€bcle . WEdIGEtTtdididid CESESEEEE SE: EX.No:o4 APRIORI ALGORITHM TO EXTRACT DATE @ior.a4| ASSOCIATION RULE 4 AIM. Wrike to implement Apriori Algoritim to extract Bssocation ywe of Data mining. ALGORITHM. SEePL: Skark tRe process, Skepa: click R> Ri 3863.22 in Desktop +o open RTeol = window step 3. Select Pac > load pack ules. Now, arules mie on ‘elleted” Tom librany, sep 4: patterns is designed in tRe nikemssiooo steps: The pattems Summary axe displayed on te Screen. step 6: items =io00 yeprenents tRe columns and trans 1000 Yepresents ‘tRe rows. step 7: Rulos of apriori is lisk the. Some Parameters ep and thon daclone § pod moe. ‘tonfitied =09 and Larget = 1 en. step & Rexk 1 call ules) function tRat Inscect “ to visit the list” and sor “ skep 4: Stop te process.De) PROGRAM: Be local (pkg select sist (sort packages (all .awailable = . TRUE)), Qvaphics= TRUE > sif (nehar( pkg) library (pkg, character. only =TRVED } ) Loading equine package: Matrin Attaching Package : ‘aoules’ The following objects ane masked from ‘package hage’ : abbreviate , wyike > UibvavyC “ovules” > Patterns random . patkerns inikeensioaa’s Summary (oatterns)s Set of 2009 itemsets mast Lrequant items: ikemuiaa Ytem34qttemo3i itemgo4 ikem 4a) (othe ba 55 54 54 53 1633 a bans “onto transacting WThems=l008, nTrans=1000, 7IMOG¢6 rans) ; > agrawal, Patterna= patterns) “aK —| | | | ! | aiargactions (Reus) ff | PPO OHEEEELELELEEEE ECE EEE EEEYY YEU>datal "Adult", >ruley( aprioy ; (Adult, povametes click (suppor:=0-6. confidence =0.4 , target=“wles">) ; Aprvioy : Parameter specification: Confidence minval Sman orem wal 2.4 +) \ non FALSE >Tuler, Set of 5a vules >inapect (hand (sovt (-yules by- “lifE"), 305 O14 { gex=Hale, ‘ hakive~country = Sie kes 42> {racswhike} o54'5421 oFosteto 0.598R1T0 10 rhood (qualikytrulesy) ; Support confidance coverage |i lank O-4I738G7 0-9)73867 = |.008@0 = Lo0p0e 4.4 807 0-4539779 1538774 L.ovop00 Wpov0D AEtED \ 2 3 O.54d5 BIS —>SFBSCIS 0-6809327 098712 gE gy 4 ©. 5G0bb5D oBbobbso F682 Of 92 44,900 > Summary mes)» Seb of Sa rules Yke TengiR duatyibulion ‘ Sizoy 12 3 4 2 1% a4 13 minifig info: dota Lronsec Hons, Suuppor bortidanes Beek U ehh eh ee hk ee Eee ee LeiBe EX no.0g K-MEANS CLUSTERING TECHNIQUE Bao DATE:31.07 25 = ° _ sp AIM: . Be hike to implomont k-means clustering techniqne = tap ALGORITHM: Se Skepa: Start tho process = Step a: Select Rhool-s Ri 3863.2.2 ; step 3: $d yaluas are Gtoved in. tRe X Variables 3 ‘ wee wad vin formed ton given valus 100 : Skee 4+ Put wy values $6, column % auld , Y ams » Step 5. Then, call kmoans vd function with a porameters — %,a ond Started in that is, Corekmaanms = (4.29). Thes function displays & clusters OS of Size 0,50 clustering mean value , a Sum. of Squaw, and oWNoiloble components. "> Shep b: Plokt> metfiod to be used +o plot by = tep Eels on the Screan witR tdi Gf aby “9 skept: Prinkt motRod be be used Tk con ; di layeat tRe conker point centroid each <| , o| step 8 Next call k-meanat) clustering function - to set Clustering starting Voluors ab 20. 2 stepg: Again. plokL> method to be used. 2 ~daplayed the different sizes oF ahead 2 : is Skop to: stop the Procoas. 2 —_ = _ 2 i| PRosiean: > =rbind (matyie(vnorm live, Sd=0-3) hdol=s), motrin Anorm (no. meary=!, $d= 0.3, Ncol =2)) > Colramosn) =e("%", "4" r= Kmeans(n,a)) k-means clustering with @ clusters of Sizes 50,50 duster means: ‘ y (105309944 = 1.085 33893 2 -O. 01609549 -0. 02649584 clustering vector: aa CY) sasaare re 22a ze aaa eee Se dorrzare 22 2Ke2a2 BRR * yal By wpyeyey ayy wet | phe Oc als ci gcluster > 4 15 2 | iF 6 a) “og2 daal °,°? yD ©. 2 49 ! 100° -95T COCCTeT eee eee eee eeTY —2 > points (eibebnkev’ Clete, Peh=8, won=ea)>x=vhind Gratrin Crnorm (loo, Sd=0-3) ncol=2), motrin Cinorm(ito0, s4=0.3), nenl=2)) matrix (ynorm (too, $420.3), neol= 2). makyix (norm (v0, means!,$d-03),neol=a), tratrin Crnprm (100, maan =}, Sd=0.3),neol=@), Mottin (norm (00, mean=1, sdz0.3), neol=2a)) >colnamas OO) =cO'"%", ”y") © 1 = Wmnaanalx.b dS K-means clustering with b dusters of Size 46,50, A421, 16, 47 Cluster moans: * y ob Hob 22 149462484 \.25pIogo> =:308 27611 ) 2 3 -0.04824088 -0-03915459 4 \45ggeon 91853549 s \eprb8 0a 30 9-80RIR34S 6 — co.6y5324a0 O-1E138673 Available Componont § : O} “cluster” “Centers” “totgs” "wi ase “Lok -wilRings” “betweanse’ “size” “ites” XS vidaul” SSLSELEEKRELDELEGELEGEREEELEEETESEUEE| a » -o3 bo Gb te ye 50 5 >poinks Cer fanters , clel.b, peh = 8) he 2, ogo be, 99 go ‘© 007 90 aoe 25 4 Be ob:O 960 9¢ 0° SS ——41— +05 09 85 lg 16 20 SVHV.ovuuesRBELELEDELE~ eS » ] to FeNo:03 HIERARCHICAL CLUSTERING a [DATE 0202.01, | ww AIM: . Write bo implement any one hierarchical clustering ALGORITHM: Step 4: Skat the PTOcoas . Stepa: th Desktop Rbool > Ri 386.22 SEPA Using 4 variable be define matrix format and iks volun ore miliolized Step4: Spearman and tefaplete method ano inikilized tditg cand by variable. step 5 plokid is used be declare \Vaninus valussand coll Re dendrogram. method using as Vouoble. Skop 6 Tide He variable unclassc) method mar height Values and label valuos one. cLisplayed. Using dondrogramc) meted to dus dordrograrn treo stuckune. Poy eps rd ond hig PLO method. qa skep 7: 2 Voluos one plotted uaing, stepq: Distance values ano erage and Stored SeeeaKeLETETETEETEEL EGER EE EEE Bs motrin fon Lowi rm i oe d= AOE (on.malvin onbeat sy) owing one. step le: ee matiiod Set of clusters [or tices cts ont orS a de) PROGRAM: SSS — | Se) > y2-matrix (xno rms), 10,5, damnames -list (pasteCg : \t0, sep=" "J, sae tpoate ("4") \5,sep=" "DD Be ><<- corlkty>, method ="Spearman”) ; =e rd ean, distlt-e) ; dhre-helust(d smethod -"Lomple te", members=NULL) rPax(mfyow =c2,29)5 SPbL Chr shang =D = Pal DOEETETEULECEETECEECEEEY 3148 $5 gn gq beELES_ os = >ple( an dondrogyamthr),ecgapar “bist (tall =4), ~ a ry] ~ | | | | | ETH | 3 | tun class (hy) D1 ol 0.2 0.3 0.406 0.7 1.0 [.2 1.4 mp $ oydoy me 0134461765210 CeeCuC ery | > sty (as dondsogromthy )) Pp -- [dondagaam wi, branehoa and \o mombers at h=19] | E '_.Cdandiogyam. wis branchas and Tmember ak h-=i ' | 1 \ \--[dendvogram w)2 branches and 4 mambers. my | | 1 plandregiam vol? branches and Smambers ak | 207 eudud |.=-[dandwoqram vale branchss and 3 mombers ot be = mh }labels [hrs Ordon] hab] ad chaxacter (o> ~ Q a. a oe s i>pay (mf yoo =eC 2,295 - | | | >Wid 2-04 dand voqnarnthy ds >ploeChyelads to 25 R 9843 9) i a we ss 4 x shel 2c sates chrdi, sarveple(s s1o))5 | >plok Girdads SSEESISTRESSEFSESSS UCU EVEVE EFF) : | | . haf a EI > labels Crd) oom eit . : 30 ‘ys sgn "ga “gb “ga” “4d “qa” gs) “ga” “gio! ® Jabels(hrda) * Cin" "RT QS" og gn Ogee nga’ ge” "a4 ge! 2seo yar™ Clumern dis dist (os. matrixtntuars,)) helust id) lush Che, Ke 3, border ="red “> >he sd » > Plot the) b >yeck he | -[- Fret auty rttirmies, Qieic Sworn, BRING FBI Ca paved pay pepe yu WEAN “comelere” ) Relask AC, Syrom gaa] | Pere? (etathuy tayshays — pasg suvarel T TT TT-T35 —— Ex:no - "A CLASSIFICATION ALGORITHM USING KNN DATE 4.03.24 AM: Write to implement classification algovithun Algoritim, Step 4: Stark the procoss Step a: selecl RT 2 Ri 38622. Step & The tow Q colummn ave mentioned at Al,Aa,A3, BI,Ba,B3 and VW-¥a Shep 4: hind) method one usedto create te tuo variables. Step 5: Tha groups ome gloyed in tratn variable. CUECEEEEEE EYEE E YEE! » | step &: plotc> metfind Is used to drow the 2 Point for Parkeuler value. | » stept: Stop the P10 cos. 2 < | ° | 2 e . > | » » 2 \ |VVSCSCUVOUSEREVEVEUEDER EEE EEE YY YEE KE PROGRAM >Al\=c(o,0) rAa=c() AB = 09,2) YBI= 6 (bb) yBaz ¢(8 6.1) >B3=¢(b.6,6) Shaun = vind (AL, Aa .A3,B1, B2.83 ) det =factoy(ccrep('A", 3), vep'B’. 8D) yheat = (4.4) >libraryclaas ) © Strmany(tnn Chron beat, #1, k= 19) ou >plot reat Tet 7 i? a >Plob(trainy ** '* ter °° oC oO | ———4| 7Sunumary Chest) Min dst Qu. Median Heo axd@a. Mare 4 4 4 4 4 + Summary (train) VI vo Min . : 0-000 Min. 0.00 Ast Qu: |a50 Act Quik 26 Meduan: 3-760 Heduan:3 50 Man: b.500 Hoax: 400 | Print (test 3 Cl 4 4 >prindltrad Prntlbtginy 4 Al 0.0 Az 10 0 | Ad 30 2 B) 6&0 & Ba 56 7 B38 65 5 >View Chest) o [RowiG ety Fol TTT TPE FOO CUELEUEUUEUEEEUEEEY YY YUKU,. pb YNo-og DECISION TREE B& DATE 2302.24 . a AIM: Wrike to implemont decision tree ALGerTHM: Step4: Start Ra process. Skepa. Create cota. set in MS Excel with weather , tempeyalure , huumodiky , wind, golf play and Sowed ik ih "Cll Sweat I Booka.csv” steps Read Lablec) metRod used to retrjewe dota from tstrcal dake set and Stored inks Play-bost Voriable step 4° Next, use print data set values using print() mothod. Steps The doka set sammory Yeport also displayed using —summany 9 method. b: The ypartt method dito du step necoatandy Kad Vale - duplay skept Plott method used to draw te "e docsion trea without ‘slo. omer Step 8 Then next meticd used bo insert Voluss inks decision treo : stepd Tha deusion & : | *e coupe veule fled He skep 0 Stop the processte Cs PROGRAM > Ploy base <- yeadtable (esi! SweakM Book 4.cev", header = TRUE, sep="5"5 Print (play-bare) x neater Lemporatuna huenady wind me poy So, @ & re Ta 33 Gita hak hgh rene wes : : Tain “sy Nat nono yeh 8 5 youn cold een fio he ff we 4 Todo fo ge 7 1 TOL. coli ah, rand Ro 8 8 tino. wort : fone 4 q fine Gold , pedi none ed 1010 Youn, worm rods. ‘feu No u y faing warm. he ‘te ho \2 12 cloud Matin ¥ hone ye fag aki = M4 M4 R warn high. few yr ySumumaoyy (ploy-bare) Ly weather temperature humadily wind golf fey Min. s100 clouds 3 Cold 4 hah prea yest | At Qe 425 fines — hot4 . 1b aig | pe Medion: 7.60 tan: una meio Hone 8 > Hens 7-50 2 oI 3rd Qu slots we Mow. :t4,00 > Cd a | es s 2~~ — - —_____35 Sw > local CLpkg z ~select .}ist (sort (- packages (all-available =TRUE.)) a ics =TROE ) HE chor (93) Wbrory (tha charade only TRUE >4) fit < ~* part (golf play ~ weathor y temperatuna & st +wind, motod ="closs”, dota = play base , Cantro}= Ypovt. control (minsplit =1) > > Summary (SE) Variable importance weatfior temperature husrodliy Wind Ad a ao 15 Node number 1:14 observations, tamgleniky payam = 0:3 Predicted clas - yen ee el eta bovine CP nod) =I Clays counts . 5 tg Probabilities: 0, 457 0643 lef som. a Hobs) night Son =3C1 bs) PP PLORED N=4 nada), split, 1, loss, yval, (yprob) ¥ denotes terminal node Nvook 44 5 yes(n-251429 0.642964) Shumadile high F & nolo.514286 0.4286 714) A)weatfor=fine 3 0 Nolkoooo0e 0.00000 0) weather = cloud, Yarn 4 I Yed(0.2800000 0-1sPe000) lo) weathar =Tair 2 1 hop. 5000000 ©. 50000000) do) wind =few 1-9 Noll. 0000000 0.00060 00) * 21) Wind stone 10 Yer (0.0900680 |.9005900) + \\) weatfer = cloud @ 0 yes (p.0000000 |.0000000)* 3D humadily = moduiam 71 yeoto.42867) 0 11429) vv CURR CEROcd » » » >plok (tit uniform “TRUE , maan = “Deniicn Tree for dataset" > We Sian wey) CRP vA [lays] | |. | | oben (fit n=TRUE » a\l=TRUE , can =0-8) ReGraphicr Don eae Ac bye) [lel] | Decision Thee toy datoretEX:No:ob CUELELEUUV IU Vue neem erin ee — LINEAR REGRESSION DATE 05 03,24 AM. Wrike to implement Vineot veg reasion. ALGORITHM: Stepa Gort tha process, crogke databane in NSEncal with brain. era og, boly-weight 4 Tndax and Ren Soe iL, ¢ Step 3: Next, print tRe data set value using Print) method. Step 4: Read Lablec) matRed used to tebriove doko from Ms-Exco) database and Stored it by using di voriable. Step 5: plokid) metod used to display volus vf bron weight ond bodyweight. Steps Ablinet) metRod is used to drow tka Stiaight \ine for the rule of Bracn_ wei and be -weight . Steet. Agouir lokO mathiod to be used. ea diiplay fa varigua types of Standardalizod on te sereans Aubpul yeaults, | step 8: Stop the process. SbedePROGRAM: od <—yead Lable (“c:/! sweat! Booka.cav”, has ve TRUE. | sepe".”) >printidey Index Brain weight fhody_weight \ \ 2.6410 46.914 75 Ao 2 2 2.9 tbo Tana 404 a 54. a2. a . o & 5400 60.151 > fila - Am (hog (Basi weight) «log (Body-weight), data=d)) >Par (mfrow =< U,199 . RRovjtia- Dacor FBR >plot Clog (dt $Body weight), Log (AL-4 Brainweight, +trlab=" 0 (Body weight y", ylab=" log (Brain weight bu Tmain="Simple linear Yeavession" > CUUUUUN UC CE Ue __Sirple linear. teqression _ Yo 3 Bram_weghtspabline (f1) ungie [neon Fagregore >Par (wtroweetaa)) “Log com cbeight ) Plot (fit) VEUHUUV UC ECU | jo a Ea | PiRsvapiics Dewicad ACTIVE), iP | Foxduala ve teed Mopral QQ 5 > ay 3 | 3384 . 5 "| 2 | = a 33%) ° 38. ° “ ° / - £ oe ot or yO - 2 al a Hel, sad ne 2, de ted Webs 3 “Thaovec tical Quanbbds 2 Gente. Locabiors Residuals vs leverage s S 3 Y 7 ~! q 3}! oy | 2S 3 ' > 2 |3 el Cache eo a & ' - . | 9 |S. et oN JT tl ° . fal ° Se Yo ro Ns o Nabe bt ot tae, y 5 s ae Fite Voliue ag | ? 2SSSVSROCHUECCEUEYUKHUEYECEC CCH EH we Tee EX\ND sof DATE C1 08.4 DATA VISUALIZATION AIM: Write to implemont dota visualization ALGORITHM: Step 1: Start the Prowas. step a Create a table in MS-Encel. stepa: The toble is Soved in the c directory, and Mud is Soved in the form of “av in extension. skepy: Th tha table, the following fields ore inserted — housing , food, cloth , errker toxin and other steps: The appropriate integer yaluos ane jnserted in Rat fields stepb: Select Ro Ri 3863.22 in Desktop to open Ric) Window cept: ‘The table is Viewed by using te Variable “ene”. chops: Plot ta points In ta table. . bunplot O motRod is used to dirgl | Shep 4 to Data Viigualiza tion groph. “ shep jo’ Stop tRa Procads.. ab 47 » PROGRAM: - »| > « » exp ¢-vead table ("¢ :// Sweet HI Rooks.csv", header =TRUE, > xe sep" ") ™p eee » » \ X housing food clotR entertain othor ' (oo. Woo joo too 200 : 2 2 2.00 Joo = 2.00 200 400 3 3 300 Ron 300 300 boo v4 pe ft 40 “joo 400 Ape Bee ape f mm boo toe oe boo p00 loo 9 + » se 2 “100 4oo0 100 Too goo » 4 a oo Boo Bou 800 70 P q qoo a90 700 qoo Joo ye © oo 100 &p0 Joo 400 ® >Viewtexp) » > | PData teup = lof] | 2 FB} 3 ts housing food | clotR [entertain | otfer | (| oo \oo \oo { » 2 |? ao | 400 200 | = 3 | 3] 300 | aoo Apo | e 4 | +] 400 Too | bon ~ [8 |5| 500 | 600 foo = 616] boo | Bee. \o9 ° 7 || too 4oo oo | &| a0 50 a | 600 | Joo | __ 5° | a 4} 40 op &oo > yo | to | 100 too | 700 - ) 7 2 E a[ > plot cong) - ] Roche Deyers ACT - IF | &@eeveees EVEN UNVC CCV UN TCU C YUE e wew Ars]