingler
Lecture 2: Sngle Layer Perceptrons Kevn Sngler kms@cs.str.ac.uk Recap: McCulloch-Ptts Neuron Ths vastly smplfed model of real neurons s also knon as a Threshold Logc Unt: W 2 A Y 3 n W n. A set of synapses
More information8 Algorithm for Binary Searching in Trees
8 Algorthm for Bnary Searchng n Trees In ths secton we present our algorthm for bnary searchng n trees. A crucal observaton employed by the algorthm s that ths problem can be effcently solved when the
More informationRisk Model of Long-Term Production Scheduling in Open Pit Gold Mining
Rsk Model of Long-Term Producton Schedulng n Open Pt Gold Mnng R Halatchev 1 and P Lever 2 ABSTRACT Open pt gold mnng s an mportant sector of the Australan mnng ndustry. It uses large amounts of nvestments,
More informationAn Integrated Approach of AHP-GP and Visualization for Software Architecture Optimization: A case-study for selection of architecture style
Internatonal Journal of Scentfc & Engneerng Research Volume 2, Issue 7, July-20 An Integrated Approach of AHP-GP and Vsualzaton for Software Archtecture Optmzaton: A case-study for selecton of archtecture
More informationVision Mouse. Saurabh Sarkar a* University of Cincinnati, Cincinnati, USA ABSTRACT 1. INTRODUCTION
Vson Mouse Saurabh Sarkar a* a Unversty of Cncnnat, Cncnnat, USA ABSTRACT The report dscusses a vson based approach towards trackng of eyes and fngers. The report descrbes the process of locatng the possble
More informationLogistic Regression. Steve Kroon
Logstc Regresson Steve Kroon Course notes sectons: 24.3-24.4 Dsclamer: these notes do not explctly ndcate whether values are vectors or scalars, but expects the reader to dscern ths from the context. Scenaro
More information1. Fundamentals of probability theory 2. Emergence of communication traffic 3. Stochastic & Markovian Processes (SP & MP)
6.3 / -- Communcaton Networks II (Görg) SS20 -- www.comnets.un-bremen.de Communcaton Networks II Contents. Fundamentals of probablty theory 2. Emergence of communcaton traffc 3. Stochastc & Markovan Processes
More informationCS 2750 Machine Learning. Lecture 3. Density estimation. CS 2750 Machine Learning. Announcements
Lecture 3 Densty estmaton Mlos Hauskrecht mlos@cs.ptt.edu 5329 Sennott Square Next lecture: Matlab tutoral Announcements Rules for attendng the class: Regstered for credt Regstered for audt (only f there
More informationA hybrid global optimization algorithm based on parallel chaos optimization and outlook algorithm
Avalable onlne www.ocpr.com Journal of Chemcal and Pharmaceutcal Research, 2014, 6(7):1884-1889 Research Artcle ISSN : 0975-7384 CODEN(USA) : JCPRC5 A hybrd global optmzaton algorthm based on parallel
More informationEfficient Project Portfolio as a tool for Enterprise Risk Management
Effcent Proect Portfolo as a tool for Enterprse Rsk Management Valentn O. Nkonov Ural State Techncal Unversty Growth Traectory Consultng Company January 5, 27 Effcent Proect Portfolo as a tool for Enterprse
More informationEnterprise Master Patient Index
Enterprse Master Patent Index Healthcare data are captured n many dfferent settngs such as hosptals, clncs, labs, and physcan offces. Accordng to a report by the CDC, patents n the Unted States made an
More informationSet. algorithms based. 1. Introduction. System Diagram. based. Exploration. 2. Index
ISSN (Prnt): 1694-0784 ISSN (Onlne): 1694-0814 www.ijcsi.org 236 IT outsourcng servce provder dynamc evaluaton model and algorthms based on Rough Set L Sh Sh 1,2 1 Internatonal School of Software, Wuhan
More informationForecasting the Direction and Strength of Stock Market Movement
Forecastng the Drecton and Strength of Stock Market Movement Jngwe Chen Mng Chen Nan Ye cjngwe@stanford.edu mchen5@stanford.edu nanye@stanford.edu Abstract - Stock market s one of the most complcated systems
More informationLatent Class Regression. Statistics for Psychosocial Research II: Structural Models December 4 and 6, 2006
Latent Class Regresson Statstcs for Psychosocal Research II: Structural Models December 4 and 6, 2006 Latent Class Regresson (LCR) What s t and when do we use t? Recall the standard latent class model
More informationSelection and Classification of Statistical Data Using Fuzzy Logic
Selecton and Classfcaton of Statstcal Data Usng Fuzzy Logc Mroslav Hudec (1), Mrko Vujoševć (2) (1) INFOSTAT Insttute of Informatcs and Statstcs, Bratslava, Slovaka (2) Faculty of Organzatonal Scences,
More informationFREQUENCY OF OCCURRENCE OF CERTAIN CHEMICAL CLASSES OF GSR FROM VARIOUS AMMUNITION TYPES
FREQUENCY OF OCCURRENCE OF CERTAIN CHEMICAL CLASSES OF GSR FROM VARIOUS AMMUNITION TYPES Zuzanna BRO EK-MUCHA, Grzegorz ZADORA, 2 Insttute of Forensc Research, Cracow, Poland 2 Faculty of Chemstry, Jagellonan
More informationHow Sets of Coherent Probabilities May Serve as Models for Degrees of Incoherence
1 st Internatonal Symposum on Imprecse Probabltes and Ther Applcatons, Ghent, Belgum, 29 June 2 July 1999 How Sets of Coherent Probabltes May Serve as Models for Degrees of Incoherence Mar J. Schervsh
More informationA Novel Methodology of Working Capital Management for Large. Public Constructions by Using Fuzzy S-curve Regression
Novel Methodology of Workng Captal Management for Large Publc Constructons by Usng Fuzzy S-curve Regresson Cheng-Wu Chen, Morrs H. L. Wang and Tng-Ya Hseh Department of Cvl Engneerng, Natonal Central Unversty,
More informationCan Auto Liability Insurance Purchases Signal Risk Attitude?
Internatonal Journal of Busness and Economcs, 2011, Vol. 10, No. 2, 159-164 Can Auto Lablty Insurance Purchases Sgnal Rsk Atttude? Chu-Shu L Department of Internatonal Busness, Asa Unversty, Tawan Sheng-Chang
More informationImplementation of Deutsch's Algorithm Using Mathcad
Implementaton of Deutsch's Algorthm Usng Mathcad Frank Roux The followng s a Mathcad mplementaton of Davd Deutsch's quantum computer prototype as presented on pages - n "Machnes, Logc and Quantum Physcs"
More informationResearch on Evaluation of Customer Experience of B2C Ecommerce Logistics Enterprises
3rd Internatonal Conference on Educaton, Management, Arts, Economcs and Socal Scence (ICEMAESS 2015) Research on Evaluaton of Customer Experence of B2C Ecommerce Logstcs Enterprses Yle Pe1, a, Wanxn Xue1,
More informationBERNSTEIN POLYNOMIALS
On-Lne Geometrc Modelng Notes BERNSTEIN POLYNOMIALS Kenneth I. Joy Vsualzaton and Graphcs Research Group Department of Computer Scence Unversty of Calforna, Davs Overvew Polynomals are ncredbly useful
More informationRecurrence. 1 Definitions and main statements
Recurrence 1 Defntons and man statements Let X n, n = 0, 1, 2,... be a MC wth the state space S = (1, 2,...), transton probabltes p j = P {X n+1 = j X n = }, and the transton matrx P = (p j ),j S def.
More informationPlanning for Marketing Campaigns
Plannng for Marketng Campagns Qang Yang and Hong Cheng Department of Computer Scence Hong Kong Unversty of Scence and Technology Clearwater Bay, Kowloon, Hong Kong, Chna (qyang, csch)@cs.ust.hk Abstract
More informationSTATISTICAL DATA ANALYSIS IN EXCEL
Mcroarray Center STATISTICAL DATA ANALYSIS IN EXCEL Lecture 6 Some Advanced Topcs Dr. Petr Nazarov 14-01-013 petr.nazarov@crp-sante.lu Statstcal data analyss n Ecel. 6. Some advanced topcs Correcton for
More informationThe Application of Fractional Brownian Motion in Option Pricing
Vol. 0, No. (05), pp. 73-8 http://dx.do.org/0.457/jmue.05.0..6 The Applcaton of Fractonal Brownan Moton n Opton Prcng Qng-xn Zhou School of Basc Scence,arbn Unversty of Commerce,arbn zhouqngxn98@6.com
More informationA DATA MINING APPLICATION IN A STUDENT DATABASE
JOURNAL OF AERONAUTICS AND SPACE TECHNOLOGIES JULY 005 VOLUME NUMBER (53-57) A DATA MINING APPLICATION IN A STUDENT DATABASE Şenol Zafer ERDOĞAN Maltepe Ünversty Faculty of Engneerng Büyükbakkalköy-Istanbul
More informationDEFINING %COMPLETE IN MICROSOFT PROJECT
CelersSystems DEFINING %COMPLETE IN MICROSOFT PROJECT PREPARED BY James E Aksel, PMP, PMI-SP, MVP For Addtonal Informaton about Earned Value Management Systems and reportng, please contact: CelersSystems,
More informationLogistic Regression. Lecture 4: More classifiers and classes. Logistic regression. Adaboost. Optimization. Multiple class classification
Lecture 4: More classfers and classes C4B Machne Learnng Hlary 20 A. Zsserman Logstc regresson Loss functons revsted Adaboost Loss functons revsted Optmzaton Multple class classfcaton Logstc Regresson
More informationReporting Forms ARF 113.0A, ARF 113.0B, ARF 113.0C and ARF 113.0D FIRB Corporate (including SME Corporate), Sovereign and Bank Instruction Guide
Reportng Forms ARF 113.0A, ARF 113.0B, ARF 113.0C and ARF 113.0D FIRB Corporate (ncludng SME Corporate), Soveregn and Bank Instructon Gude Ths nstructon gude s desgned to assst n the completon of the FIRB
More informationInvestment Portfolio Evaluation by the Fuzzy Approach
Investment Portfolo Evaluaton by the Fuzzy Approach Lambovska Maya, Marchev Angel Abstract Ths paper presents a new fuzzy approach for the evaluaton of nvestment portfolo, where the approach s vewed by
More informationAbstract. Clustering ensembles have emerged as a powerful method for improving both the
Clusterng Ensembles: {topchyal, Models jan, of punch}@cse.msu.edu Consensus and Weak Parttons * Alexander Topchy, Anl K. Jan, and Wllam Punch Department of Computer Scence and Engneerng, Mchgan State Unversty
More informationDetecting Credit Card Fraud using Periodic Features
Detectng Credt Card Fraud usng Perodc Features Alejandro Correa Bahnsen, Djamla Aouada, Aleksandar Stojanovc and Björn Ottersten Interdscplnary Centre for Securty, Relablty and Trust Unversty of Luxembourg,
More informationStudy on Model of Risks Assessment of Standard Operation in Rural Power Network
Study on Model of Rsks Assessment of Standard Operaton n Rural Power Network Qngj L 1, Tao Yang 2 1 Qngj L, College of Informaton and Electrcal Engneerng, Shenyang Agrculture Unversty, Shenyang 110866,
More informationCourse outline. Financial Time Series Analysis. Overview. Data analysis. Predictive signal. Trading strategy
Fnancal Tme Seres Analyss Patrck McSharry patrck@mcsharry.net www.mcsharry.net Trnty Term 2014 Mathematcal Insttute Unversty of Oxford Course outlne 1. Data analyss, probablty, correlatons, vsualsaton
More informationHow To Know The Components Of Mean Squared Error Of Herarchcal Estmator S
S C H E D A E I N F O R M A T I C A E VOLUME 0 0 On Mean Squared Error of Herarchcal Estmator Stans law Brodowsk Faculty of Physcs, Astronomy, and Appled Computer Scence, Jagellonan Unversty, Reymonta
More informationSection 5.4 Annuities, Present Value, and Amortization
Secton 5.4 Annutes, Present Value, and Amortzaton Present Value In Secton 5.2, we saw that the present value of A dollars at nterest rate per perod for n perods s the amount that must be deposted today
More informationAssessing Student Learning Through Keyword Density Analysis of Online Class Messages
Assessng Student Learnng Through Keyword Densty Analyss of Onlne Class Messages Xn Chen New Jersey Insttute of Technology xc7@njt.edu Brook Wu New Jersey Insttute of Technology wu@njt.edu ABSTRACT Ths
More informationBrigid Mullany, Ph.D University of North Carolina, Charlotte
Evaluaton And Comparson Of The Dfferent Standards Used To Defne The Postonal Accuracy And Repeatablty Of Numercally Controlled Machnng Center Axes Brgd Mullany, Ph.D Unversty of North Carolna, Charlotte
More informationPRACTICE 1: MUTUAL FUNDS EVALUATION USING MATLAB.
PRACTICE 1: MUTUAL FUNDS EVALUATION USING MATLAB. INDEX 1. Load data usng the Edtor wndow and m-fle 2. Learnng to save results from the Edtor wndow. 3. Computng the Sharpe Rato 4. Obtanng the Treynor Rato
More informationCalculating the high frequency transmission line parameters of power cables
< ' Calculatng the hgh frequency transmsson lne parameters of power cables Authors: Dr. John Dcknson, Laboratory Servces Manager, N 0 RW E B Communcatons Mr. Peter J. Ncholson, Project Assgnment Manager,
More informationUsing Series to Analyze Financial Situations: Present Value
2.8 Usng Seres to Analyze Fnancal Stuatons: Present Value In the prevous secton, you learned how to calculate the amount, or future value, of an ordnary smple annuty. The amount s the sum of the accumulated
More informationIntelligent stock trading system by turning point confirming and probabilistic reasoning
Expert Systems wth Applcatons Expert Systems wth Applcatons 34 (2008) 620 627 www.elsever.com/locate/eswa Intellgent stock tradng system by turnng pont confrmng and probablstc reasonng Depe Bao *, Zehong
More informationAn Interest-Oriented Network Evolution Mechanism for Online Communities
An Interest-Orented Network Evoluton Mechansm for Onlne Communtes Cahong Sun and Xaopng Yang School of Informaton, Renmn Unversty of Chna, Bejng 100872, P.R. Chna {chsun,yang}@ruc.edu.cn Abstract. Onlne
More informationKeywords : classifier, Association rules, data mining, healthcare, Associative Classifiers, CBA, CMAR, CPAR, MCAR. GJCST Classification : H.2.
Global Journal of Computer Scence and Technology Volume 11 Issue 22 Verson 1.0 Type: Double Blnd Peer Revewed Internatonal Research Journal Publsher: Global Journals Inc. (USA) Onlne ISSN: 0975-4172 &
More informationFault tolerance in cloud technologies presented as a service
Internatonal Scentfc Conference Computer Scence 2015 Pavel Dzhunev, PhD student Fault tolerance n cloud technologes presented as a servce INTRODUCTION Improvements n technques for vrtualzaton and performance
More informationL10: Linear discriminants analysis
L0: Lnear dscrmnants analyss Lnear dscrmnant analyss, two classes Lnear dscrmnant analyss, C classes LDA vs. PCA Lmtatons of LDA Varants of LDA Other dmensonalty reducton methods CSCE 666 Pattern Analyss
More informationSketching Sampled Data Streams
Sketchng Sampled Data Streams Florn Rusu, Aln Dobra CISE Department Unversty of Florda Ganesvlle, FL, USA frusu@cse.ufl.edu adobra@cse.ufl.edu Abstract Samplng s used as a unversal method to reduce the
More informationData Visualization by Pairwise Distortion Minimization
Communcatons n Statstcs, Theory and Methods 34 (6), 005 Data Vsualzaton by Parwse Dstorton Mnmzaton By Marc Sobel, and Longn Jan Lateck* Department of Statstcs and Department of Computer and Informaton
More informationCalculation of Sampling Weights
Perre Foy Statstcs Canada 4 Calculaton of Samplng Weghts 4.1 OVERVIEW The basc sample desgn used n TIMSS Populatons 1 and 2 was a two-stage stratfed cluster desgn. 1 The frst stage conssted of a sample
More informationMethodology to Determine Relationships between Performance Factors in Hadoop Cloud Computing Applications
Methodology to Determne Relatonshps between Performance Factors n Hadoop Cloud Computng Applcatons Lus Eduardo Bautsta Vllalpando 1,2, Alan Aprl 1 and Alan Abran 1 1 Department of Software Engneerng and
More informationPrediction Model for Characteristics of Implementation of Information Systems in Small and Medium Enterprises
Predcton Model for Characterstcs of Implementaton of Informaton Systems n Small and Medum Enterprses I. Nazor, K. Fertalj, and D. Kalpc Abstract The process of choosng an Enterprse Resource Plannng (ERP)
More informationConversion between the vector and raster data structures using Fuzzy Geographical Entities
Converson between the vector and raster data structures usng Fuzzy Geographcal Enttes Cdála Fonte Department of Mathematcs Faculty of Scences and Technology Unversty of Combra, Apartado 38, 3 454 Combra,
More informationBUSINESS PROCESS PERFORMANCE MANAGEMENT USING BAYESIAN BELIEF NETWORK. 0688, dskim@ssu.ac.kr
Proceedngs of the 41st Internatonal Conference on Computers & Industral Engneerng BUSINESS PROCESS PERFORMANCE MANAGEMENT USING BAYESIAN BELIEF NETWORK Yeong-bn Mn 1, Yongwoo Shn 2, Km Jeehong 1, Dongsoo
More informationLearning to Classify Ordinal Data: The Data Replication Method
Journal of Machne Learnng Research 8 (7) 393-49 Submtted /6; Revsed 9/6; Publshed 7/7 Learnng to Classfy Ordnal Data: The Data Replcaton Method Jame S. Cardoso INESC Porto, Faculdade de Engenhara, Unversdade
More informationA PROBABILITY-MAPPING ALGORITHM FOR CALIBRATING THE POSTERIOR PROBABILITIES: A DIRECT MARKETING APPLICATION
Document de traval du LEM 2011-06 A PROBABILITY-MAPPIG ALGORITHM FOR CALIBRATIG THE POSTERIOR PROBABILITIES: A DIRECT MARKETIG APPLICATIO Krstof Coussement *, Wouter Bucknx ** * IESEG School of Management
More informationIDENTIFICATION AND CORRECTION OF A COMMON ERROR IN GENERAL ANNUITY CALCULATIONS
IDENTIFICATION AND CORRECTION OF A COMMON ERROR IN GENERAL ANNUITY CALCULATIONS Chrs Deeley* Last revsed: September 22, 200 * Chrs Deeley s a Senor Lecturer n the School of Accountng, Charles Sturt Unversty,
More informationDraft. Evaluation of project and portfolio Management Information Systems with the use of a hybrid IFS-TOPSIS method
Intellgent Decson Technologes 7 (2013) 91 105 91 DOI 10.3233/IDT-120153 IOS Press Evaluaton of project and portfolo Management Informaton Systems wth the use of a hybrd IFS-TOPSIS method Vassls C. Geroganns
More informationMETHODOLOGY TO DETERMINE RELATIONSHIPS BETWEEN PERFORMANCE FACTORS IN HADOOP CLOUD COMPUTING APPLICATIONS
METHODOLOGY TO DETERMINE RELATIONSHIPS BETWEEN PERFORMANCE FACTORS IN HADOOP CLOUD COMPUTING APPLICATIONS Lus Eduardo Bautsta Vllalpando 1,2, Alan Aprl 1 and Alan Abran 1 1 Department of Software Engneerng
More informationComplex Service Provisioning in Collaborative Cloud Markets
Melane Sebenhaar, Ulrch Lampe, Tm Lehrg, Sebastan Zöller, Stefan Schulte, Ralf Stenmetz: Complex Servce Provsonng n Collaboratve Cloud Markets. In: W. Abramowcz et al. (Eds.): Proceedngs of the 4th European
More informationLuby s Alg. for Maximal Independent Sets using Pairwise Independence
Lecture Notes for Randomzed Algorthms Luby s Alg. for Maxmal Independent Sets usng Parwse Independence Last Updated by Erc Vgoda on February, 006 8. Maxmal Independent Sets For a graph G = (V, E), an ndependent
More informationHow To Calculate The Accountng Perod Of Nequalty
Inequalty and The Accountng Perod Quentn Wodon and Shlomo Ytzha World Ban and Hebrew Unversty September Abstract Income nequalty typcally declnes wth the length of tme taen nto account for measurement.
More informationMATHEMATICAL ENGINEERING TECHNICAL REPORTS. Sequential Optimizing Investing Strategy with Neural Networks
MATHEMATICAL ENGINEERING TECHNICAL REPORTS Sequental Optmzng Investng Strategy wth Neural Networks Ryo ADACHI and Akmch TAKEMURA METR 2010 03 February 2010 DEPARTMENT OF MATHEMATICAL INFORMATICS GRADUATE
More informationPerformance attribution for multi-layered investment decisions
Performance attrbuton for mult-layered nvestment decsons 880 Thrd Avenue 7th Floor Ne Yor, NY 10022 212.866.9200 t 212.866.9201 f qsnvestors.com Inna Oounova Head of Strategc Asset Allocaton Portfolo Management
More informationTopic Identification based on Bayesian Belief Networks in the context of an Air Traffic Control Task
Procesamento del Lenguaje Natural, núm. 35 (2005), pp. 327-332 recbdo 06-05-2005; aceptado 01-06-2005 Topc Identfcaton based on Bayesan Belef Networs n the context of an Ar Traffc Control Tas F. Fernández,
More informationOn-Line Fault Detection in Wind Turbine Transmission System using Adaptive Filter and Robust Statistical Features
On-Lne Fault Detecton n Wnd Turbne Transmsson System usng Adaptve Flter and Robust Statstcal Features Ruoyu L Remote Dagnostcs Center SKF USA Inc. 3443 N. Sam Houston Pkwy., Houston TX 77086 Emal: ruoyu.l@skf.com
More informationAnts Can Schedule Software Projects
Ants Can Schedule Software Proects Broderck Crawford 1,2, Rcardo Soto 1,3, Frankln Johnson 4, and Erc Monfroy 5 1 Pontfca Unversdad Católca de Valparaíso, Chle FrstName.Name@ucv.cl 2 Unversdad Fns Terrae,
More informationUsing Supervised Clustering Technique to Classify Received Messages in 137 Call Center of Tehran City Council
Usng Supervsed Clusterng Technque to Classfy Receved Messages n 137 Call Center of Tehran Cty Councl Mahdyeh Haghr 1*, Hamd Hassanpour 2 (1) Informaton Technology engneerng/e-commerce, Shraz Unversty (2)
More informationA Hybrid Model for Forecasting Sales in Turkish Paint Industry
Internatonal Journal of Computatonal Intellgence Systems, Vol.2, No. 3 (October, 2009), 277-287 A Hybrd Model for Forecastng Sales n Turksh Pant Industry Alp Ustundag * Department of Industral Engneerng,
More informationBusiness Process Improvement using Multi-objective Optimisation K. Vergidis 1, A. Tiwari 1 and B. Majeed 2
Busness Process Improvement usng Mult-objectve Optmsaton K. Vergds 1, A. Twar 1 and B. Majeed 2 1 Manufacturng Department, School of Industral and Manufacturng Scence, Cranfeld Unversty, Cranfeld, MK43
More information1 Example 1: Axis-aligned rectangles
COS 511: Theoretcal Machne Learnng Lecturer: Rob Schapre Lecture # 6 Scrbe: Aaron Schld February 21, 2013 Last class, we dscussed an analogue for Occam s Razor for nfnte hypothess spaces that, n conjuncton
More informationFast Fuzzy Clustering of Web Page Collections
Fast Fuzzy Clusterng of Web Page Collectons Chrstan Borgelt and Andreas Nürnberger Dept. of Knowledge Processng and Language Engneerng Otto-von-Guercke-Unversty of Magdeburg Unverstätsplatz, D-396 Magdeburg,
More informationRing structure of splines on triangulations
www.oeaw.ac.at Rng structure of splnes on trangulatons N. Vllamzar RICAM-Report 2014-48 www.rcam.oeaw.ac.at RING STRUCTURE OF SPLINES ON TRIANGULATIONS NELLY VILLAMIZAR Introducton For a trangulated regon
More information