Combining Statistics and Semantics via Ensemble Model for Document Clustering

Size: px
Start display at page:

Download "Combining Statistics and Semantics via Ensemble Model for Document Clustering"

Transcription

1 ombining tatitic and emantic via Enemble Model or Document lutering amah Jamal Fodeh Michigan tate Univerity Eat Laning, MI, William F Punch Michigan tate Univerity Eat Laning, MI, punch@mu.edu Pang-Ning an Michigan tate Univerity Eat Laning, MI, ptan@mu.edu ABRA Incorporating background knowledge into data mining algorithm i an important but challenging problem. urrent approache in emi-upervied learning require explicit knowledge provided by domain expert, knowledge peciic to the particular data et. In thi tudy, we propoe an enemble model that couple two ource o inormation: tatitic inormation that i derived rom the data et, and ene inormation retrieved rom WordNet that i ued to build a emantic binary model. We evaluated the eicacy o uing our combined enemble model on the Reuter and 20newgroup data et. Keyword WordNet, enemble learning, text clutering, diambiguation. 1. INRODUION he rapidly growing availability o large tract o textual data uch a online new eed, blog poting, , and dicuion board meage, ha made the need or improved text clutering an important current reearch area. However, depite the extenive reearch, clutering untructured, textual inormation remain a challenging problem. For example, the nature o the untructured textual inormation make it hard or current clutering algorithm to capture the intrinic tructure that we deire [3]. Individual data et alo have unique characteritic which add more complexity to mapping or deciding upon the clutering methodology that work bet or a particular data et. Moreover, the lack o labeled example in unupervied clutering make the partitioning tak an ill-poed problem ince there i no adopted methodology wellknown to produce the ideal clutering [3]. o overcome thee challenge, reearcher have begun to invetigate alternative clutering approache that incorporate background knowledge to guide each partitioning tak and thu alleviate the diiculty o inding a ingle, bet approach [2][6]. One way to add background knowledge i through emiupervied clutering [1], where the domain inormation i Permiion to make digital or hard copie o all or part o thi work or peronal or claroom ue i granted without ee provided that copie are not made or ditributed or proit or commercial advantage and that copie bear thi notice and the ull citation on the irt page. o copy otherwie, or republih, to pot on erver or to reditribute to lit, require prior peciic permiion and/or a ee. A 09, March 8-12, 2009, Honolulu, Hawaii, U..A. opyright 2009 AM /09/03 $5.00. provided in the orm o labeled example or mut-link (ML) and cannot-link (L) contraint. hi explicit inormation ugget the availability o an expert in the domain who would annotate the label o the document or ummarize the important aociation between document in the data et. In practice, thi human intervention can be expenive and could produce inaccurate reult depending on the reliability o the inormation provided. o addre thee problem, there ha been ome recent work that attempt to incorporate background knowledge into the partitioning tak without any uer or expert dynamic interaction [8]. Unlike emi-upervied clutering, thi kind o background knowledge provide general inormation about the relationhip between the eature and i applicable to any data et with imilar type o eature. For document clutering, recent interet ha ocued on incorporating contextual knowledge in the orm o linguitic ontologie into clutering algorithm uch a WordNet [2][6][8] WordNet i an ontology that include not only the ene o the word, but alo their relationhip with each other. WordNet addree the ynonymy and polyemy problem o text document by replacing the word by their mot appropriate ene a ued in the context o the document. In thi paper, we invetigate the eectivene o combining term tatitic with emantic knowledge acquired rom WordNet to improve document clutering. Our analyi how that a traightorward replacement o the word by their correponding ene rom WordNet may not alway improve the clutering reult. hi i becaue the clutering algorithm mut deal with iue uch a the increaing dimenionality o the data (when the noun are replaced by their correponding ene) and noie (when incorrect ene are elected a eature). We propoe to addre thee iue uing a compound enemble clutering algorithm that combine the tatitic inormation rom the data with the ene inormation rom WordNet. Our approach or combining the model take into account the conitency o each clutering olution and i applicable to any clutering algorithm (including k-mean). We evaluated the eectivene our method uing two benchmark dataet: Reuter and 20newgroup. Our experimental reult ugget that the propoed method help to improve the clutering reult igniicantly when applied to data et where ene inormation i valuable to diambiguate word that are ued in multiple context. 2. emantic imilarity uing WordNet WordNet i a hierarchical linguitic ontology that group word into ynet. Each ynet deine a certain concept. ynet are

2 linked by emantic relation uch a hypernym and hyponym, indicating cla-ubcla relationhip between the ynet. everal emantic ditance meaure can be ued to compute the emantic imilarity between two ynet in the Wordnet hierarchy. In particular, we ued the Wu-Palmer meaure, which utilize inormation uch a ene depth. Wu-Palmer compute the imilarity between two ene by inding the leat common ubumer (L) node that connect their ene. he L o two ene, p and q, i the lowet interecting node between the path o p and q rom the root o the WordNet cla-ubcla hierarchy. Once the L ha been identiied, the Wu-Palmer ditance i given by the ollowing equation: 2d Wu-Palmer( p, q ) (1) L L 2d p where d i the depth o the L rom the root, L p i the path length between p and the L, and L q i the path length between q and the L. A key tep in our propoed approach or incorporating emantic knowledge rom WordNet i to identiy the mot appropriate ene aociated with each noun in a given document. Our approach i baed on the aumption that the ene o a noun i determined by the context in which it i being ued in the document. For example, conider the word cat, which ha eight meaning a a noun in WordNet. I it i ued in a document that contain other word uch a kitten, Perian, and pet, we expect it ene reer to the eline mammal ene o cat, and not one o the other even (uch a a arm machine or a particular type o X-ray). However, i the word cat appear in a document that contain other noun uch a contruction and builder, it ene mot likely reer to a aterpillar, which i a large tracked vehicle ued or moving earth in contruction. Let i = { i1, i2, ik } denote the et o all ene aociated with the noun t i according to the WordNet ontology. Given a document d, we determine the mot appropriate ene o a noun t i i by computing the um o it imilarity to other noun ene in d, i.e.: i argmax max ( il, jm) il i jm j t j d where ( p, q ) i the WordNet imilarity between two ene, p and q. Furthermore, ince the ene o a given noun in the WordNet hierarchy are arranged in decending order according to their popularity, we retrict our conideration to the top 3 ene o each given noun. Once the appropriate ene have been choen or the noun, we tranorm each document vector o term into a binary vector o their correponding ene. 3. ENEMBLE LUERING he propoed enemble clutering ramework combine the clutering olution obtained rom the emantic imilarity o the document (emantic binary model) with thoe obtained baed on requency imilaritie (noun requency model). Our rationale or uing enemble clutering i that, although individual clutering olution (uing either requency or emantic imilarity) may make poor deciion regarding the cluter aignment or ome document, one may be able to improve clutering reult by q (2) conidering their collective deciion 1. he propoed ramework i alo highly lexible becaue it may accommodate any baeline clutering algorithm a well a methodology or creating dierent intance o the enemble. he enemble clutering ramework ue two type o data input: (1) a document-noun requency matrix A, and (2) a documentene binary matrix A. A et o requency-baed clutering olution, M Fenemble, i generated rom A uing the methodology preented in ection 3.1. Analogouly, a ene o ene-baed clutering olution, M enemble, i obtained by applying the methodology given in ection 3.2. he clutering olution rom M Fenemble and M enemble are then aggregated to obtain the inal clutering uing the approach decribed in ection Noun Frequency Model he noun requency model i generated rom applying enemble clutering to the document-noun requency matrix A. Firt, we randomly ample a ubet o the noun, V V. o alleviate bia in the ample ize V, the number o noun to be ampled i an integer choen randomly between m/2 and m-1, where m i the total number o noun in the original dataet. A truncated document-noun requency matrix, A i then created by chooing only the noun in each document that belong to the ubet V. Once the truncated matrix i obtained, the weight or each document vector are urther normalized uing the FIDF method. Finally we apply the tandard k-mean algorithm to obtain an N k requency-baed cluter memberhip matrix,, whoe (i,j) th element i equal to 1 i the document d i belong to cluter j and 0 otherwie. 3.2 emantic Binary Model Our approach or creating the emantic binary model i quite imilar to the approach decribed in the previou ection. However, intead o uing the document-noun requency matrix, we ue the document-ene matrix a input to the enemble clutering algorithm. he ample ize i a random integer choen between m/2 and m-1. A truncated document-ene binary matrix A i then created by removing all the ene not elected by the ample. Ater normalization uing the FIDF method, we apply the k-mean clutering algorithm to obtain the ene-baed cluter memberhip matrix,. 3.3 ombined Enemble lutering hi ection decribe our propoed approach or combining the noun requency model with the emantic binary model. Firt, an N N weighted co-aociation matrix i computed rom the et o requency-baed cluter memberhip matrice in M Fenemble. he co-aociation matrix repreent the number o time a pair o document i aigned to the ame cluter in the enemble, weighted by the quality o the individual clutering olution. Formally it i iteratively computed rom the requency-baed cluter memberhip matrice a ollow: ( t1) ( t ) wt where the matrix product t) (3) ( i a binary 0/1 matrix that indicate whether a pair o document belong to the ame cluter 1 Auming each clutering olution i independent and i doing better than random cluter aignment.

3 during the t th iteration o the enemble and the weighting actor w t meaure the quality o the clutering. he matrix product i alo known a an incidence matrix in clutering literature. he weighting actor w t i then computed by correlating the coine imilarity matrix between each pair o document with the incidence matrix. he higher the correlation i, the greater the level o agreement between the clutering reult and the document imilarity matrix. Equation (3) i iteratively updated uing all the clutering olution rom the noun requency model. he weighted co-aociation matrix eectively encode the likelihood that a pair o document i in the ame cluter baed on it term requency inormation. It i certainly poible to apply a clutering algorithm uch a k-mean on to produce a inal clutering or the noun requency model. imilarly, we may repeat thi procedure to obtain a ene-baed weighted co-aociation matrix : ( t1) where the incidence matrix w t (4) depend on the clutering olution o the emantic binary model and the weighting actor w t i the correlation coeicient between the coine imilarity o document (computed rom the document-ene matrix A ) and the incidence matrix. he overall emantic binary model can be obtained by applying the kmean algorithm to the weighted coaociation matrix. However, ince our intention i to combine the noun requency model with emantic binary model, we may aggregate their weighted co-aociation matrice a ollow: ( 1 ) (5) ombined where α i a parameter that govern the tradeo between uing both model. We will apply a clutering algorithm uch a kmean on the combined weighted co-aociation matrix to obtain the inal clutering reult. 4. EXPERIMEN electing a data et to tet our approach upon ha to be done careully. We wanted to work with a data et that reveal the power o ubtituting the noun by it ene. o teting with a dataet that i rich with vocabulary ued to expre the ame meaning, or vocabulary that ha multiple meaning would give our algorithm a better chance to demontrate the eect o incorporating the emantic knowledge rom WordNet. wo benchmark data et were elected or our experiment the Reuter and 20newgroup data et. For Reuter dataet we elected only document that belong to the top 20 larget categorie or our experiment. We then ampled at mot 200 document rom each category. he inal ize o our text corpu wa 2655 document. For 20newgroup dataet: we aggregated the original training et into 6 ditinct categorie. For example, we aggregate categorie uch a rec.auto, rec.motorcycle, rec.port.baeball, and rec.port.hockey into one cla. he original data wa divided into 60% training et and 40% tet et. For our experiment, we ampled 6000 article rom the training et and applied our algorithm to detect the 6 ditinct categorie. All the experiment conducted in thi tudy employ kmean a the baeline algorithm. he number o cluter i equal to the number o categorie in the original data. ince kmean i enitive to the choice o initial centroid, we repeat each experiment 50 time and report their average entropy or purity. We compared our method againt Latent emantic Indexing (LI) which i a tatitical method or identiying the latent tructure in a et o document by analyzing the relationhip between the document and their correponding term. able 1: Intra-Document lutering or Reuter Data et. Document call rate, loan creen luter degree, monetary_value department, militia corn, grain, barley, wheat, oat liberation,agribuine average, iodine national, ubtitute,orghum hundredweight,two, three, our,ix tallow tallion, anadian authority, agreement purchae, bargain department, point, Hondura tranhipment, agribuine corn, wheat port, creen, hip meaure, metric_ton board etimate, intervention, beginning metric_ton, let, French eaon barley, cereal, wheat, corn prognoi, manner_o_peaking 4.1 Intra Document ene Diambiguation hi experiment aim to demontrate the eectivene o our method in electing the mot appropriate ene to ubtitute or a noun in the context o a particular document. One way to do thi i to cluter the ene and examine the reulting cluter. I we oberve ditinctive topic in the cluter then thi indicate that the elected ene were able to reveal ome o the emantic content o the document. In order to apply clutering within the document, we built a imilarity matrix or the ene uing Wu- Palmer imilarity meaure [10]. Intra-document clutering i then perormed uing the complete-link agglomerative hierarchical clutering algorithm. able 1 how the reult o clutering three document that belong to the grain category in the Reuter dataet. Document 2 how well-related emantic cluter uch a (purchae, bargain), (corn, wheat), and (meaure, metric_ton). hee cluter repreent the dierent emantic concept that are important in the document. Neverthele, we do oberve ome impure cluter uing the ene that were choen. For example, in document 1, although we ound a cloely-related emantic cluter (corn, grain, barley, wheat, oat), we alo mied a ene (orghum) that hould be added to thi cluter. Intead, orghum wa added to a cluter that doe not have a high emantic imilarity with gain.

4 hee reult ugget that our technique or ixing the ene o a word uing WordNet wa quite ucceul to orm a group o emantically related ene. We thereore argue that thi method wa able -to a good extent- to eparate the dierent topic in the document which hould improve document clutering reult. 4.2 Document lutering uing ene Ater ixing the ene o each word uing the approach decribed in ection 2, we tranorm each document vector into a binary vector o their ene. For the Reuter dataet, we tarted with 5922 noun a eature. Ater applying WordNet, the data i tranormed into a document-ene matrix with 6559 ene. Likewie, in the 20newgroup dataet, our method convert the noun into ene. able 2 how the number o eature ued or the dierent approache (including LI). Dataet able 2. Number o eature ued Noun-requency emantic model binary model LI 20newgroup Reuter he number o eature ha increaed ater word ene diambiguation ince the algorithm may reolve the ame noun into multiple ene depending on the context o the document. Next, we apply k-mean clutering on both the document-noun requency matrix A and document-ene binary matrix A. able 3 how the reult o the clutering, which are baed on the average entropy or 50 iteration o applying kmean with dierent initial centroid. hee reult however do not eem to jutiy the need to replace the noun with their correponding ene. In act, the ue o ene inormation eem to degrade the entropy igniicantly or the Reuter data et. able 3. omparion o average entropy or K-mean Dataet Noun-requency emantic model binary model 20newgroup Reuter One poible explanation or the poor reult i due to increaing number o eature when we replace the noun by their ene. o invetigate the eect o the number o dimenion, we have perormed ingular Value Decompoition (VD) on both the document-noun requency matrix and document-ene binary matrix prior to applying the kmean algorithm. he ormer i almot equivalent to a Latent emantic Indexing (LI) except we had removed the adjective, verb, and noun that are not regitered in WordNet. Figure 1 how the reult o applying k- mean clutering on the document-noun requency matrix, document-ene binary matrix, and LI. For the 20newgroup data, the entropy value improve igniicantly ater incorporating ene inormation. Neverthele, thee value are till wore than the reult without applying VD (ee able 3). For the Reuter data, the entropy value or clutering uing ene inormation are generally wore than thoe obtained by applying VD on the noun and the LI method. he reult provided in thi ection ugget that tranorming the noun into ene alone i inuicient to improve the clutering reult. In act, the reult may be wore due to the increaed dimenionality o the data or when the wrong ene i choen to replace ome o the noun. Figure 1. omparion o Avg entropy or K-mean with VD Neverthele, we till expect ome document to be correctly placed in the right cluter becaue o the word ene diambiguation. What i needed i a clutering algorithm that: (1) better utilize the ene inormation in combination with the term tatitic inormation. (2) deal with the increaing number o dimenion when replacing the noun by their ene. (3) tolerant to noie when incorrect ene are ued. Becaue o it lexibility and robutne, we conjecture that enemble clutering i an appropriate approach or combining term tatitic with emantic knowledge or document clutering. 4.3 Enemble lutering In thi work, we combine the clutering reult rom both the emantic binary model (denoted a M enemble ) and the noun requency model (denoted a M Fenemble ) into one inal clutering (denoted a M enemble ). Our motivation or uing enemble clutering i becaue o it lexibility to accommodate any input data matrix (either the document-noun requency matrix, the document-ene binary matrix, or both), it ability to deal with high dimenionality via eature ubampling, and it reilience to noie and other variability in the data. Figure 2. omparion o entropy or enemble clutering Figure 2 how the reult o applying the three enemble clutering method to the Reuter and 20newgroup data. he number o iteration or the enemble in thi experiment i varied rom 10 to 30 run. For each number o iteration, we combine hal o the clutering reult rom M enemble with another hal rom M Fenemble to generate the inal clutering M enemble. A mentioned in ection 3.3, each clutering olution in the enemble will be

5 weighted according to the quality o their cluter (which i meaured in term o the correlation between the coine imilarity o the document and the reulting incidence matrix o the cluter). We oberved that the weighting actor w t aociated with the olution in M Fenemble were generally greater than the weighting actor or M enemble. For example, or the Reuter dataet the average o the weight o the M Fenemble wa approximately.5 compared to.3 in the M enemble. hi obervation i conitent with our reult in ection 4.2 where the noun requency model appear to produce better cluter than uing the emantic binary model. he weighted cluter in each enemble were aggregated in a co-aociation matrix (with α = 0.8) that relect the conenu o the individual run on allocating the document acro the cluter. Figure 2 how a comparion o the entropy value between the three enemble or both data et. For the 20newgroup dataet, the compound enemble M enemble achieved the lowet entropy core. For example, ater 20 iteration, the entropy core or the compound enemble i coniderably lower (.559) than the core or the noun requency model (.636) and emantic binary model (.787). hi reult ugget that our compound enemble method i capable o enhancing the clutering reult by taking advantage o the variability in the clutering olution obtained rom the term tatitic and emantic knowledge. Even though the emantic binary model ha higher entropy than noun requency model, it till provide ueul olution that can be exploited by our compound enemble. Furthermore, all the enemble clutering reult are igniicantly better than the reult or individual run (ection 4.2) even though it ue only a ample o the original eature. In the Reuter data et, no igniicant improvement wa oberved or the compound enemble M enemble over the requency enemble M Fenemble. hi reult ugget that the nature o the data et alo play a igniicant role in determining the eectivene o incorporating emantic knowledge rom WordNet. Figure 3. omparion o purity or enemble clutering Figure 3 how the purity value or the dierent enemble clutering method. Once again, the compound enemble achieved the highet purity core (.802) or the 20newgroup data et wherea the noun requency enemble model ha the highet purity or the Reuter data et. Finally, it i worth noting that our propoed method uing the compound enemble clutering igniicantly improved the cluter quality compared to LI or both dataet. he lowet entropy obtained or the Reuter in LI wa 1.01 uing 100 component compared to.947 ater 10 iteration uing our compound enemble method. For the 20- newgroup dataet, LI achieved an entropy o 1.13 with 100 component compared to.559 when applying the compound enemble at 20 iteration. 5. ONLUION hi paper preent a methodology or combining term tatitic with emantic knowledge rom WordNet or document clutering. Our analyi how that a traightorward replacement o the word with their ene may not necearily improve the clutering reult, which i conitent with ome o the previou reult reported in [6] and [8]. he clutering algorithm need to be lexible and robut enough to deal with the higher dimenionality and noie due to improper election o ene. o overcome thee challenge, we propoe an enemble clutering method that ytematically combine clutering olution rom a noun requency model with thoe rom a emantic binary model baed on the conitency o their clutering olution. Our experimental reult how that the enemble method i eective on ome but not all data. We are preently doing urther reearch to better determine what characteritic o the data are mot uitable or the approach. 6. AKNOWLEDGMEN We acknowledge the upport o the National Archive and Record Adminitration and the MARIX group at Michigan tate Univerity. 7. REFERENE [1] Bradley P., Bennett K., and Demiriz A., ontrained k- mean clutering. Microot Reearch echnical Report, MR-R , [2] Hotho A., taab., tumme G, WordNet improve text document clutering. In Proc. o the IGIR 2003 emantic Web Workhop, 2003, [3] Goe J., an P. N., and heng H., emi-upervied lutering with Partial Background Inormation. In Proc. o IAM Int l on on Data Mining, Betheda, MD [4] Mann H. B., Whitney D. R. On a tet whether one o two random variable i tochatically larger than the other. Annal o Mathmatical tatitic, 18, 1947, [5] Miller J., WordNet: a lexical databae or Englih, ommunication o the AM [6] edding J., Kazakov D., WordNet-baed text document clutering. In Proc. o the 3 rd Workhop on Robut Method in Analyi o Natural Language Proceing Data. 2004, [7] teinbach M. and Karypi G. and Kumar V., A comparion o document clutering technique. In proc. o KDD Workhop on ext Mining, [8] ermier A., Rouet M, ebag M, ombining tatitic and emantic or word and document clutering, In Proc. o IJAI, 2001, [9] opchy A., Jain A.K., Punch W., A mixture model or clutering enemble, In Proc. o IAM onerence on Data Mining, 2004, [10] Wu Z. and Palmer M. Verb emantic and Lexical election. In Proc. o the 32nd Annual Meeting o the Aoc. or omputational Linguitic, 1994,

A Spam Message Filtering Method: focus on run time

A Spam Message Filtering Method: focus on run time , pp.29-33 http://dx.doi.org/10.14257/atl.2014.76.08 A Spam Meage Filtering Method: focu on run time Sin-Eon Kim 1, Jung-Tae Jo 2, Sang-Hyun Choi 3 1 Department of Information Security Management 2 Department

More information

Queueing systems with scheduled arrivals, i.e., appointment systems, are typical for frontal service systems,

Queueing systems with scheduled arrivals, i.e., appointment systems, are typical for frontal service systems, MANAGEMENT SCIENCE Vol. 54, No. 3, March 28, pp. 565 572 in 25-199 ein 1526-551 8 543 565 inform doi 1.1287/mnc.17.82 28 INFORMS Scheduling Arrival to Queue: A Single-Server Model with No-Show INFORMS

More information

Partial optimal labeling search for a NP-hard subclass of (max,+) problems

Partial optimal labeling search for a NP-hard subclass of (max,+) problems Partial optimal labeling earch for a NP-hard ubcla of (max,+) problem Ivan Kovtun International Reearch and Training Center of Information Technologie and Sytem, Kiev, Uraine, ovtun@image.iev.ua Dreden

More information

Unit 11 Using Linear Regression to Describe Relationships

Unit 11 Using Linear Regression to Describe Relationships Unit 11 Uing Linear Regreion to Decribe Relationhip Objective: To obtain and interpret the lope and intercept of the leat quare line for predicting a quantitative repone variable from a quantitative explanatory

More information

1 Introduction. Reza Shokri* Privacy Games: Optimal User-Centric Data Obfuscation

1 Introduction. Reza Shokri* Privacy Games: Optimal User-Centric Data Obfuscation Proceeding on Privacy Enhancing Technologie 2015; 2015 (2):1 17 Reza Shokri* Privacy Game: Optimal Uer-Centric Data Obfucation Abtract: Conider uer who hare their data (e.g., location) with an untruted

More information

A technical guide to 2014 key stage 2 to key stage 4 value added measures

A technical guide to 2014 key stage 2 to key stage 4 value added measures A technical guide to 2014 key tage 2 to key tage 4 value added meaure CONTENTS Introduction: PAGE NO. What i value added? 2 Change to value added methodology in 2014 4 Interpretation: Interpreting chool

More information

DISTRIBUTED DATA PARALLEL TECHNIQUES FOR CONTENT-MATCHING INTRUSION DETECTION SYSTEMS

DISTRIBUTED DATA PARALLEL TECHNIQUES FOR CONTENT-MATCHING INTRUSION DETECTION SYSTEMS DISTRIBUTED DATA PARALLEL TECHNIQUES FOR CONTENT-MATCHING INTRUSION DETECTION SYSTEMS Chritopher V. Kopek Department of Computer Science Wake Foret Univerity Winton-Salem, NC, 2709 Email: kopekcv@gmail.com

More information

DISTRIBUTED DATA PARALLEL TECHNIQUES FOR CONTENT-MATCHING INTRUSION DETECTION SYSTEMS. G. Chapman J. Cleese E. Idle

DISTRIBUTED DATA PARALLEL TECHNIQUES FOR CONTENT-MATCHING INTRUSION DETECTION SYSTEMS. G. Chapman J. Cleese E. Idle DISTRIBUTED DATA PARALLEL TECHNIQUES FOR CONTENT-MATCHING INTRUSION DETECTION SYSTEMS G. Chapman J. Cleee E. Idle ABSTRACT Content matching i a neceary component of any ignature-baed network Intruion Detection

More information

Maximizing Acceptance Probability for Active Friending in Online Social Networks

Maximizing Acceptance Probability for Active Friending in Online Social Networks Maximizing for Active Friending in Online Social Network De-Nian Yang, Hui-Ju Hung, Wang-Chien Lee, Wei Chen Academia Sinica, Taipei, Taiwan The Pennylvania State Univerity, State College, Pennylvania,

More information

Optical Illusion. Sara Bolouki, Roger Grosse, Honglak Lee, Andrew Ng

Optical Illusion. Sara Bolouki, Roger Grosse, Honglak Lee, Andrew Ng Optical Illuion Sara Bolouki, Roger Groe, Honglak Lee, Andrew Ng. Introduction The goal of thi proect i to explain ome of the illuory phenomena uing pare coding and whitening model. Intead of the pare

More information

Assessing the Discriminatory Power of Credit Scores

Assessing the Discriminatory Power of Credit Scores Aeing the Dicriminatory Power of Credit Score Holger Kraft 1, Gerald Kroiandt 1, Marlene Müller 1,2 1 Fraunhofer Intitut für Techno- und Wirtchaftmathematik (ITWM) Gottlieb-Daimler-Str. 49, 67663 Kaierlautern,

More information

Profitability of Loyalty Programs in the Presence of Uncertainty in Customers Valuations

Profitability of Loyalty Programs in the Presence of Uncertainty in Customers Valuations Proceeding of the 0 Indutrial Engineering Reearch Conference T. Doolen and E. Van Aken, ed. Profitability of Loyalty Program in the Preence of Uncertainty in Cutomer Valuation Amir Gandomi and Saeed Zolfaghari

More information

T-test for dependent Samples. Difference Scores. The t Test for Dependent Samples. The t Test for Dependent Samples. s D

T-test for dependent Samples. Difference Scores. The t Test for Dependent Samples. The t Test for Dependent Samples. s D The t Tet for ependent Sample T-tet for dependent Sample (ak.a., Paired ample t-tet, Correlated Group eign, Within- Subject eign, Repeated Meaure,.. Repeated-Meaure eign When you have two et of core from

More information

1) Assume that the sample is an SRS. The problem state that the subjects were randomly selected.

1) Assume that the sample is an SRS. The problem state that the subjects were randomly selected. 12.1 Homework for t Hypothei Tet 1) Below are the etimate of the daily intake of calcium in milligram for 38 randomly elected women between the age of 18 and 24 year who agreed to participate in a tudy

More information

Improving the Performance of Web Service Recommenders Using Semantic Similarity

Improving the Performance of Web Service Recommenders Using Semantic Similarity Improving the Performance of Web Service Recommender Uing Semantic Similarity Juan Manuel Adán-Coello, Carlo Miguel Tobar, Yang Yuming Faculdade de Engenharia de Computação, Pontifícia Univeridade Católica

More information

A Resolution Approach to a Hierarchical Multiobjective Routing Model for MPLS Networks

A Resolution Approach to a Hierarchical Multiobjective Routing Model for MPLS Networks A Reolution Approach to a Hierarchical Multiobjective Routing Model for MPLS Networ Joé Craveirinha a,c, Rita Girão-Silva a,c, João Clímaco b,c, Lúcia Martin a,c a b c DEEC-FCTUC FEUC INESC-Coimbra International

More information

TIME SERIES ANALYSIS AND TRENDS BY USING SPSS PROGRAMME

TIME SERIES ANALYSIS AND TRENDS BY USING SPSS PROGRAMME TIME SERIES ANALYSIS AND TRENDS BY USING SPSS PROGRAMME RADMILA KOCURKOVÁ Sileian Univerity in Opava School of Buine Adminitration in Karviná Department of Mathematical Method in Economic Czech Republic

More information

Redesigning Ratings: Assessing the Discriminatory Power of Credit Scores under Censoring

Redesigning Ratings: Assessing the Discriminatory Power of Credit Scores under Censoring Redeigning Rating: Aeing the Dicriminatory Power of Credit Score under Cenoring Holger Kraft, Gerald Kroiandt, Marlene Müller Fraunhofer Intitut für Techno- und Wirtchaftmathematik (ITWM) Thi verion: June

More information

Group Mutual Exclusion Based on Priorities

Group Mutual Exclusion Based on Priorities Group Mutual Excluion Baed on Prioritie Karina M. Cenci Laboratorio de Invetigación en Sitema Ditribuido Univeridad Nacional del Sur Bahía Blanca, Argentina kmc@c.un.edu.ar and Jorge R. Ardenghi Laboratorio

More information

Control of Wireless Networks with Flow Level Dynamics under Constant Time Scheduling

Control of Wireless Networks with Flow Level Dynamics under Constant Time Scheduling Control of Wirele Network with Flow Level Dynamic under Contant Time Scheduling Long Le and Ravi R. Mazumdar Department of Electrical and Computer Engineering Univerity of Waterloo,Waterloo, ON, Canada

More information

TRADING rules are widely used in financial market as

TRADING rules are widely used in financial market as Complex Stock Trading Strategy Baed on Particle Swarm Optimization Fei Wang, Philip L.H. Yu and David W. Cheung Abtract Trading rule have been utilized in the tock market to make profit for more than a

More information

AN OVERVIEW ON CLUSTERING METHODS

AN OVERVIEW ON CLUSTERING METHODS IOSR Journal Engineering AN OVERVIEW ON CLUSTERING METHODS T. Soni Madhulatha Aociate Preor, Alluri Intitute Management Science, Warangal. ABSTRACT Clutering i a common technique for tatitical data analyi,

More information

CASE STUDY BRIDGE. www.future-processing.com

CASE STUDY BRIDGE. www.future-processing.com CASE STUDY BRIDGE TABLE OF CONTENTS #1 ABOUT THE CLIENT 3 #2 ABOUT THE PROJECT 4 #3 OUR ROLE 5 #4 RESULT OF OUR COLLABORATION 6-7 #5 THE BUSINESS PROBLEM THAT WE SOLVED 8 #6 CHALLENGES 9 #7 VISUAL IDENTIFICATION

More information

REDUCTION OF TOTAL SUPPLY CHAIN CYCLE TIME IN INTERNAL BUSINESS PROCESS OF REAMER USING DOE AND TAGUCHI METHODOLOGY. Abstract. 1.

REDUCTION OF TOTAL SUPPLY CHAIN CYCLE TIME IN INTERNAL BUSINESS PROCESS OF REAMER USING DOE AND TAGUCHI METHODOLOGY. Abstract. 1. International Journal of Advanced Technology & Engineering Reearch (IJATER) REDUCTION OF TOTAL SUPPLY CHAIN CYCLE TIME IN INTERNAL BUSINESS PROCESS OF REAMER USING DOE AND Abtract TAGUCHI METHODOLOGY Mr.

More information

A note on profit maximization and monotonicity for inbound call centers

A note on profit maximization and monotonicity for inbound call centers A note on profit maximization and monotonicity for inbound call center Ger Koole & Aue Pot Department of Mathematic, Vrije Univeriteit Amterdam, The Netherland 23rd December 2005 Abtract We conider an

More information

Risk Management for a Global Supply Chain Planning under Uncertainty: Models and Algorithms

Risk Management for a Global Supply Chain Planning under Uncertainty: Models and Algorithms Rik Management for a Global Supply Chain Planning under Uncertainty: Model and Algorithm Fengqi You 1, John M. Waick 2, Ignacio E. Gromann 1* 1 Dept. of Chemical Engineering, Carnegie Mellon Univerity,

More information

FEDERATION OF ARAB SCIENTIFIC RESEARCH COUNCILS

FEDERATION OF ARAB SCIENTIFIC RESEARCH COUNCILS Aignment Report RP/98-983/5/0./03 Etablihment of cientific and technological information ervice for economic and ocial development FOR INTERNAL UE NOT FOR GENERAL DITRIBUTION FEDERATION OF ARAB CIENTIFIC

More information

Trusted Document Signing based on use of biometric (Face) keys

Trusted Document Signing based on use of biometric (Face) keys Truted Document Signing baed on ue of biometric (Face) Ahmed B. Elmadani Department of Computer Science Faculty of Science Sebha Univerity Sebha Libya www.ebhau.edu.ly elmadan@yahoo.com ABSTRACT An online

More information

Cluster-Aware Cache for Network Attached Storage *

Cluster-Aware Cache for Network Attached Storage * Cluter-Aware Cache for Network Attached Storage * Bin Cai, Changheng Xie, and Qiang Cao National Storage Sytem Laboratory, Department of Computer Science, Huazhong Univerity of Science and Technology,

More information

Project Management Basics

Project Management Basics Project Management Baic A Guide to undertanding the baic component of effective project management and the key to ucce 1 Content 1.0 Who hould read thi Guide... 3 1.1 Overview... 3 1.2 Project Management

More information

Optimal Digital Content Distribution Strategy in the Presence of the Consumer-to-Consumer Channel

Optimal Digital Content Distribution Strategy in the Presence of the Consumer-to-Consumer Channel Optimal Digital Content Ditribution Strategy in the Preence o the Conumer-to-Conumer Channel Yunang Feng, Zhiling Guo, and Wei-yu Kevin Chiang Yu n a n g Fe n g received hi PhD in Inormation Sytem rom

More information

NETWORK TRAFFIC ENGINEERING WITH VARIED LEVELS OF PROTECTION IN THE NEXT GENERATION INTERNET

NETWORK TRAFFIC ENGINEERING WITH VARIED LEVELS OF PROTECTION IN THE NEXT GENERATION INTERNET Chapter 1 NETWORK TRAFFIC ENGINEERING WITH VARIED LEVELS OF PROTECTION IN THE NEXT GENERATION INTERNET S. Srivatava Univerity of Miouri Kana City, USA hekhar@conrel.ice.umkc.edu S. R. Thirumalaetty now

More information

Two Dimensional FEM Simulation of Ultrasonic Wave Propagation in Isotropic Solid Media using COMSOL

Two Dimensional FEM Simulation of Ultrasonic Wave Propagation in Isotropic Solid Media using COMSOL Excerpt from the Proceeding of the COMSO Conference 0 India Two Dimenional FEM Simulation of Ultraonic Wave Propagation in Iotropic Solid Media uing COMSO Bikah Ghoe *, Krihnan Balaubramaniam *, C V Krihnamurthy

More information

6. Friction, Experiment and Theory

6. Friction, Experiment and Theory 6. Friction, Experiment and Theory The lab thi wee invetigate the rictional orce and the phyical interpretation o the coeicient o riction. We will mae ue o the concept o the orce o gravity, the normal

More information

A Note on Profit Maximization and Monotonicity for Inbound Call Centers

A Note on Profit Maximization and Monotonicity for Inbound Call Centers OPERATIONS RESEARCH Vol. 59, No. 5, September October 2011, pp. 1304 1308 in 0030-364X ein 1526-5463 11 5905 1304 http://dx.doi.org/10.1287/opre.1110.0990 2011 INFORMS TECHNICAL NOTE INFORMS hold copyright

More information

Independent Samples T- test

Independent Samples T- test Independent Sample T- tet With previou tet, we were intereted in comparing a ingle ample with a population With mot reearch, you do not have knowledge about the population -- you don t know the population

More information

Assigning Tasks for Efficiency in Hadoop

Assigning Tasks for Efficiency in Hadoop Aigning Tak for Efficiency in Hadoop [Extended Abtract] Michael J. Ficher Computer Science Yale Univerity P.O. Box 208285 New Haven, CT, USA michael.ficher@yale.edu Xueyuan Su Computer Science Yale Univerity

More information

Scheduling of Jobs and Maintenance Activities on Parallel Machines

Scheduling of Jobs and Maintenance Activities on Parallel Machines Scheduling of Job and Maintenance Activitie on Parallel Machine Chung-Yee Lee* Department of Indutrial Engineering Texa A&M Univerity College Station, TX 77843-3131 cylee@ac.tamu.edu Zhi-Long Chen** Department

More information

Bi-Objective Optimization for the Clinical Trial Supply Chain Management

Bi-Objective Optimization for the Clinical Trial Supply Chain Management Ian David Lockhart Bogle and Michael Fairweather (Editor), Proceeding of the 22nd European Sympoium on Computer Aided Proce Engineering, 17-20 June 2012, London. 2012 Elevier B.V. All right reerved. Bi-Objective

More information

Performance of a Browser-Based JavaScript Bandwidth Test

Performance of a Browser-Based JavaScript Bandwidth Test Performance of a Brower-Baed JavaScript Bandwidth Tet David A. Cohen II May 7, 2013 CP SC 491/H495 Abtract An exiting brower-baed bandwidth tet written in JavaScript wa modified for the purpoe of further

More information

SELF-MANAGING PERFORMANCE IN APPLICATION SERVERS MODELLING AND DATA ARCHITECTURE

SELF-MANAGING PERFORMANCE IN APPLICATION SERVERS MODELLING AND DATA ARCHITECTURE SELF-MANAGING PERFORMANCE IN APPLICATION SERVERS MODELLING AND DATA ARCHITECTURE RAVI KUMAR G 1, C.MUTHUSAMY 2 & A.VINAYA BABU 3 1 HP Bangalore, Reearch Scholar JNTUH, Hyderabad, India, 2 Yahoo, Bangalore,

More information

Queueing Models for Multiclass Call Centers with Real-Time Anticipated Delays

Queueing Models for Multiclass Call Centers with Real-Time Anticipated Delays Queueing Model for Multicla Call Center with Real-Time Anticipated Delay Oualid Jouini Yve Dallery Zeynep Akşin Ecole Centrale Pari Koç Univerity Laboratoire Génie Indutriel College of Adminitrative Science

More information

Acceleration-Displacement Crash Pulse Optimisation A New Methodology to Optimise Vehicle Response for Multiple Impact Speeds

Acceleration-Displacement Crash Pulse Optimisation A New Methodology to Optimise Vehicle Response for Multiple Impact Speeds Acceleration-Diplacement Crah Pule Optimiation A New Methodology to Optimie Vehicle Repone for Multiple Impact Speed D. Gildfind 1 and D. Ree 2 1 RMIT Univerity, Department of Aeropace Engineering 2 Holden

More information

Unobserved Heterogeneity and Risk in Wage Variance: Does Schooling Provide Earnings Insurance?

Unobserved Heterogeneity and Risk in Wage Variance: Does Schooling Provide Earnings Insurance? TI 011-045/3 Tinbergen Intitute Dicuion Paper Unoberved Heterogeneity and Rik in Wage Variance: Doe Schooling Provide Earning Inurance? Jacopo Mazza Han van Ophem Joop Hartog * Univerity of Amterdam; *

More information

SCM- integration: organiational, managerial and technological iue M. Caridi 1 and A. Sianei 2 Dipartimento di Economia e Produzione, Politecnico di Milano, Italy E-mail: maria.caridi@polimi.it Itituto

More information

CASE STUDY ALLOCATE SOFTWARE

CASE STUDY ALLOCATE SOFTWARE CASE STUDY ALLOCATE SOFTWARE allocate caetud y TABLE OF CONTENTS #1 ABOUT THE CLIENT #2 OUR ROLE #3 EFFECTS OF OUR COOPERATION #4 BUSINESS PROBLEM THAT WE SOLVED #5 CHALLENGES #6 WORKING IN SCRUM #7 WHAT

More information

Research Article An (s, S) Production Inventory Controlled Self-Service Queuing System

Research Article An (s, S) Production Inventory Controlled Self-Service Queuing System Probability and Statitic Volume 5, Article ID 558, 8 page http://dxdoiorg/55/5/558 Reearch Article An (, S) Production Inventory Controlled Self-Service Queuing Sytem Anoop N Nair and M J Jacob Department

More information

Software Engineering Management: strategic choices in a new decade

Software Engineering Management: strategic choices in a new decade Software Engineering : trategic choice in a new decade Barbara Farbey & Anthony Finkeltein Univerity College London, Department of Computer Science, Gower St. London WC1E 6BT, UK {b.farbey a.finkeltein}@ucl.ac.uk

More information

Support Vector Machine Based Electricity Price Forecasting For Electricity Markets utilising Projected Assessment of System Adequacy Data.

Support Vector Machine Based Electricity Price Forecasting For Electricity Markets utilising Projected Assessment of System Adequacy Data. The Sixth International Power Engineering Conference (IPEC23, 27-29 November 23, Singapore Support Vector Machine Baed Electricity Price Forecating For Electricity Maret utiliing Projected Aement of Sytem

More information

Figure 2.1. a. Block diagram representation of a system; b. block diagram representation of an interconnection of subsystems

Figure 2.1. a. Block diagram representation of a system; b. block diagram representation of an interconnection of subsystems Figure. a. Block diagram repreentation o a ytem; b. block diagram repreentation o an interconnection o ubytem REVIEW OF THE LAPLACE TRANSFORM Table. Laplace tranorm table Table. Laplace tranorm theorem

More information

Sector Concentration in Loan Portfolios and Economic Capital. Abstract

Sector Concentration in Loan Portfolios and Economic Capital. Abstract Sector Concentration in Loan Portfolio and Economic Capital Klau Düllmann and Nancy Machelein 2 Thi verion: September 2006 Abtract The purpoe of thi paper i to meaure the potential impact of buine-ector

More information

QUANTIFYING THE BULLWHIP EFFECT IN THE SUPPLY CHAIN OF SMALL-SIZED COMPANIES

QUANTIFYING THE BULLWHIP EFFECT IN THE SUPPLY CHAIN OF SMALL-SIZED COMPANIES Sixth LACCEI International Latin American and Caribbean Conference for Engineering and Technology (LACCEI 2008) Partnering to Succe: Engineering, Education, Reearch and Development June 4 June 6 2008,

More information

Pekka Helkiö, 58490K Antti Seppälä, 63212W Ossi Syd, 63513T

Pekka Helkiö, 58490K Antti Seppälä, 63212W Ossi Syd, 63513T Pekka Helkiö, 58490K Antti Seppälä, 63212W Oi Syd, 63513T Table of Content 1. Abtract...1 2. Introduction...2 2.1 Background... 2 2.2 Objective and Reearch Problem... 2 2.3 Methodology... 2 2.4 Scoping

More information

Name: SID: Instructions

Name: SID: Instructions CS168 Fall 2014 Homework 1 Aigned: Wedneday, 10 September 2014 Due: Monday, 22 September 2014 Name: SID: Dicuion Section (Day/Time): Intruction - Submit thi homework uing Pandagrader/GradeScope(http://www.gradecope.com/

More information

Evaluating Teaching in Higher Education. September 2008. Bruce A. Weinberg The Ohio State University *, IZA, and NBER weinberg.27@osu.

Evaluating Teaching in Higher Education. September 2008. Bruce A. Weinberg The Ohio State University *, IZA, and NBER weinberg.27@osu. Evaluating Teaching in Higher Education September 2008 Bruce A. Weinberg The Ohio State Univerity *, IZA, and NBER weinberg.27@ou.edu Belton M. Fleiher The Ohio State Univerity * and IZA fleiher.1@ou.edu

More information

Mobility Improves Coverage of Sensor Networks

Mobility Improves Coverage of Sensor Networks Mobility Improve Coverage of Senor Networ Benyuan Liu Dept. of Computer Science Univerity of Maachuett-Lowell Lowell, MA 1854 Peter Bra Dept. of Computer Science City College of New Yor New Yor, NY 131

More information

MANAGING DATA REPLICATION IN MOBILE AD- HOC NETWORK DATABASES (Invited Paper) *

MANAGING DATA REPLICATION IN MOBILE AD- HOC NETWORK DATABASES (Invited Paper) * MANAGING DATA REPLICATION IN MOBILE AD- HOC NETWORK DATABASES (Invited Paper) * Praanna Padmanabhan School of Computer Science The Univerity of Oklahoma Norman OK, USA praannap@yahoo-inc.com Dr. Le Gruenwald

More information

A New Optimum Jitter Protection for Conversational VoIP

A New Optimum Jitter Protection for Conversational VoIP Proc. Int. Conf. Wirele Commun., Signal Proceing (Nanjing, China), 5 pp., Nov. 2009 A New Optimum Jitter Protection for Converational VoIP Qipeng Gong, Peter Kabal Electrical & Computer Engineering, McGill

More information

Proceedings of Power Tech 2007, July 1-5, Lausanne

Proceedings of Power Tech 2007, July 1-5, Lausanne Second Order Stochatic Dominance Portfolio Optimization for an Electric Energy Company M.-P. Cheong, Student Member, IEEE, G. B. Sheble, Fellow, IEEE, D. Berleant, Senior Member, IEEE and C.-C. Teoh, Student

More information

462 Machine Translation Systems for Europe

462 Machine Translation Systems for Europe 462 Machine Tranlation Sytem for Europe Philipp Koehn School of Informatic Univerity of Edinburgh pkoehn@inf.ed.ac.uk Alexandra Birch School of Informatic Univerity of Edinburgh a.birch@m.ed.ac.uk Ralf

More information

A Life Contingency Approach for Physical Assets: Create Volatility to Create Value

A Life Contingency Approach for Physical Assets: Create Volatility to Create Value A Life Contingency Approach for Phyical Aet: Create Volatility to Create Value homa Emil Wendling 2011 Enterprie Rik Management Sympoium Society of Actuarie March 14-16, 2011 Copyright 2011 by the Society

More information

Review of Multiple Regression Richard Williams, University of Notre Dame, http://www3.nd.edu/~rwilliam/ Last revised January 13, 2015

Review of Multiple Regression Richard Williams, University of Notre Dame, http://www3.nd.edu/~rwilliam/ Last revised January 13, 2015 Review of Multiple Regreion Richard William, Univerity of Notre Dame, http://www3.nd.edu/~rwilliam/ Lat revied January 13, 015 Aumption about prior nowledge. Thi handout attempt to ummarize and yntheize

More information

Mobile Network Configuration for Large-scale Multimedia Delivery on a Single WLAN

Mobile Network Configuration for Large-scale Multimedia Delivery on a Single WLAN Mobile Network Configuration for Large-cale Multimedia Delivery on a Single WLAN Huigwang Je, Dongwoo Kwon, Hyeonwoo Kim, and Hongtaek Ju Dept. of Computer Engineering Keimyung Univerity Daegu, Republic

More information

Bidding for Representative Allocations for Display Advertising

Bidding for Representative Allocations for Display Advertising Bidding for Repreentative Allocation for Diplay Advertiing Arpita Ghoh, Preton McAfee, Kihore Papineni, and Sergei Vailvitkii Yahoo! Reearch. {arpita, mcafee, kpapi, ergei}@yahoo-inc.com Abtract. Diplay

More information

Growing Self-Organizing Maps for Surface Reconstruction from Unstructured Point Clouds

Growing Self-Organizing Maps for Surface Reconstruction from Unstructured Point Clouds Growing Self-Organizing Map for Surface Recontruction from Untructured Point Cloud Renata L. M. E. do Rêgo, Aluizio F. R. Araújo, and Fernando B.de Lima Neto Abtract Thi work introduce a new method for

More information

Morningstar Fixed Income Style Box TM Methodology

Morningstar Fixed Income Style Box TM Methodology Morningtar Fixed Income Style Box TM Methodology Morningtar Methodology Paper Augut 3, 00 00 Morningtar, Inc. All right reerved. The information in thi document i the property of Morningtar, Inc. Reproduction

More information

The Arms Race on American Roads: The Effect of SUV s and Pickup Trucks on Traffic Safety

The Arms Race on American Roads: The Effect of SUV s and Pickup Trucks on Traffic Safety The Arm Race on American Road: The Effect of SUV and Pickup Truck on Traffic Safety Michelle J. White Univerity of California, San Diego, and NBER Abtract Driver have been running an arm race on American

More information

DUE to the small size and low cost of a sensor node, a

DUE to the small size and low cost of a sensor node, a 1992 IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 14, NO. 10, OCTOBER 2015 A Networ Coding Baed Energy Efficient Data Bacup in Survivability-Heterogeneou Senor Networ Jie Tian, Tan Yan, and Guiling Wang

More information

Graph Analyi I Network Meaure of the Networked Adaptive Agents

Graph Analyi I Network Meaure of the Networked Adaptive Agents Uing Graph Analyi to Study Network of Adaptive Agent Sherief Abdallah Britih Univerity in Dubai, United Arab Emirate Univerity of Edinburgh, United Kingdom hario@ieee.org ABSTRACT Experimental analyi of

More information

Online story scheduling in web advertising

Online story scheduling in web advertising Online tory cheduling in web advertiing Anirban Dagupta Arpita Ghoh Hamid Nazerzadeh Prabhakar Raghavan Abtract We tudy an online job cheduling problem motivated by toryboarding in web advertiing, where

More information

Performance of Multiple TFRC in Heterogeneous Wireless Networks

Performance of Multiple TFRC in Heterogeneous Wireless Networks Performance of Multiple TFRC in Heterogeneou Wirele Network 1 Hyeon-Jin Jeong, 2 Seong-Sik Choi 1, Firt Author Computer Engineering Department, Incheon National Univerity, oaihjj@incheon.ac.kr *2,Correponding

More information

THE IMPACT OF MULTIFACTORIAL GENETIC DISORDERS ON CRITICAL ILLNESS INSURANCE: A SIMULATION STUDY BASED ON UK BIOBANK ABSTRACT KEYWORDS

THE IMPACT OF MULTIFACTORIAL GENETIC DISORDERS ON CRITICAL ILLNESS INSURANCE: A SIMULATION STUDY BASED ON UK BIOBANK ABSTRACT KEYWORDS THE IMPACT OF MULTIFACTORIAL GENETIC DISORDERS ON CRITICAL ILLNESS INSURANCE: A SIMULATION STUDY BASED ON UK BIOBANK BY ANGUS MACDONALD, DELME PRITCHARD AND PRADIP TAPADAR ABSTRACT The UK Biobank project

More information

Exposure Metering Relating Subject Lighting to Film Exposure

Exposure Metering Relating Subject Lighting to Film Exposure Expoure Metering Relating Subject Lighting to Film Expoure By Jeff Conrad A photographic expoure meter meaure ubject lighting and indicate camera etting that nominally reult in the bet expoure of the film.

More information

CLUSTBIGFIM-FREQUENT ITEMSET MINING OF BIG DATA USING PRE-PROCESSING BASED ON MAPREDUCE FRAMEWORK

CLUSTBIGFIM-FREQUENT ITEMSET MINING OF BIG DATA USING PRE-PROCESSING BASED ON MAPREDUCE FRAMEWORK CLUSTBIGFIM-FREQUENT ITEMSET MINING OF BIG DATA USING PRE-PROCESSING BASED ON MAPREDUCE FRAMEWORK Sheela Gole 1 and Bharat Tidke 2 1 Department of Computer Engineering, Flora Intitute of Technology, Pune,

More information

Progress 8 measure in 2016, 2017, and 2018. Guide for maintained secondary schools, academies and free schools

Progress 8 measure in 2016, 2017, and 2018. Guide for maintained secondary schools, academies and free schools Progre 8 meaure in 2016, 2017, and 2018 Guide for maintained econdary chool, academie and free chool July 2016 Content Table of figure 4 Summary 5 A ummary of Attainment 8 and Progre 8 5 Expiry or review

More information

Benchmarking Bottom-Up and Top-Down Strategies for SPARQL-to-SQL Query Translation

Benchmarking Bottom-Up and Top-Down Strategies for SPARQL-to-SQL Query Translation Benchmarking Bottom-Up and Top-Down Strategie for SPARQL-to-SQL Query Tranlation Kahlev a, Chebotko b,c, John Abraham b, Pearl Brazier b, and Shiyong Lu a a Department of Computer Science, Wayne State

More information

Risk-Sharing within Families: Evidence from the Health and Retirement Study

Risk-Sharing within Families: Evidence from the Health and Retirement Study Rik-Sharing within Familie: Evidence from the Health and Retirement Study Ş. Nuray Akın and Okana Leukhina December 14, 2014 We report trong empirical upport for the preence of elf-interet-baed rik haring

More information

Control Theory based Approach for the Improvement of Integrated Business Process Interoperability

Control Theory based Approach for the Improvement of Integrated Business Process Interoperability www.ijcsi.org 201 Control Theory baed Approach for the Improvement of Integrated Buine Proce Interoperability Abderrahim Taoudi 1, Bouchaib Bounabat 2 and Badr Elmir 3 1 Al-Qualadi Reearch & Development

More information

January 21, 2015. Abstract

January 21, 2015. Abstract T S U I I E P : T R M -C S J. R January 21, 2015 Abtract Thi paper evaluate the trategic behavior of a monopolit to influence environmental policy, either with taxe or with tandard, comparing two alternative

More information

INFORMATION Technology (IT) infrastructure management

INFORMATION Technology (IT) infrastructure management IEEE TRANSACTIONS ON CLOUD COMPUTING, VOL. 2, NO. 1, MAY 214 1 Buine-Driven Long-term Capacity Planning for SaaS Application David Candeia, Ricardo Araújo Santo and Raquel Lope Abtract Capacity Planning

More information

Measuring the Ability of Score Distributions to Model Relevance

Measuring the Ability of Score Distributions to Model Relevance Meauring the Ability of Score Ditribution to Model Relevance Ronan Cummin Department of Information Technology National Univerity of Ireland, Galway ronan.cummin@nuigalway.ie Abtract. Modelling the core

More information

Return on Investment and Effort Expenditure in the Software Development Environment

Return on Investment and Effort Expenditure in the Software Development Environment International Journal of Applied Information ytem (IJAI) IN : 2249-0868 Return on Invetment and Effort Expenditure in the oftware Development Environment Dineh Kumar aini Faculty of Computing and IT, ohar

More information

Senior Thesis. Horse Play. Optimal Wagers and the Kelly Criterion. Author: Courtney Kempton. Supervisor: Professor Jim Morrow

Senior Thesis. Horse Play. Optimal Wagers and the Kelly Criterion. Author: Courtney Kempton. Supervisor: Professor Jim Morrow Senior Thei Hore Play Optimal Wager and the Kelly Criterion Author: Courtney Kempton Supervior: Profeor Jim Morrow June 7, 20 Introduction The fundamental problem in gambling i to find betting opportunitie

More information

Growth and Sustainability of Managed Security Services Networks: An Economic Perspective

Growth and Sustainability of Managed Security Services Networks: An Economic Perspective Growth and Sutainability of Managed Security Service etwork: An Economic Perpective Alok Gupta Dmitry Zhdanov Department of Information and Deciion Science Univerity of Minneota Minneapoli, M 55455 (agupta,

More information

Unusual Option Market Activity and the Terrorist Attacks of September 11, 2001*

Unusual Option Market Activity and the Terrorist Attacks of September 11, 2001* Allen M. Potehman Univerity of Illinoi at Urbana-Champaign Unuual Option Market Activity and the Terrorit Attack of September 11, 2001* I. Introduction In the aftermath of the terrorit attack on the World

More information

Pipe Flow Calculations

Pipe Flow Calculations Pipe Flow Calculation R. Shankar Subramanian epartment o Chemical and Biomolecular Engineering Clarkon Univerity We begin with ome reult that we hall ue when making riction lo calculation or teady, ully

More information

Trade Elasticities PRELIMINARY. DO NOT CIRCULATE. Abstract

Trade Elasticities PRELIMINARY. DO NOT CIRCULATE. Abstract Trade Elaticitie Jean Imb y Iabelle Méjean z PRELIMINARY. DO NOT CIRCULATE Abtract Arkolaki et al (2011) how the welfare gain from trade can be ummarized by import hare and the price elaticity of import.

More information

Growth and Sustainability of Managed Security Services Networks: An Economic Perspective

Growth and Sustainability of Managed Security Services Networks: An Economic Perspective Growth and Sutainability of Managed Security Service etwork: An Economic Perpective Alok Gupta Dmitry Zhdanov Department of Information and Deciion Science Univerity of Minneota Minneapoli, M 55455 (agupta,

More information

TOWARDS AUTOMATED LIDAR BORESIGHT SELF-CALIBRATION

TOWARDS AUTOMATED LIDAR BORESIGHT SELF-CALIBRATION TOWARDS AUTOMATED LIDAR BORESIGHT SELF-CALIBRATION J. Skaloud a, *, P. Schaer a a TOPO Lab, Ecole Polytechnique Fédérale de Lauanne (EPFL), Station 18, 1015 Lauanne, Switzerland KEY WORDS: airborne laer

More information

2. METHOD DATA COLLECTION

2. METHOD DATA COLLECTION Key to learning in pecific ubject area of engineering education an example from electrical engineering Anna-Karin Cartenen,, and Jonte Bernhard, School of Engineering, Jönköping Univerity, S- Jönköping,

More information

Distributed, Secure Load Balancing with Skew, Heterogeneity, and Churn

Distributed, Secure Load Balancing with Skew, Heterogeneity, and Churn Ditributed, Secure Load Balancing with Skew, Heterogeneity, and Churn Jonathan Ledlie and Margo Seltzer Diviion of Engineering and Applied Science Harvard Univerity Abtract Numerou propoal exit for load

More information

MSc Financial Economics: International Finance. Bubbles in the Foreign Exchange Market. Anne Sibert. Revised Spring 2013. Contents

MSc Financial Economics: International Finance. Bubbles in the Foreign Exchange Market. Anne Sibert. Revised Spring 2013. Contents MSc Financial Economic: International Finance Bubble in the Foreign Exchange Market Anne Sibert Revied Spring 203 Content Introduction................................................. 2 The Mone Market.............................................

More information

BUILT-IN DUAL FREQUENCY ANTENNA WITH AN EMBEDDED CAMERA AND A VERTICAL GROUND PLANE

BUILT-IN DUAL FREQUENCY ANTENNA WITH AN EMBEDDED CAMERA AND A VERTICAL GROUND PLANE Progre In Electromagnetic Reearch Letter, Vol. 3, 51, 08 BUILT-IN DUAL FREQUENCY ANTENNA WITH AN EMBEDDED CAMERA AND A VERTICAL GROUND PLANE S. H. Zainud-Deen Faculty of Electronic Engineering Menoufia

More information

Corporate Tax Aggressiveness and the Role of Debt

Corporate Tax Aggressiveness and the Role of Debt Corporate Tax Aggreivene and the Role of Debt Akankha Jalan, Jayant R. Kale, and Cotanza Meneghetti Abtract We examine the effect of leverage on corporate tax aggreivene. We derive the optimal level of

More information

Socially Optimal Pricing of Cloud Computing Resources

Socially Optimal Pricing of Cloud Computing Resources Socially Optimal Pricing of Cloud Computing Reource Ihai Menache Microoft Reearch New England Cambridge, MA 02142 t-imena@microoft.com Auman Ozdaglar Laboratory for Information and Deciion Sytem Maachuett

More information

Report 4668-1b 30.10.2010. Measurement report. Sylomer - field test

Report 4668-1b 30.10.2010. Measurement report. Sylomer - field test Report 4668-1b Meaurement report Sylomer - field tet Report 4668-1b 2(16) Contet 1 Introduction... 3 1.1 Cutomer... 3 1.2 The ite and purpoe of the meaurement... 3 2 Meaurement... 6 2.1 Attenuation of

More information

Tax Evasion and Self-Employment in a High-Tax Country: Evidence from Sweden

Tax Evasion and Self-Employment in a High-Tax Country: Evidence from Sweden Tax Evaion and Self-Employment in a High-Tax Country: Evidence from Sweden by Per Engtröm * and Bertil Holmlund ** Thi verion: May 17, 2006 Abtract Self-employed individual have arguably greater opportunitie

More information

CHARACTERISTICS OF WAITING LINE MODELS THE INDICATORS OF THE CUSTOMER FLOW MANAGEMENT SYSTEMS EFFICIENCY

CHARACTERISTICS OF WAITING LINE MODELS THE INDICATORS OF THE CUSTOMER FLOW MANAGEMENT SYSTEMS EFFICIENCY Annale Univeritati Apuleni Serie Oeconomica, 2(2), 200 CHARACTERISTICS OF WAITING LINE MODELS THE INDICATORS OF THE CUSTOMER FLOW MANAGEMENT SYSTEMS EFFICIENCY Sidonia Otilia Cernea Mihaela Jaradat 2 Mohammad

More information

Bio-Plex Analysis Software

Bio-Plex Analysis Software Multiplex Supenion Array Bio-Plex Analyi Software The Leader in Multiplex Immunoaay Analyi Bio-Plex Analyi Software If making ene of your multiplex data i your challenge, then Bio-Plex data analyi oftware

More information

HUMAN CAPITAL AND THE FUTURE OF TRANSITION ECONOMIES * Michael Spagat Royal Holloway, University of London, CEPR and Davidson Institute.

HUMAN CAPITAL AND THE FUTURE OF TRANSITION ECONOMIES * Michael Spagat Royal Holloway, University of London, CEPR and Davidson Institute. HUMAN CAPITAL AND THE FUTURE OF TRANSITION ECONOMIES * By Michael Spagat Royal Holloway, Univerity of London, CEPR and Davidon Intitute Abtract Tranition economie have an initial condition of high human

More information