Statistical Techniques Employed in Atmospheric Sampling

Size: px
Start display at page:

Download "Statistical Techniques Employed in Atmospheric Sampling"

Transcription

1 Appedx A Statstcal Techques Employed Atmospherc Samplg A.1 Itroducto Proper use of statstcs ad statstcal techques s ecessary for assessg the qualty of ambet ar samplg data. For a comprehesve dscusso of the subject of data qualty assessmet (DQA), revew EPA s techcal assstace documet, Gudace for Data Qualty Assessmet, Practcal Methods of Data Aalyss, EPA QA/G-9 (EPA/600/R-96/084), Jauary Ths referece documet provdes practcal demostratos o how to use the data qualty assessmet (DQA) techque evaluatg evrometal data sets ad shows how to apply some graphc ad statstcal tools for performg DQA. Ths chapter s teded as a troducto to statstcs ad statstcal cocepts ad ther use aalyzg ambet ar samplg data. Topcs addressed clude: (a) Data Qualty Objectves (DQO), (b) Data Plottg, (c) Measures of Cetral Tedecy, (d) Measures of Dsperso, ad (e) Dstrbuto Curves. Although these topcs are ot smple, they ca be uderstood ad used by o-statstcas. If a detaled statstcal aalyss of data s requred, t s recommeded that a expereced statstca be cosulted. Studets who could beeft from a revew of basc mathematcs ambet motorg are ecouraged to access the EPA Ar Polluto Trag Isttute course, SI 100: Mathematcs Revew for Ar Polluto Cotrol. Ths selfstructo course ca be foud at: I addto, the Uversty of Illos-Chcago, School of Publc Health- Evrometal ad Occupatoal Health Dvso, has developed a Iteretbased program ettled Itroducto to Evrometal Statstcs. Ths program s preseted as a vdeo seres three modules o topcs whch clude terpretg motorg data, samplg ad aalytcal lmtatos ad sample detecto lmts, ad qualty assurace ad qualty cotrol. Ths program ca be foud at: A-1

2 It s mportat to ote that the statstcal calculatos dscussed ths Appedx are best ad more easly performed by employg oe of may commercally avalable computer-based statstcal software packages. A. The Data Qualty Objectves (DQO) Process Whle the Data Qualty Objectves (DQO) Process s ot a statstcal techque per se, t s mportat because t helps to establsh crtera for data qualty ad the developmet of data collecto desgs. DQOs provde the approprate cotext for uderstadg the purpose of the ambet ar samplg ad aalyss data collecto effort. Also, they establsh the qualtatve ad quattatve crtera for assessg the qualty of the collected data set, based o the predefed teded use of data. Specfc formato o the Data Qualty Objectves Process ca be foud EPA documet, Gudace o Systematc Plag Usg the Data Qualty Objectve Process (EPA QA/G-4), at: DQOs are qualtatve ad quattatve statemets derved from the outputs of the frst sx steps of the DQO Process that ecompass the followg: Clarfy the study objectve. Defe the most approprate type of data to collect. Determe the most approprate codtos from whch to collect the data. Specfy tolerable lmts o decso errors whch wll be used as the bass for establshg the quatty ad qualty of data eeded to support the decso. The DQOs are the used to develop a scetfc ad resource-effectve data collecto desg. The Seve Steps of the DQO Process Step 1: Step : Step 3: Step 4: Step 5: State the Problem. Cocsely descrbe the problem to be studed. Revew pror studes ad exstg formato to ga a suffcet uderstadg to defe the problem. Idetfy the Goal of the Study. Idetfy what questos the study wll attempt to aswer. Idetfy Iformato Iputs. Idetfy the formato that eeds to be obtaed ad the measuremets that eed to be take to resolve the decso statemet. Defe Boudares of the Study. Specfy the tme perods ad spatal area to whch decsos wll apply. Determe whe ad where data should be collected. Develop the Aalytcal Approach. Defe the statstcal parameters of terest, specfy the acto level, ad tegrate the prevous DQO outputs to a sgle statemet that descrbes the logcal bass for choosg amog alteratve actos. A-

3 Step 6: Step 7: Specfy the Performace or Acceptace Crtera. Defe the decso maker s tolerable decso error rates based o a cosderato of the cosequeces of makg a correct decso. Develop the Pla for Obtag Data. Evaluate formato from the prevous steps ad geerate alteratve data collecto desgs. Choose the most resource-effectve desg that meets all DQOs. Outputs of the DQO Process The DQO Process leads to the developmet of a quattatve ad qualtatve framework for a study. Each step of the Process derves valuable crtera that wll be used to establsh the fal data collecto desg. The frst fve steps of the DQO Process detfy mostly qualtatve crtera, such as what problem has tated the study ad what decso t attempts to resolve. These steps also defe the type of data that wll be collected, where ad whe the data wll be collected, ad a decso rule that specfes how the decso wll be made. The sxth step defes quattatve crtera expressed as lmts o decso errors that the decso maker ca tolerate. The fal step s used to develop a data collecto desg based o the crtera developed the frst sx steps. The fal product of the DQO Process s a data collecto desg that meets the quattatve ad qualtatve eeds of the study. A.3 Data Collecto Desg A data collecto desg specfes the fal cofgurato of the evrometal motorg or measuremet effort requred to satsfy the DQOs. It desgates: the types ad quattes of samples or motorg formato to be collected; where, whe, ad uder what codtos they should be collected; what varables are to be measured; ad QA/QC procedures to esure that samplg desg ad measuremet errors are cotrolled suffcetly to meet the tolerable decso error rates specfed the DQOs. Data Plottg Data s usually uterpretable the form whch t s collected. I ths secto, we shall cosder the graphcal techques of summarzg such data so that the meagful formato ca be extracted from t. There are two kds of varables to whch we assg data: cotuous varables ad dscrete varables. A cotuous varable s oe that ca assume ay value some terval of values. Examples of cotuous varables are weght, volume, legth, tme, ad temperature. Most ar polluto data are take from cotuous varables. Dscrete varables, o the other had, are those varables whose possble values are tegers. Therefore, they volve coutg rather tha measurg. Examples of dscrete varables are the umber of sample statos, umber of people a room, ad umber of tmes a cotrol stadard s volated. Sce ay measurg A-3

4 devce s of lmted accuracy, measuremets real lfe are actually dscrete ature rather tha cotuous, but ths should ot keep us from regardg such varables as cotuous. Whe a weght s recorded as 165 pouds, t s assumed that the actual weght s somewhere betwee ad pouds. A.4 Graphcal Aalyss Frequecy Tables Let us cosder the set of data Table A-1, whch represets SO levels for a gve hour for 5 days. The frst step summarzg the data s to form a frequecy table. A frequecy table s a table prepared by dvdg a data set to selected uts or class tervals, the coutg ad sertg the umber of pots (frequecy of occurreces) wth the uts or class tervals. Table A- s a frequecy table prepared from the SO data set gve Table A-1. I costructg the frequecy table, we have dvded the 5 pots the data set to 11 class tervals wth each terval beg 15 uts legth. The choce of dvdg the data to 11 tervals was purely arbtrary. However, dealg wth data t s best to choose the legth of the class terval such that 8 to 15 tervals wll clude all of the data uder cosderato. Dervg the frequecy of occurrece colum volves othg more tha coutg the umber of values each terval. The relatve frequecy colum s obtaed by dvdg the umber of pots or frequecy of occurreces wth a ut by the total umber of evets wth the data set, whch ths example s 5. From observato of the frequecy table, we ca ow see the data takg form. The values appear to be clustered betwee 5 ad 85 ppb. I fact, early 80% are ths terval. A-4

5 Table A-1. SO levels. Days SO Cocetrato (ppb)* *ppb = parts per bllo collected SO levels. Class Iterval (ppb) Table A-. Frequecy table. Frequecy of Occurrece (total 5) Relatve Frequecy /5 = /5 = /5= /5= /5 = /5= /5= /5 = A-5

6 The Frequecy Polygo The ext step s to graph the formato the frequecy table. Oe way of dog ths would be to plot the frequecy for the mdpot of each class terval. The sold le coectg the pots of Fgure A-1 forms a frequecy polygo. Fgure A-1. Polluto cocetrato (mdpot of class terval) frequecy polygo. The Hstogram Aother method of graphg the formato would be by costructg a hstogram as show Fgure A-. The hstogram s a two-dmesoal graph whch the legth of the class terval s take to cosderato. The hstogram ca be a very useful tool statstcs, especally f we covert the gve frequecy scale to a relatve scale so that the sum of all the ordates equals oe. Ths s show Fgure A-3. Thus, each ordate value s derved by dvdg the orgal value by the umber of observatos the sample, ths case, 5. The advatage costructg a hstogram lke ths oe s that we ca read probabltes from t, f we ca assume a scale o the abscssa such that a gve value wll fall ay oe terval the area uder the curve that terval. For example, the probablty that a value wll fall betwee 55 ad 70 s equal to ts assocated terval's porto of the total area of tervals, whch s A-6

7 Fgure A-. Pollutat cocetrato hstogram of frequecy dstrbuto curve. Fgure A-3. Hstogram of percet frequecy dstrbuto curve. A-7

8 The Cumulatve Frequecy Dstrbuto Usg the frequecy table ad hstogram dscussed above, we ca costruct a cumulatve frequecy table ad curve as show Table A-3 ad Fgure A-4. Table A-3. Cumulatve frequecy table. SO level Cumulatve frequecy Relatve cumulatve frequecy Uder Fgure A-4. Cumulatve frequecy dstrbuto curve. A-8

9 The cumulatve frequecy table gves the umber of observatos less tha a gve value. Probabltes ca be read from the cumulatve frequecy curve or cumulatve frequecy table. For example, to fd the probablty that a value wll be less tha 85, we read up to the curve at the pot x = 85 ad across to the value 0.76 o the y-axs. A alteratve way to use the table s to go to the row where the SO level shows uder 85, the go across to the relatve cumulatve frequecy value of Dstrbuto of Data Whe we draw a hstogram for a set of data, we are represetg the dstrbuto of the data. Dfferet sets of data wll vary relato to oe aother ad, cosequetly, ther hstograms wll look dfferet. I ths chapter, we detfy three characterstcs that wll dstgush the dstrbutos of dfferet sets of data. These are cetral locato, dsperso, ad skewess. These are characterzed Fgure A-5. Curves A ad B have the same cetral locato, but B s more dspersed. However, both A ad B are symmetrcal ad are, therefore, sad ot to be skewed. Curve C s skewed to the rght ad has a dfferet cetral locato tha A ad B. Mathematcal measures of cetral locato ad dsperso wll be dscussed later. Fgure A-5. Relatve frequecy dstrbuto showg: Curve A ad B both cetrally located, Curve B beg more dspersed tha Curve A, ad the skewess of Curve C. Trasformato of Data I most statstcal work, data that closely approxmate a partcular symmetrcal curve, called the ormal curve, are requred. Both curves A ad B Fgure A-5 are examples of ormal curves. I dealg wth skewed curves, such as C the same fgure, t s desrable to trasform the data some way so that a symmetrcal curve resemblg the ormal curve s obtaed. Referrg to the frequecy table (Table A-) ad hstogram (Fgure A-) of the data used earler, t A-9

10 ca be see that for ths set of data, the dstrbuto s skewed ( the opposte drecto as Curve C above), hece the data are ot ormally dstrbuted. The Logarthmc Trasformato Oe of the most successful ways of dervg a symmetrcal dstrbuto from a skewed dstrbuto s by expressg the orgal data terms of logarthms. The logarthms of the orgal data are gve Table A-4. Arbtrarly dvdg the logarthmc data to e class tervals, each of 0.1 ut legth, we ca prepare the logarthmc frequecy table Table A-5. As ca be see Fgure A-6, a frequecy plot of the log trasformed data more closely approxmates a symmetrcal curve tha the arthmetc plot of the orgal data. Table A-4. Logarthmc trasformato. Day Pollutat coc. Log A-10

11 Class terval Table A-5. Logarthmc frequecy table. Frequecy of occurrece Cumulatve frequecy Relatve cumulatve frequecy Probablty Graph Paper Probablty graph paper s used the aalyss of cumulatve frequecy curves; for example, the graph paper ca be used as a rough test of whether the arthmetc or the logarthmc scale best approxmates a ormal dstrbuto. The scale, arthmetc or logarthmc, o whch the cumulatve frequecy dstrbuto of the data s more early a straght le, s the oe provdg the better approxmato to a ormal dstrbuto. Plottg the cumulatve dstrbuto curve of the data above o the two scales shows that the logarthmc scale yelds the better ft (Fgure A-6). Fgure A-6. Normalzed data plot vs. o-trasformed data. A-11

12 These probablty plots ca be used, f the data are ormally dstrbuted, to estmate the mea ad stadard devato of the data. The estmate of the mea, as wll be show later, s the 50 th percetle pot, ad the estmato of the stadard devato s the dstace from the 50 th percetle to the 16 th percetle. A percetle s a measure of the relatve posto of oe of several observatos relato to all of the observatos, ad provdes a measure of relatve stadg that s useful for summarzg data. Least-Square Lear Regresso If the lear relatoshp betwee two varables s sgfcat, a least-square lear regresso le, or le of best ft, may be draw to represet the data. Ths relatoshp ca the be used to determe the value of a ukow varable. For example, f the ambet ar cocetrato s ukow, but learly related to the respose of a ambet ar motor, we ca estmate the ambet ar cocetrato based o a observed respose from the ar motor. Algebracally, a straght le has the followg form: (Eq. A-1) y mx b Where: y = depedet varable plotted o the ordate (y-axs) x = explaatory varable (depedet varable) plotted o the abscssa (x-axs) b = the pot at whch the le tercepts the y-axs at x = 0 m = slope, whch shows how much of a chage of 1 ut of x affects y Lear regresso mmzes the vertcal dstace betwee all data pots ad the straght le (Fgure A-7). Fgure A-7. Lear regresso curve. The costats m ad b for the least-square le ca be determed usg the followg equatos: A-1

13 (Eq. A-) x y xy m x x (Eq. A-3) b y mx Where: = umber of observatos y = y ; x x Example Problem Calbrato of a ambet ar aalyzer s requred before t ca be used to provde relable ambet ar cocetrato measuremets. A typcal calbrato cossts of the troducto of kow ad certfed stadard cocetratos, typcally parts per mllo (ppm) over the lear operatoal rage of the strumet, ad the recordg of the correspodg respose of the strumet uts such as volts. Based o the recorded resposes ad the kow cocetratos, a least-square lear relatoshp betwee the varables ca be calculated ad subsequetly used to determe ambet cocetratos based o the respose of the aalyzer. The followg data were collected durg a calbrato of a chemlumescet NO aalyzer. x = Cocetrato NO x (ppm) y = Istrumet respose (volts) Values for m ad b for the least-square or best ft le ca be calculated from: x, y, x, x y,, y, ad x. Soluto: x x y xy 5 x x y y A-13

14 7.33 m b The equato for ths calbrato curve would be y = 1.6x 0.10, where y (the strumet respose volts) s equal to the ambet cocetrato ppm tmes the slope of the le whch s 1.6, plus the y-tercept of x, whch s To calculate ambet cocetratos ppm, we solve the equato for x : x x ppm ppm y b m y A.5 Measures of Cetral Tedecy Arthmetc Average, or Mea A basc way of summarzg data s by the computato of a cetral value. The most commoly used cetral value statstc s the arthmetc average, or the mea. Ths statstc s partcularly useful whe appled to a set of data havg a farly symmetrcal dstrbuto. The mea s a effcet statstc that t summarzes all the data the set, ad because each pece of data s take to accout ts computato. The formula for computg the mea s: (Eq. A-4) Where: = arthmetc mea = th measuremet = total umber of observatos The arthmetc mea s ot a perfect measure of the true cetral value of a gve data set. Arthmetc meas overemphasze the mportace of oe or two extreme data pots. May measuremets of a ormally dstrbuted data set wll have a arthmetc mea that closely approxmates the true cetral value. A-14

15 Example Problem Calculate the mea of 3.0,.5,., 3.4, 3.. Soluto: Meda Whe a dstrbuto of data s asymmetrcal, such as that of Fgure A-8, t s sometmes desrable to compute a dfferet measure of cetral value. Ths secod measure, kow as the meda, s smply the mddle value of a dstrbuto, or the quatty above whch half the data le ad below whch the other half of the data le. If data are lsted ther order of magtude (from lowest to hghest), the meda s the [(+1)/] value. If the umber of data s eve, the the umercal data of the meda s the value mdway betwee the two data earest the mddle. The meda, beg a postoal value, s less flueced by extreme values a dstrbuto tha the mea. Fgure A-8. Example of a asymmetrcal dstrbuto of data (meda vs. mea). A-15

16 Example Problem Fd the meda of, 10, 15, 8, 13, 18. Soluto: The data must frst be arraged order of magtude, such as: 8, 10, 13, 15, 18, Sce = 6, the meda s the 7/ = 3.5 value, thus the meda s 14, or the value halfway betwee 13 ad 15, sce ths data set has a eve umber of measuremets. Geometrc Mea Aother measure of cetral tedecy used more specalzed applcatos s the geometrc mea ( z ). The geometrc mea s defed by usg the followg equato: (Eq. A-5)... z 1 If scetfc calculators are ot avalable, a formula that more readly leds tself to a four-fucto calculator s: 1 Log 10 z Log 10 The formula s derved as follows. Log... Log 10 z Log Where: log s to base 10 1 but 1 1 Log Log ad Log Y Log LogY Therefore: Log Log Log 1 Log... 1 Log Log A-16

17 The geometrc mea s most ofte used for data whose causes behave expoetally rather tha learly, such as the growth of bactera, measuremets that are ratos, or logormal dstrbutos. I a dstrbuto shaped lke that of Fgure A-8, the geometrc mea, lke the meda, wll yeld a value closer to the ma cluster of values tha wll the mea. The arthmetc mea s always hgher tha the geometrc mea. Example Problem Calculate the geometrc mea of 3.0,.5,., 3.4, 3.. Soluto: Log z or Log 10 z z z A.6 Measures of Dsperso Measures of cetral tedecy are more meagful f accompaed by formato o measures of dsperso. Measures of dsperso descrbe how the data spread out from the ceter. Examples of measures of dsperso a data set clude the rage, sample stadard devato, coeffcet of varato, ad the stadard geometrc devato. The Rage The easest measure of dsperso of a set of data s the dfferece betwee the maxmum ad the mmum values the set, termed the rage. The rage does ot make full use of the formato cotaed the data, sce oly two of the data pots are take to accout. Thus the rage s a useful measure of varablty for data sets of 10 or less. A-17

18 Fgure A-9. Dsperso characterstc curves. Stadard Devato The most commoly used measure of dsperso, or varablty, of sets of data s the stadard devato. Its defg formula s gve by the expresso: (Eq. A-6) s 1 Where: s = the stadard devato (always postve) = th measuremet = the mea of the data sample = the umber of observatos The expresso shows how the devato of each measuremet from the overall mea s corporated to the stadard devato. A algebracally equvalet formula that makes computato much easer s: s 1 where the varables are defed as above. A-18

19 Example Problem: Stadard Devato Usg the data provded the followg table, calculate the stadard devato: Soluto: s s s s s s s Coeffcet of Varato The coeffcet of varato (CV) s a utless measure that allows the comparso of dsperso across several sets of data. It s the stadard devato dvded by the sample mea. The CV s ofte used evrometal applcatos because varablty (expressed as stadard devato) s ofte proportoal to the mea. A-19

20 (Eq. A-7) Where: s = stadard = sample mea CV s Example Problem: Coeffcet of Varato Use the data preseted the prevous example problem to solve for the CV. CV s CV CV Stadard Geometrc Devato Dsperso of skewed data such as logormal dstrbutos s measured by the stadard geometrc devato. The stadard geometrc devato s very smlar to the stadard devato. The dsperso the log of the measuremets s measured by the geometrc stadard devato stead of the dsperso of the measuremets whch would provde a arthmetc stadard devato. The log calculato ormalzes the data to better approxmate a ormal dstrbuto. The formula for calculatg the stadard geometrc devato s: (Eq. A-8) s z atlog Where: log s to the base 10 s z = stadard geometrc devato = th measuremet = the mea of the sample log 1 log The followg formula s mathematcally detcal, yet t s much easer to use calculato: s z atlog log 1 log 1 1 A-0

21 Example Problem: Stadard Geometrc Devato Usg the data provded the followg table, calculate the stadard geometrc devato: log log log. 541 log log [.e. (.541) ] s z atlog log 1 log 1 s z atlog s z atlog 4 1 s z atlog sz atlog sz atlog 0786 s z or 1.0 A-1

22 A.7 Dstrbuto Curves Dstrbuto curves are graphcal dsplays of the dvdual data pots a data set ad are mportat because they ca detfy patters ad treds data that mght go uotced f the data were ot plotted. May types of dstrbuto curves exst: bomal, t, ch, F, ormal, ad logormal are just a few of the exstg dstrbutos. However, ar polluto measuremets, the ormal ad logormal are the most commoly occurrg oes. Thus, oly these two wll be dscussed. The Normal Dstrbuto Oe reaso the ormal (Gaussa) dstrbuto s so mportat s that a umber of atural pheomea are ormally dstrbuted or closely approxmate t. I fact, may expermets whe repeated a large umber of tmes wll approach the ormal dstrbuto curve. I ts pure form, the ormal curve s a cotuous symmetrcal, smooth curve shaped lke the oe show Fgure A-10. Naturally, a fte dstrbuto of dscrete data ca oly approxmate ths curve. The ormal curve has the followg defte relatos to the descrptve measures of a dstrbuto. Fgure A-10. Normal dstrbuto curve. A-

23 The Mea ad Meda The ormal dstrbuto curve s symmetrcal; therefore, the mea ad the meda are equal ad are foud at the ceter of the curve. Recall that, geeral, the mea ad meda of a asymmetrcal dstrbuto do ot cocde. The Rage The ormal curve rages alog the x-axs from mus fty to plus fty. Therefore, the rage of a ormal dstrbuto s fte. The Stadard Devato The stadard devato, s, becomes a most meagful measure whe related to the ormal curve. A total of 68.% of the area lyg uder a ormal curve s cluded by the part ragg from 1 stadard devato below to 1 stadard devato above the mea. A total of 95.4% les + stadard devatos from the mea ad 99.7% les wth 3 stadard devatos (Fgure A-11). By usg tables foud statstcs texts ad hadbooks, oe ca determe the area lyg uder ay part of the ormal curve. Fgure A-11. Characterstcs of the ormal dstrbuto. These areas uder the ormal dstrbuto curve ca be gve probablty terpretatos. For example, f a expermet yelds a early ormal dstrbuto wth a mea equal to 30 ad a stadard devato of 10, we ca expect about 68% of a large umber of expermetal results to rage from 0 to 40, so that the probablty of ay partcular expermetal result's havg a value betwee 0 ad 40 s about I applyg the propertes of the ormal curve to the testg of data readgs, oe ca determe whether a chage the codtos beg measured s show A-3

24 or whether oly chace fluctuatos the readgs are represeted. For a wellestablshed set of crtero data, a frequetly used set of cotrol lmts s ± 3 stadard devatos. That s, a specal vestgato of data readgs tryg these lmts ca be used to determe whether the codtos uder whch the orgal data were take have chaged. Sce the lmts of 3 stadard devatos o ether sde of the mea clude 99.7% of the area uder the ormal curve, t s very ulkely that a readg outsde these lmts s due to the codtos producg the crtero set of data. The purpose of ths techque s to separate the purely chace fluctuatos from the other causes of varato. For example, f a log seres of observatos of a evrometal measuremet yeld a mea of 50 ad a stadard devato of 10, the cotrol lmts wll be set up as 50 ± 30 - other words, ± 3 stadard devatos, or from 0 to 80. So, a value of 81 would suggest that the uderlyg codtos have chaged, ad that a large umber of smlar observatos at ths tme would yeld a dstrbuto of results wth a mea dfferet (larger) tha 50. Ths process of determg whether a value represets a sgfcat chage s closely related to the use of cotrol charts. I settg up cotrol lmts, t s ofte ecessary to dvde the avalable data to subgroups ad calculate the mea ad stadard devatos of each of these groups, makg careful ote of the codtos prevalg uder each subgroup. I collectg data to establsh cotrol lmts, as much formato as possble should be gathered about the causes ad codtos effect durg the perod of obtag a crtero set of data. Geerally, the codtos durg ths perod should be ormal, or as much cotrol as possble. I the stuato where oe takes readgs of some evrometal quatty, the appearace of data beyod the cotrol lmts mght suggest the startg of a ew data groupg to further ascerta whether the uderlyg evrometal varable has chaged. It should be kept md that the lmts of ± 3 stadard devatos are tradtoal rather tha absolute. They have bee foud through experece to be very useful may cotrol stuatos, but each expermeter must decde what lmts would be most sutable for a gve purpose by determg what levels of probablty would be eeded to quatfy acceptace ad rejecto bouds. Logormal Dstrbutos Logormal dstrbutos ca best be demostrated by meas of a example: If hourly sulfur doxde cocetratos are plotted agast frequecy of occurrece as the Data Plottg Secto, a skewed dstrbuto would exst smlar to the oe Fgure A-1. Such a curve dcates that may cocetratos are close to zero ad that few are very hgh. Ulke temperature, sulfur doxde cocetratos are blocked o the left because values less tha zero do ot exst. Because umerous ads exst for ormal dstrbutos, t s desrable to ormalze ths type of dstrbuto. By plottg the log of hourly SO cocetratos agast the frequecy of occurrece, a bell-shaped curve smlar to Fgure A-10 s obtaed. By makg ths ample ormalzg feature, all exstg ormal dstrbuto tables ca be used to make probablty terpretatos. A-4

25 Fgure A-1. Frequecy vs. cocetrato of SO. A-5

26 Refereces Ambet Motorg Techology Iformato Ceter [Iteret]. Avalable at: The Clea Ar Act. 4 USC 85, School of Publc Health, Evrometal ad Occuapatoal Health Dvso, Uversty of Illos-Chcago. Course: Itroducto to Evrometal Statstcs [Iteret]. Avalable at: U.S. Evrometal Protecto Agecy July CFR Pt. 50, Appedx A. U.S. Evrometal Protecto Agecy. 40 CFR Pt. 53. U.S. Evrometal Protecto Agecy. 40 CFR Pt. 58. U.S. Evrometal Protecto Agecy Dec Fed. Reg U.S. Evrometal Protecto Agecy Oct Fed. Reg., o. 194, pp U.S. Evrometal Protecto Agecy, Ar Polluto Trag Isttute. Course SI 100: Mathematcs Revew for Ar Polluto Cotrol. U.S. Evrometal Protecto Agecy. Gudace o systematc plag usg the Data Qualty Objectves Process. EPA QA/G-4 (EPA/40/B-06/001). U.S. Evrometal Protecto Agecy. Qualty assurace hadbook for ar polluto measuremet systems. EPA 454/R A-6

Chapter 3 Descriptive Statistics Numerical Summaries

Chapter 3 Descriptive Statistics Numerical Summaries Secto 3.1 Chapter 3 Descrptve Statstcs umercal Summares Measures of Cetral Tedecy 1. Mea (Also called the Arthmetc Mea) The mea of a data set s the sum of the observatos dvded by the umber of observatos.

More information

Bezier curves. 1. Defining a Bezier curve. A closed Bezier curve can simply be generated by closing its characteristic polygon

Bezier curves. 1. Defining a Bezier curve. A closed Bezier curve can simply be generated by closing its characteristic polygon Curve represetato Copyrght@, YZU Optmal Desg Laboratory. All rghts reserved. Last updated: Yeh-Lag Hsu (--). Note: Ths s the course materal for ME55 Geometrc modelg ad computer graphcs, Yua Ze Uversty.

More information

CS 2710 Foundations of AI Lecture 22. Machine learning. Machine Learning

CS 2710 Foundations of AI Lecture 22. Machine learning. Machine Learning CS 7 Foudatos of AI Lecture Mache learg Mlos Hauskrecht mlos@cs.ptt.edu 539 Seott Square Mache Learg The feld of mache learg studes the desg of computer programs (agets) capable of learg from past eperece

More information

APR 1965 Aggregation Methodology

APR 1965 Aggregation Methodology Sa Joaqu Valley Ar Polluto Cotrol Dstrct APR 1965 Aggregato Methodology Approved By: Sged Date: March 3, 2016 Araud Marjollet, Drector of Permt Servces Backgroud Health rsk modelg ad the collecto of emssos

More information

Review Statistics review 1: Presenting and summarising data Elise Whitley* and Jonathan Ball

Review Statistics review 1: Presenting and summarising data Elise Whitley* and Jonathan Ball Crtcal Care February Vol 6 No Whtley ad Ball Revew Statstcs revew : Presetg ad summarsg data Else Whtley* ad Joatha Ball *Lecturer Medcal Statstcs, Uversty of Brstol, Brstol, UK Lecturer Itesve Care Medce,

More information

Point Estimation-III: General Methods for Obtaining Estimators

Point Estimation-III: General Methods for Obtaining Estimators Pot Estmato-III: Geeral Methods for Obtag Estmators RECAP 0.-0.6 Data: Radom Sample from a Populato of terest o Real valued measuremets: o Assumpto (Hopefully Reasoable) o Model: Specfed Probablty Dstrbuto

More information

Descriptive Statistics: Measures of Center

Descriptive Statistics: Measures of Center Secto 2.3 Descrptve Statstcs: Measures of Ceter Frequec dstrbutos are helpful provdg formato about categorcal data, but wth umercal data we ma wat more formato. Statstc: s a umercal measure calculated

More information

1-D matrix method. U 4 transmitted. incident U 2. reflected U 1 U 5 U 3 L 2 L 3 L 4. EE 439 matrix method 1

1-D matrix method. U 4 transmitted. incident U 2. reflected U 1 U 5 U 3 L 2 L 3 L 4. EE 439 matrix method 1 -D matrx method We ca expad the smple plae-wave scatterg for -D examples that we ve see to a more versatle matrx approach that ca be used to hadle may terestg -D problems. The basc dea s that we ca break

More information

Machine Learning: Algorithms and Applications

Machine Learning: Algorithms and Applications /03/ Mache Learg: Algorthms ad Applcatos Florao Z Free Uversty of Boze-Bolzao Faculty of Computer Scece Academc Year 0-0 Lecture 3: th March 0 Naïve Bayes classfer ( Problem defto A trag set X, where each

More information

For all questions, answer choice E) NOTA" means none of the above answers is correct. A) 50,500 B) 500,000 C) 500,500 D) 1,001,000 E) NOTA

For all questions, answer choice E) NOTA means none of the above answers is correct. A) 50,500 B) 500,000 C) 500,500 D) 1,001,000 E) NOTA For all questos, aswer choce " meas oe of the above aswers s correct.. What s the sum of the frst 000 postve tegers? A) 50,500 B) 500,000 C) 500,500 D),00,000. What s the sum of the tegers betwee 00 ad

More information

A Comparison of Univariate Smoothing Models: Application to Heart Rate Data Marcus Beal, Member, IEEE

A Comparison of Univariate Smoothing Models: Application to Heart Rate Data Marcus Beal, Member, IEEE A Comparso of Uvarate Smoothg Models: Applcato to Heart Rate Data Marcus Beal, Member, IEEE E-mal: bealm@pdx.edu Abstract There are a umber of uvarate smoothg models that ca be appled to a varety of olear

More information

Beijing University of Technology, Beijing , China; Beijing University of Technology, Beijing , China;

Beijing University of Technology, Beijing , China; Beijing University of Technology, Beijing , China; d Iteratoal Coferece o Machery, Materals Egeerg, Chemcal Egeerg ad Botechology (MMECEB 5) Research of error detecto ad compesato of CNC mache tools based o laser terferometer Yuemg Zhag, a, Xuxu Chu, b

More information

ChEn 475 Statistical Analysis of Regression Lesson 1. The Need for Statistical Analysis of Regression

ChEn 475 Statistical Analysis of Regression Lesson 1. The Need for Statistical Analysis of Regression Statstcal-Regresso_hadout.xmcd Statstcal Aalss of Regresso ChE 475 Statstcal Aalss of Regresso Lesso. The Need for Statstcal Aalss of Regresso What do ou do wth dvdual expermetal data pots? How are the

More information

ANALYSIS OF VARIANCE WITH PARETO DATA

ANALYSIS OF VARIANCE WITH PARETO DATA Proceedgs of the th Aual Coferece of Asa Pacfc Decso Sceces Isttute Hog Kog, Jue -8, 006, pp. 599-609. ANALYSIS OF VARIANCE WITH PARETO DATA Lakhaa Watthaacheewakul Departmet of Mathematcs ad Statstcs,

More information

APPLICATION OF CLUSTERING METHODS IN BANK S PROPENSITY MODEL

APPLICATION OF CLUSTERING METHODS IN BANK S PROPENSITY MODEL APPLICATION OF CLUSTERING METHODS IN BANK S PROPENSITY MODEL Sergej Srota Haa Řezaková Abstract Bak s propesty models are beg developed for busess support. They should help to choose clets wth a hgher

More information

Eight Solved and Eight Open Problems in Elementary Geometry

Eight Solved and Eight Open Problems in Elementary Geometry Eght Solved ad Eght Ope Problems Elemetary Geometry Floret Smaradache Math & Scece Departmet Uversty of New Mexco, Gallup, US I ths paper we revew eght prevous proposed ad solved problems of elemetary

More information

Estimation of Co-efficient of Variation in PPS sampling.

Estimation of Co-efficient of Variation in PPS sampling. It. Statstcal Ist.: Proc. 58th World Statstcal Cogress, 0, Dubl (Sesso CPS00) p.409 Estmato of Co-effcet of Varato PPS samplg. Archaa. V ( st Author) Departmet of Statstcs, Magalore Uverst Magalagagotr,

More information

Clustering documents with vector space model using n-grams

Clustering documents with vector space model using n-grams Clusterg documets wth vector space model usg -grams Klas Skogmar, d97ksk@efd.lth.se Joha Olsso, d97jo@efd.lth.se Lud Isttute of Techology Supervsed by: Perre Nugues, Perre.Nugues@cs.lth.se Abstract Ths

More information

Optimal Allocation of Complex Equipment System Maintainability

Optimal Allocation of Complex Equipment System Maintainability Optmal Allocato of Complex Equpmet System ataablty X Re Graduate School, Natoal Defese Uversty, Bejg, 100091, Cha edcal Protecto Laboratory, Naval edcal Research Isttute, Shagha, 200433, Cha Emal:rexs841013@163.com

More information

LP: example of formulations

LP: example of formulations LP: eample of formulatos Three classcal decso problems OR: Trasportato problem Product-m problem Producto plag problem Operatos Research Massmo Paolucc DIBRIS Uversty of Geova Trasportato problem The decso

More information

Blind Steganalysis for Digital Images using Support Vector Machine Method

Blind Steganalysis for Digital Images using Support Vector Machine Method 06 Iteratoal Symposum o Electrocs ad Smart Devces (ISESD) November 9-30, 06 Bld Stegaalyss for Dgtal Images usg Support Vector Mache Method Marcelus Hery Meor School of Electrcal Egeerg ad Iformatcs Badug

More information

Face Recognition using Supervised & Unsupervised Techniques

Face Recognition using Supervised & Unsupervised Techniques Natoal Uversty of Sgapore EE5907-Patter recogto-2 NAIONAL UNIVERSIY OF SINGAPORE EE5907 Patter Recogto Project Part-2 Face Recogto usg Supervsed & Usupervsed echques SUBMIED BY: SUDEN NAME: harapa Reddy

More information

Fitting. We ve learned how to detect edges, corners, blobs. Now what? We would like to form a. compact representation of

Fitting. We ve learned how to detect edges, corners, blobs. Now what? We would like to form a. compact representation of Fttg Fttg We ve leared how to detect edges, corers, blobs. Now what? We would lke to form a hgher-level, h l more compact represetato of the features the mage b groupg multple features accordg to a smple

More information

ITEM ToolKit Technical Support Notes

ITEM ToolKit Technical Support Notes ITEM ToolKt Notes Fault Tree Mathematcs Revew, Ic. 2875 Mchelle Drve Sute 300 Irve, CA 92606 Phoe: +1.240.297.4442 Fax: +1.240.297.4429 http://www.itemsoft.com Page 1 of 15 6/1/2016 Copyrght, Ic., All

More information

Process Quality Evaluation based on Maximum Entropy Principle. Yuhong Wang, Chuanliang Zhang, Wei Dai a and Yu Zhao

Process Quality Evaluation based on Maximum Entropy Principle. Yuhong Wang, Chuanliang Zhang, Wei Dai a and Yu Zhao Appled Mechacs ad Materals Submtted: 204-08-26 ISSN: 662-7482, Vols. 668-669, pp 625-628 Accepted: 204-09-02 do:0.4028/www.scetfc.et/amm.668-669.625 Ole: 204-0-08 204 Tras Tech Publcatos, Swtzerlad Process

More information

MATHEMATICAL PROGRAMMING MODEL OF THE CRITICAL CHAIN METHOD

MATHEMATICAL PROGRAMMING MODEL OF THE CRITICAL CHAIN METHOD MATHEMATICAL PROGRAMMING MODEL OF THE CRITICAL CHAIN METHOD TOMÁŠ ŠUBRT, PAVLÍNA LANGROVÁ CUA, SLOVAKIA Abstract Curretly there s creasgly dcated that most of classcal project maagemet methods s ot sutable

More information

Eight Solved and Eight Open Problems in Elementary Geometry

Eight Solved and Eight Open Problems in Elementary Geometry Eght Solved ad Eght Ope Problems Elemetary Geometry Floret Smaradache Math & Scece Departmet Uversty of New Mexco, Gallup, US I ths paper we revew eght prevous proposed ad solved problems of elemetary

More information

A hybrid method using FAHP and TOPSIS for project selection Xuan Lia, Jiang Jiangb and Su Deng c

A hybrid method using FAHP and TOPSIS for project selection Xuan Lia, Jiang Jiangb and Su Deng c 5th Iteratoal Coferece o Computer Sceces ad Automato Egeerg (ICCSAE 205) A hybrd method usg FAHP ad TOPSIS for project selecto Xua La, Jag Jagb ad Su Deg c College of Iformato System ad Maagemet, Natoal

More information

Nine Solved and Nine Open Problems in Elementary Geometry

Nine Solved and Nine Open Problems in Elementary Geometry Ne Solved ad Ne Ope Problems Elemetary Geometry Floret Smaradache Math & Scece Departmet Uversty of New Mexco, Gallup, US I ths paper we revew e prevous proposed ad solved problems of elemetary D geometry

More information

Differentiated Service of Streaming Media Playback Technology

Differentiated Service of Streaming Media Playback Technology Iteratoal Coferece o Advaced Iformato ad Commucato Techology for Educato (ICAICTE 2013) Dfferetated Servce of Streamg Meda Playback Techology CHENG Z-ao 1 MENG Bo 1 WANG Da-hua 1 ZHAO Yue 1 1 Iformatzato

More information

Area and Power Efficient Modulo 2^n+1 Multiplier

Area and Power Efficient Modulo 2^n+1 Multiplier Iteratoal Joural of Moder Egeerg Research (IJMER) www.jmer.com Vol.3, Issue.3, May-Jue. 013 pp-137-1376 ISSN: 49-6645 Area ad Power Effcet Modulo ^+1 Multpler K. Ptambar Patra, 1 Saket Shrvastava, Sehlata

More information

Machine Learning. CS 2750 Machine Learning. Administration. Lecture 1. Milos Hauskrecht 5329 Sennott Square, x4-8845

Machine Learning. CS 2750 Machine Learning. Administration. Lecture 1. Milos Hauskrecht 5329 Sennott Square, x4-8845 CS 75 Mache Learg Lecture Mache Learg Mlos Hauskrecht mlos@cs.ptt.edu 539 Seott Square, 5 people.cs.ptt.edu/~mlos/courses/cs75/ Admstrato Istructor: Prof. Mlos Hauskrecht mlos@cs.ptt.edu 539 Seott Square,

More information

COMSC 2613 Summer 2000

COMSC 2613 Summer 2000 Programmg II Fal Exam COMSC 63 Summer Istructos: Name:. Prt your ame the space provded Studet Id:. Prt your studet detfer the space Secto: provded. Date: 3. Prt the secto umber of the secto whch you are

More information

COMPARISON OF PARAMETERIZATION METHODS USED FOR B-SPLINE CURVE INTERPOLATION

COMPARISON OF PARAMETERIZATION METHODS USED FOR B-SPLINE CURVE INTERPOLATION Europea Joural of Techc COMPARISON OF PARAMETERIZATION METHODS USED FOR B-SPLINE CURVE INTERPOLATION Sıtı ÖZTÜRK, Cegz BALTA, Melh KUNCAN 2* Kocael Üverstes, Mühedsl Faültes, Eletro ve Haberleşme Mühedslğ

More information

Performance Impact of Load Balancers on Server Farms

Performance Impact of Load Balancers on Server Farms erformace Impact of Load Balacers o Server Farms Ypg Dg BMC Software Server Farms have gaed popularty for provdg scalable ad relable computg / Web servces. A load balacer plays a key role ths archtecture,

More information

Journal of Chemical and Pharmaceutical Research, 2015, 7(3): Research Article

Journal of Chemical and Pharmaceutical Research, 2015, 7(3): Research Article Avalable ole www.ocpr.com Joural of Chemcal ad Pharmaceutcal Research, 2015, 73):476-481 Research Artcle ISSN : 0975-7384 CODENUSA) : JCPRC5 Research o cocept smlarty calculato method based o sematc grd

More information

COMBINATORIAL METHOD OF POLYNOMIAL EXPANSION OF SYMMETRIC BOOLEAN FUNCTIONS

COMBINATORIAL METHOD OF POLYNOMIAL EXPANSION OF SYMMETRIC BOOLEAN FUNCTIONS COMBINATORIAL MTHOD O POLYNOMIAL XPANSION O SYMMTRIC BOOLAN UNCTIONS Dala A. Gorodecky The Uted Isttute of Iformatcs Prolems of Natoal Academy of Sceces of Belarus, Msk,, Belarus, dala.gorodecky@gmal.com.

More information

On a Sufficient and Necessary Condition for Graph Coloring

On a Sufficient and Necessary Condition for Graph Coloring Ope Joural of Dscrete Matheatcs, 04, 4, -5 Publshed Ole Jauary 04 (http://wwwscrporg/joural/ojd) http://dxdoorg/0436/ojd04400 O a Suffcet ad Necessary Codto for raph Colorg Maodog Ye Departet of Matheatcs,

More information

Regression Analysis. Acknowledgments

Regression Analysis. Acknowledgments PT 3 - Lear Regresso Regresso Aalyss How to develop ad assess a CER All models are wrog, but some are useful. -George Box I mathematcs, cotext obscures structure. I data aalyss, cotext provdes meag. -George

More information

Office Hours. COS 341 Discrete Math. Office Hours. Homework 8. Currently, my office hours are on Friday, from 2:30 to 3:30.

Office Hours. COS 341 Discrete Math. Office Hours. Homework 8. Currently, my office hours are on Friday, from 2:30 to 3:30. Oce Hours Curretly, my oce hours are o Frday, rom :30 to 3:30. COS 31 Dscrete Math 1 Oce Hours Curretly, my oce hours are o Frday, rom :30 to 3:30. Nobody seems to care. Chage oce hours? Tuesday, 8 PM

More information

NEURO FUZZY MODELING OF CONTROL SYSTEMS

NEURO FUZZY MODELING OF CONTROL SYSTEMS NEURO FUZZY MODELING OF CONTROL SYSTEMS Efré Gorrosteta, Carlos Pedraza Cetro de Igeería y Desarrollo Idustral CIDESI, Av Pe de La Cuesta 70. Des. Sa Pablo. Querétaro, Qro, Méxco gorrosteta@teso.mx pedraza@cdes.mx

More information

TDT-2004: ADAPTIVE TOPIC TRACKING AT MARYLAND

TDT-2004: ADAPTIVE TOPIC TRACKING AT MARYLAND TDT-2004: ADAPTIVE TOPIC TRACKING AT MARYLAND Tamer Elsayed, Douglas W. Oard, Davd Doerma Isttute for Advaced r Studes Uversty of Marylad, College Park, MD 20742 Cotact author: telsayed@cs.umd.edu Gary

More information

Two step approach for Software Process Control: HLSRGM

Two step approach for Software Process Control: HLSRGM Iteratoal Joural of Emergg Treds & Techology Computer Scece (IJETTCS Web Ste: wwwjettcsorg Emal: edtor@jettcsorg, edtorjettcs@gmalcom Volume, Issue 4, July August 03 ISS 78-686 Two step approach for Software

More information

OMAE HOW TO CARRY OUT METOCEAN STUDIES

OMAE HOW TO CARRY OUT METOCEAN STUDIES Proceedgs of the ASME 20 30th Iteratoal Coferece o Ocea, Offshore ad Arctc Egeerg OMAE20 Jue 9-24, 20, Rotterdam, The Netherlads OMAE20-490 HOW TO CARRY OUT METOCEAN STUDIES Judth va Os Hydraulc Egeerg

More information

International Mathematical Forum, 1, 2006, no. 31, ON JONES POLYNOMIALS OF GRAPHS OF TORUS KNOTS K (2, q ) Tamer UGUR, Abdullah KOPUZLU

International Mathematical Forum, 1, 2006, no. 31, ON JONES POLYNOMIALS OF GRAPHS OF TORUS KNOTS K (2, q ) Tamer UGUR, Abdullah KOPUZLU Iteratoal Mathematcal Forum,, 6, o., 57-54 ON JONES POLYNOMIALS OF RAPHS OF TORUS KNOTS K (, q ) Tamer UUR, Abdullah KOPUZLU Atatürk Uverst Scece Facult Dept. of. Math. 54 Erzurum, Turkey tugur@atau.edu.tr

More information

CLUSTERING ASSISTED FUNDAMENTAL MATRIX ESTIMATION

CLUSTERING ASSISTED FUNDAMENTAL MATRIX ESTIMATION CLUSERING ASSISED FUNDAMENAL MARIX ESIMAION Hao Wu ad Y Wa School of Iformato Scece ad Egeerg, Lazhou Uversty, Cha wuhao1195@163com, wayjs@163com ABSRAC I computer vso, the estmato of the fudametal matrx

More information

Reliable Surface Extraction from Point-Clouds using Scanner-Dependent Parameters

Reliable Surface Extraction from Point-Clouds using Scanner-Dependent Parameters 1 Relable Surface Extracto from Pot-Clouds usg Scaer-Depedet Parameters Hrosh Masuda 1, Ichro Taaka 2, ad Masakazu Eomoto 3 1 The Uversty of Tokyo, masuda@sys.t.u-tokyo.ac.jp 2 Tokyo Dek Uversty, taaka@cck.deda.ac.jp

More information

Process Capability Analysis by Using Statistical Process Control of Rice Polished Cylinder Turning Practice

Process Capability Analysis by Using Statistical Process Control of Rice Polished Cylinder Turning Practice World Academy of Scece, Egeerg ad Techology Iteratoal Joural of Idustral ad Maufacturg Egeerg Vol:8, No:, 04 Process Capablty Aalyss by Usg Statstcal Process Cotrol of ce Polshed Cylder Turg Practce S.

More information

A Genetic K-means Clustering Algorithm Applied to Gene Expression Data

A Genetic K-means Clustering Algorithm Applied to Gene Expression Data A Geetc K-meas Clusterg Algorthm Appled to Gee Expresso Data Fag-Xag Wu, W. J. Zhag, ad Athoy J. Kusal Dvso of Bomedcal Egeerg, Uversty of Sasatchewa, Sasatoo, S S7N 5A9, CANADA faw34@mal.usas.ca, zhagc@egr.usas.ca

More information

Prof. Feng Liu. Winter /24/2019

Prof. Feng Liu. Winter /24/2019 Prof. Feg Lu Wter 209 http://www.cs.pd.edu/~flu/courses/cs40/ 0/24/209 Last Tme Feature detecto 2 Toda Feature matchg Fttg The followg sldes are largel from Prof. S. Lazebk 3 Wh etract features? Motvato:

More information

Dhaka University of Engineering & Technology Gazipur-1700, Bangladesh. b Department of Physics

Dhaka University of Engineering & Technology Gazipur-1700, Bangladesh. b Department of Physics Egeerg e-trasacto (ISSN 183-6379 Vol., No.1, Jue 7, pp 15-19 Ole at http//ejum.fsktm.um.edu.my Receved 1 Ja, 7; Accepted 17Aprl, 7 MEASUREMENT OF MANUFATURING PROESS APABILITY A ASE STUDY M. Mostaqur Rahma

More information

Simulator for Hydraulic Excavator

Simulator for Hydraulic Excavator Smulator for Hydraulc Excavator Tae-Hyeog Lm*, Hog-Seo Lee ** ad Soo-Yog Yag *** * Departmet of Mechacal ad Automotve Egeerg, Uversty of Ulsa,Ulsa, Korea (Tel : +82-52-259-273; E-mal: bulbaram@mal.ulsa.ac.kr)

More information

A SOFTWARE QUALITY EVALUATION METHOD BASED ON THE PRINCIPLE OF MAXIMUM COORDINATION AND SUBORDINATION

A SOFTWARE QUALITY EVALUATION METHOD BASED ON THE PRINCIPLE OF MAXIMUM COORDINATION AND SUBORDINATION Joural of Theoretcal ad Appled Iformato Techology 1 th Jauary 213. Vol. 47 No.1 25-213 JATIT & LLS. All rghts reserved. ISSN: 1992-8645 www.att.org E-ISSN: 1817-3195 A SOFTWARE QUALITY EVALUATION METHOD

More information

Enumerating XML Data for Dynamic Updating

Enumerating XML Data for Dynamic Updating Eumeratg XML Data for Dyamc Updatg Lau Ho Kt ad Vcet Ng Departmet of Computg, The Hog Kog Polytechc Uversty, Hug Hom, Kowloo, Hog Kog cstyg@comp.polyu.edu.h Abstract I ths paper, a ew mappg model, called

More information

MINIMIZATION OF THE VALUE OF DAVIES-BOULDIN INDEX

MINIMIZATION OF THE VALUE OF DAVIES-BOULDIN INDEX MIIMIZATIO OF THE VALUE OF DAVIES-BOULDI IDEX ISMO ÄRÄIE ad PASI FRÄTI Departmet of Computer Scece, Uversty of Joesuu Box, FI-800 Joesuu, FILAD ABSTRACT We study the clusterg problem whe usg Daves-Bould

More information

Spatial Interpolation Using Neural Fuzzy Technique

Spatial Interpolation Using Neural Fuzzy Technique Wog, K.W., Gedeo, T., Fug, C.C. ad Wog, P.M. (00) Spatal terpolato usg eural fuzzy techque. I: Proceedgs of the 8th Iteratoal Coferece o Neural Iformato Processg (ICONIP), Shagha, Cha Spatal Iterpolato

More information

Outline. Area objects and spatial autocorrelation. Types of area object

Outline. Area objects and spatial autocorrelation. Types of area object Area objects ad spatal autocorrelato Outle Itroducto Geometrc propertes of areas Spatal autocorrelato: jos cout approach Spatal autocorrelato: Mora s I Spatal autocorrelato: Geary s C Spatal autocorrelato:

More information

Software reliability is defined as the probability of failure

Software reliability is defined as the probability of failure Evolutoary Regresso Predcto for Software Cumulatve Falure Modelg: a comparatve study M. Beaddy, M. Wakrm & S. Aljahdal 2 : Dept. of Math. & Ifo. Equpe MMS, Ib Zohr Uversty Morocco. beaddym@yahoo.fr 2:

More information

Transistor/Gate Sizing Optimization

Transistor/Gate Sizing Optimization Trasstor/Gate Szg Optmzato Gve: Logc etwork wth or wthout cell lbrary Fd: Optmal sze for each trasstor/gate to mmze area or power, both uder delay costrat Statc szg: based o tmg aalyss ad cosder all paths

More information

2 General Regression Neural Network (GRNN)

2 General Regression Neural Network (GRNN) 4 Geeral Regresso Neural Network (GRNN) GRNN, as proposed b oald F. Specht [Specht 9] falls to the categor of probablstc eural etworks as dscussed Chapter oe. Ths eural etwork lke other probablstc eural

More information

Network Security Evaluation Based on Variable Weight Fuzzy Cloud Model

Network Security Evaluation Based on Variable Weight Fuzzy Cloud Model 207 2 d Iteratoal Coferece o Computer Scece ad Techology (CST 207) ISBN: 978--60595-46-5 Networ Securty Evaluato Based o Varable Weght Fuzzy Cloud Model Yag JIANG a*, Cheg-ha LI, Zh-peg LI ad Mg-ca SUN

More information

New Fuzzy Integral for the Unit Maneuver in RTS Game

New Fuzzy Integral for the Unit Maneuver in RTS Game New Fuzzy Itegral for the Ut Maeuver RTS Game Peter Hu Fug Ng, YgJe L, ad Smo Ch Keug Shu Departmet of Computg, The Hog Kog Polytechc Uversty, Hog Kog {cshfg,csyjl,csckshu}@comp.polyu.edu.hk Abstract.

More information

GUI Simulation Platform for RFID Indoor Tracking System

GUI Simulation Platform for RFID Indoor Tracking System Sesors & Trasducers 2014 by IFSA Publshg, S. L. http://www.sesorsportal.com GUI Smulato Platform for RFID Idoor Trackg System 1 Be-Be Mao, 2 JIN Xue-Bo School of Computer ad Iformato Egeerg, Bejg Techology

More information

Fingerprint Classification Based on Spectral Features

Fingerprint Classification Based on Spectral Features Fgerprt Classfcato Based o Spectral Features Hosse Pourghassem Tarbat Modares Uversty h_poorghasem@modares.ac.r Hassa Ghassema Tarbat Modares Uversty ghassem@modares.ac.r Abstract: Fgerprt s oe of the

More information

A Perception of Statistical Inference in Data Mining

A Perception of Statistical Inference in Data Mining Iteratoal Joural of Computer Scece & Commucato Vol., No., July-December 00, pp. 373-378 A Percepto of Statstcal Iferece Data Mg Sajay Gaur & M. S. Dulawat, Departmet of Mathematcs & Statstcs, Mohalal Sukhada

More information

Effective Steganalysis Based on Statistical Moments of Wavelet Characteristic Function

Effective Steganalysis Based on Statistical Moments of Wavelet Characteristic Function Effectve Stegaalyss Based o Statstcal Momets of Wavelet Characterstc Fucto Yu Q. Sh 1, Guorog Xua, Chegyu Yag, Jaog Gao, Zhepg Zhag, Peq Cha, Deku Zou 1, Chuhua Che 1, We Che 1 1 New Jersey Isttute of

More information

A New Hybrid Audio Classification Algorithm Based on SVM Weight Factor and Euclidean Distance

A New Hybrid Audio Classification Algorithm Based on SVM Weight Factor and Euclidean Distance Proceedgs of the 2007 WSEAS Iteratoal Coferece o Computer Egeerg ad Applcatos, Gold Coast, Australa, Jauary 7-9, 2007 52 A New Hybrd Audo Classfcato Algorthm Based o SVM Weght Factor ad Eucldea Dstace

More information

QUADRATURE POINTS ON POLYHEDRAL ELEMENTS

QUADRATURE POINTS ON POLYHEDRAL ELEMENTS QUADRATURE POINTS ON POLYHEDRAL ELEMENTS Tobas Pck, Peter Mlbradt 2 ABSTRACT The method of the fte elemets s a flexble umerc procedure both for terpolato ad approxmato of the solutos of partal dfferetal

More information

DEEP (Displacement Estimation Error Back-Propagation) Method for Cascaded ViSPs (Visually Servoed Paired Structured Light Systems)

DEEP (Displacement Estimation Error Back-Propagation) Method for Cascaded ViSPs (Visually Servoed Paired Structured Light Systems) DEEP (Dsplacemet Estmato Error Back-Propagato) Method for Cascaded VSPs (Vsually Servoed Pared Structured Lght Systems) Haem Jeo 1), Jae-Uk Sh 2), Wachoel Myeog 3), Yougja Km 4), ad *Hyu Myug 5) 1), 3),

More information

Vanishing Point Detection: Representation Analysis and New Approaches

Vanishing Point Detection: Representation Analysis and New Approaches Publshed the Proceedgs of the th Iteratoal Coferece o Image Aalyss ad Processg (ICIAP ). IEEE. Persoal use of ths materal s permtted. However, permsso to reprt/republsh ths materal for advertsg or promotoal

More information

Cubic fuzzy H-ideals in BF-Algebras

Cubic fuzzy H-ideals in BF-Algebras OSR Joural of Mathematcs (OSR-JM) e-ssn: 78-578 p-ssn: 39-765X Volume ssue 5 Ver (Sep - Oct06) PP 9-96 wwwosrjouralsorg Cubc fuzzy H-deals F-lgebras Satyaarayaa Esraa Mohammed Waas ad U du Madhav 3 Departmet

More information

Multi-dimensional Characteristics Analysis of Decathlon Champion Achievement of Modern Olympic Game

Multi-dimensional Characteristics Analysis of Decathlon Champion Achievement of Modern Olympic Game Iteratoal Workshop o Computer Scece Sports (IWCSS 203) Mult-dmesoal Characterstcs Aalyss of Decathlo Champo Achevemet of Moder Olympc Game Chagme Huag Isttute of Physcal Educato Hua Uversty of Scece ad

More information

Mode Changes in Priority Pre-emptively Scheduled Systems. K. W. Tindell, A. Burns, A. J. Wellings

Mode Changes in Priority Pre-emptively Scheduled Systems. K. W. Tindell, A. Burns, A. J. Wellings ode hages rorty re-emptvely Scheduled Systems. W. dell, A. Burs, A.. Wellgs Departmet of omputer Scece, Uversty of York, Eglad Abstract may hard real tme systems the set of fuctos that a system s requred

More information

Region Matching by Optimal Fuzzy Dissimilarity

Region Matching by Optimal Fuzzy Dissimilarity Rego Matchg by Optmal Fuzzy Dssmlarty Zhaggu Zeg, Ala Fu ad Hog Ya School of Electrcal ad formato Egeerg The Uversty of Sydey Phoe: +6--935-6659 Fax: +6--935-3847 Emal: zzeg@ee.usyd.edu.au Abstract: Ths

More information

Multi-modal Image Registration by Quantitative-Qualitative Measure of Mutual Information (Q-MI) *

Multi-modal Image Registration by Quantitative-Qualitative Measure of Mutual Information (Q-MI) * Mult-modal Image Regstrato by Quattatve-Qualtatve Measure of Mutual Iformato (Q-MI) * Hogxa Lua 1, Fehu Q 1, ad Dggag She 2 1 Departmet of Computer Scece ad Egeerg, Shagha Jao Tog Uversty, Shagha, Cha

More information

A Framework for Block-Based Timing Sensitivity Analysis

A Framework for Block-Based Timing Sensitivity Analysis 39.3 Framework for Block-Based Tmg Sestvty alyss Sajay V. Kumar Chadramoul V. Kashyap Sach S. Sapatekar Uversty of Mesota Itel Corporato Uversty of Mesota Meapols MN 55455 Hllsboro OR 973 Meapols MN 55455

More information

A Disk-Based Join With Probabilistic Guarantees*

A Disk-Based Join With Probabilistic Guarantees* A Dsk-Based Jo Wth Probablstc Guaratees* Chrstopher Jermae, Al Dobra, Subramaa Arumugam, Shatau Josh, Abhjt Pol Computer ad Iformato Sceces ad Egeerg Departmet Uversty of Florda, Gaesvlle {cjerma, adobra,

More information

NUMERICAL INTEGRATION BY GENETIC ALGORITHMS. Vladimir Morozenko, Irina Pleshkova

NUMERICAL INTEGRATION BY GENETIC ALGORITHMS. Vladimir Morozenko, Irina Pleshkova 5 Iteratoal Joural Iformato Theores ad Applcatos, Vol., Number 3, 3 NUMERICAL INTEGRATION BY GENETIC ALGORITHMS Vladmr Morozeko, Ira Pleshkova Abstract: It s show that geetc algorthms ca be used successfully

More information

Classification Web Pages By Using User Web Navigation Matrix By Mementic Algorithm

Classification Web Pages By Using User Web Navigation Matrix By Mementic Algorithm Classfcato Web Pages By Usg User Web Navgato Matrx By Memetc Algorthm 1 Parvaeh roustae 2 Mehd sadegh zadeh 1 Studet of Computer Egeerg Software EgeergDepartmet of ComputerEgeerg, Bushehr brach,

More information

Unsupervised Discretization Using Kernel Density Estimation

Unsupervised Discretization Using Kernel Density Estimation Usupervsed Dscretzato Usg Kerel Desty Estmato Maregle Bba, Floraa Esposto, Stefao Ferll, Ncola D Mauro, Teresa M.A Basle Departmet of Computer Scece, Uversty of Bar Va Oraboa 4, 7025 Bar, Italy {bba,esposto,ferll,dm,basle}@d.uba.t

More information

Clustered Signatures for Modeling and Recognizing 3D Rigid Objects

Clustered Signatures for Modeling and Recognizing 3D Rigid Objects World Academy of Scece, Egeerg ad Techology 4 008 Clustered Sgatures for Modelg ad Recogzg 3D Rgd Obects H. B. Darbad, M. R. Ito, ad J. Lttle Abstract Ths paper descrbes a probablstc method for three-dmesoal

More information

Automated approach for the surface profile measurement of moving objects based on PSP

Automated approach for the surface profile measurement of moving objects based on PSP Uversty of Wollogog Research Ole Faculty of Egeerg ad Iformato Sceces - Papers: Part B Faculty of Egeerg ad Iformato Sceces 207 Automated approach for the surface profle measuremet of movg objects based

More information

Estimating Feasibility Using Multiple Surrogates and ROC Curves

Estimating Feasibility Using Multiple Surrogates and ROC Curves Estmatg Feasblty Usg Multple Surrogates ad ROC Curves Arba Chaudhur * Uversty of Florda, Gaesvlle, Florda, 3601 Rodolphe Le Rche École Natoale Supéreure des Mes de Sat-Étee, Sat-Étee, Frace ad CNRS LIMOS

More information

RESEARCH ON SPATIAL INTERRELATIONS OF GEOMETRIC DEVIATIONS DETERMINED IN COORDINATE MEASUREMENTS OF FREE-FORM SURFACES

RESEARCH ON SPATIAL INTERRELATIONS OF GEOMETRIC DEVIATIONS DETERMINED IN COORDINATE MEASUREMENTS OF FREE-FORM SURFACES Metrol. Meas. Syst. Vol. XVI (2009), No 3, pp. 50-50 METROLOGY AND MEASUREMENT SYSTEMS Idex 330930, ISSN 0860-8229 www.metrology.pg.gda.pl RESEARCH ON SPATIAL INTERRELATIONS OF GEOMETRIC DEVIATIONS DETERMINED

More information

Active Bayesian Learning For Mixture Models

Active Bayesian Learning For Mixture Models Actve Bayesa Learg For Mxture Models Ia Davdso Slco Graphcs 300 Crttede L, MS 876 Mouta Vew, CA 94587 pd@hotmal.com Abstract Tradtoally, Bayesa ductve learg volves fdg the most probable model from the

More information

A PROCEDURE FOR SOLVING INTEGER BILEVEL LINEAR PROGRAMMING PROBLEMS

A PROCEDURE FOR SOLVING INTEGER BILEVEL LINEAR PROGRAMMING PROBLEMS ISSN: 39-8753 Iteratoal Joural of Iovatve Research Scece, Egeerg ad Techology A ISO 397: 7 Certfed Orgazato) Vol. 3, Issue, Jauary 4 A PROCEDURE FOR SOLVING INTEGER BILEVEL LINEAR PROGRAMMING PROBLEMS

More information

Application Research for Ultrasonic Flaw Identification Based on Support Vector Machine Jing Huang 1, a, Binglei Guan 1, b

Application Research for Ultrasonic Flaw Identification Based on Support Vector Machine Jing Huang 1, a, Binglei Guan 1, b 4th Iteratoal Coferece o Mechatrocs, Materals, Chemstry ad Computer Egeerg (ICMMCCE 205) Applcato Research for Ultrasoc Flaw Idetfcato Based o Support Vector Mache Jg Huag, a, Bgle Gua, b School of Electroc

More information

Method to reduce the effect of miagrafic and sensory noise with isolating the isoline on ECG signal

Method to reduce the effect of miagrafic and sensory noise with isolating the isoline on ECG signal MATEC Web of Cofereces 1, 05017 (017 DTS-017 DOI: 10.1051/mateccof/017105017 Method to reduce the effect of marafc ad sesory ose wth solat the sole o ECG sal Evey Semeshchev 1,*, Dmtry Cheryshov 1, Ilya

More information

Application of Genetic Algorithm for Computing a Global 3D Scene Exploration

Application of Genetic Algorithm for Computing a Global 3D Scene Exploration Joural of Software Egeerg ad Applcatos, 2011, 4, 253-258 do:10.4236/jsea.2011.44028 Publshed Ole Aprl 2011 (http://www.scrp.org/joural/jsea) 253 Applcato of Geetc Algorthm for Computg a Global 3D Scee

More information

FRAMELET-BASED MULTIRESOLUTION IMAGE FUSION WITH AN IMPROVED INTENSITY-HUE-SATURATION TRANSFORM

FRAMELET-BASED MULTIRESOLUTION IMAGE FUSION WITH AN IMPROVED INTENSITY-HUE-SATURATION TRANSFORM FRAMELET-BASED MULTIRESOLUTION IMAGE FUSION WITH AN IMPROVED INTENSITY-HUE-SATURATION TRANSFORM M. J. Cho *, D. H. Lee, H. S. Lm Satellte Iformato Research Isttute, KARI, 45, Eoeu-dog, Yuseog-gu, Daejeo,

More information

ABSTRACT Keywords

ABSTRACT Keywords A Preprocessg Scheme for Hgh-Cardalty Categorcal Attrbutes Classfcato ad Predcto Problems Daele Mcc-Barreca ClearCommerce Corporato 1100 Metrc Blvd. Aust, TX 78732 ABSTRACT Categorcal data felds characterzed

More information

A modified Logic Scoring Preference method for dynamic Web services evaluation and selection

A modified Logic Scoring Preference method for dynamic Web services evaluation and selection A modfed Logc Scorg Preferece method for dyamc Web servces evaluato ad selecto Hog Qg Yu ad Herá Mola 2 Departmet of Computer Scece, Uversty of Lecester, UK hqy@mcs.le.ac.uk 2 Departmet of Iformatcs, School

More information

A Novel Clustering Algorithm Based on Graph Matching

A Novel Clustering Algorithm Based on Graph Matching JOURNAL OF SOFTWARE, VOL. 8, NO. 4, APRIL 203 035 A Novel Clusterg Algorthm Based o raph Matchg uoyua L School of Computer Scece ad Techology, Cha Uversty of Mg ad Techology, Xuzhou, Cha State Key Laboratory

More information

Summary of Curve Smoothing Technology Development

Summary of Curve Smoothing Technology Development RESEARCH ARTICLE Summary of Curve Smoothg Techology Developmet Wu Yze,ZhagXu,Jag Mgyag (College of Mechacal Egeerg, Shagha Uversty of Egeerg Scece, Shagha, Cha) Abstract: Wth the cotuous developmet of

More information

DESIGN AND EVALUATION OF EXPERIMENTS WITH SAS

DESIGN AND EVALUATION OF EXPERIMENTS WITH SAS XIX IMEKO World Cogress Fudametal ad ppled Metrology September 6, 009, Lsbo, Portugal DESIGN ND EVLUTION OF EXPERIMENTS WITH draa Horíová Uversty of Ecoomcs Bratslava (Faculty of Ecoomcs Iformatcs, Isttute

More information

Multiple fault diagnosis of down-hole conditions of sucker-rod pumping wells based on Freeman chain code and DCA

Multiple fault diagnosis of down-hole conditions of sucker-rod pumping wells based on Freeman chain code and DCA Pet.Sc.(3):37-3 DOI.7/s8-3-83-37 Multple fault dagoss of dow-hole codtos of sucker-rod pumpg wells based o Freema cha code ad DCA LI Ku, GAO Xa-we, YANG We-bg, DAI Yg-log ad TIAN Zhog-da College of Iformato

More information

Reconstruction of Orthogonal Polygonal Lines

Reconstruction of Orthogonal Polygonal Lines Recostructo of Orthogoal Polygoal Les Alexader Grbov ad Eugee Bodasky Evrometal System Research Isttute (ESRI) 380 New ork St. Redlads CA 9373-800 USA {agrbov ebodasky}@esr.com Abstract. A orthogoal polygoal

More information

A Traffic Camera Calibration Method Based on Multi-rectangle

A Traffic Camera Calibration Method Based on Multi-rectangle Traffc Camera Calbrato Method ased o Mult-rectagle Lyg Lu Xaobo Lu Sapg J Che Tog To cte ths verso: Lyg Lu Xaobo Lu Sapg J Che Tog. Traffc Camera Calbrato Method ased o Multrectagle. Zhogzh Sh; Zhaohu

More information

Toward Undetected Operating System Fingerprinting

Toward Undetected Operating System Fingerprinting Toward Udetected Operatg System Fgerprtg Lloyd G. Greewald ad Tavars J. Thomas LGS Bell Labs Iovatos {lgreewald, tthomas}@lgsovatos.com Abstract Tools for actve remote operatg system fgerprtg geerate may

More information

Delay based Duplicate Transmission Avoid (DDA) Coordination Scheme for Opportunistic routing

Delay based Duplicate Transmission Avoid (DDA) Coordination Scheme for Opportunistic routing Delay based Duplcate Trasmsso Avod (DDA) Coordato Scheme for Opportustc routg Ng L, Studet Member IEEE, Jose-Fera Martez-Ortega, Vcete Heradez Daz Abstract-Sce the packet s trasmtted to a set of relayg

More information