— SREE VIDYANIKETHAN ENGINEERING COLLEGE (AUTONOMOUS) SaRE SAINATH NAGAR, A. RANGAMPET~517 102. (Affiliated to Jawaharlal NehruTechnetgnes University Anantapur, Ananthapuramu) Nameofthestudent ; Dillaesaint = rme [9] [To] [als [2) fe rr : st E 7 Year & Semester fo oe prance AUAL Nameofthesubject ; Machine Learnin Academic Year:__2.07 3 Ome Signature of the Student: Signature of the Faculty: Q.No. 1 2 2 P I = Marks, Awarded eee ae Cl "(Start writing from here) @ LJlok tthe diene blu avn—linear | ea ond multiple — vearssion , end how ave they applied in sali beal onalytis 2 B lonchan, Pye alenclnmr Rguion 4 « =———+ nem ot yenlaor anal in which dale is fe a model anc thea caprtsed apa a enadhemed cal furl ion 5 i ine learning . Regression i 4k i a methine tvnng | vlutrpl pe fo pedi a denndent arable | algorithm ae | pith two (i) more pe dors patios 1A ipl Regression has numerous real- 1 j Mu Jone tn three lomains. | | ' }es ton-linear reyretsion q mutt reareisione: ee pd: ers es Meslne rasan and maliipl regression, ow tae dllewat “kechniues ued an llhial orale, aur leculanly ution vine centre a ef enalusis i yidich fo. chad fr anedling 4 uilationshp Hui a dy sere independent ae Hee are the ley (OSE enh a Harem | ne dlieg with applieitione : deat — vavtable end of @) ef the” Relationship = fee ate ze Gn the, the onlay he de vartable onc) athe. inde variableg) i¢ mek assume! to be ‘bnear. Tplisdia she fosnye tcHellvartousi can lia. pe ae expeneectial, tegevithmic ply ony athe non-linesr fusetian | ] « cfuple, egremton <= on thi athe Hand’, agunae rlaonship By the depuadeak variable @ | sicliok variables. 4¢ model, this ius | | uteg a lnnar equition there the eo-dfsaen |oepact che dle Pilea tadepnclnt varrabl on_dhe_clped ; ferns, Maennear Regression hae ene vegesten Yypically involue nealincor furitins ond Poranales md ae crlimadd Jo but fsb the te EAatelemmiensth. Pesusli based ts spreilied on the native of the relaltonehip yo went to ego rata and 4aker tee the fake , whee Reoveston x In this, the equation te) hice ensuite, Ye the epinclat verrable , po tc the tale ae PE fivcde, are the coe Mnat fer the indepeodens sean(iee + nonlinear, Regrion i doce *, cs ees (@) conlant vorante OF errors, conch ty a We assumption 9 cutipl rts A cae taly cuumpliens i) Pe enan ele ae furction chosen. seple Rayaaton + Ht afeens tne, ind: padar E = moneda hit ly Ceonstont varione of tinees ) 9 cornell hia crnts, variables x4, x2, ele4) Wedel tompliy =——x lone \h R yasion 3- ele I fpanike e Se se con ie sgncenly Mme compler thea linear models. The ae of | non-Unear hunitions Bnew parameter , ond the Need] for initial porander ermater can | contribute lo model tom aa Eeconmalee inenellinieay models ma Veg Wire alieated onkimreation echrigus ond can be eomptledienall lemon. i i j= wlausple | Regrsvion ee made “on gers simpler 3 drudin , jhe model eoneits ob & equation with a defined set of ete > Siena ty con be on adluontege in devs ob Race rolememocciy talerpretatcent linear ’ erhimatron 3 ond i terapalazonal et iuensy h blir i athete ave some of the ey ie . dite Wea AE (aeael ited i clepea de nq the nature af your date and the vaderlyeng rela enh p you wank Jo caprtare.| Diag roms beds ayant the ve he 4g Ucts 3 Si of the pe eros, 4 I 4 ee a gO x ker Regression | S- i Lagiske Regrecion ieee Py Or pubes eres |Agpted on OB tebe analysis > rey Bol non-linear reqreson y muthple regresion oe aped in dlalistical anayitt de medal)’ aad undivdlend reletionlape Ho varia bles Aeo'e how eoth of these echaiguer applied an alotirhica) — analy, Applications of rlenLinear Rlegrerton ——— — © Mocdhling cen Relakonshipt w- JE ir used cohen thee ae «eb Comet ean ks redondyp Hut the denndenk ond todepeadiot variables a abe qustly clsivibed by 2 bincor model, Fu fre dk is pled in stludtiny wshee the ~~ ale ae pone) gristh,, gcturaline cay other “nonelineer pullers @ Phenomena ith Aeduval bodle yyy porteolaly | ural or modeling dale sith paluval tw, | where the depacleat vlevahle centet gross | | tndefrately , fer daclude populsction gtouth models easyme kinele ond a adeplon ries ae— a a) cane Fakbe St is wed ify tuwe liking, tchere the ic lo find the best Sidbing cena te fuition that denaiber the relalton dp ~ Alas variables sth tc commun a fils such ax leg yy chenisy phys fb enginennye | a) Fie Series Analyts te 4k con bh applied in time ent onalanice eae pollens that ahébit un bnear behewos over time, Applicctons ot Mawnple Regrevon SSS ar © Bereme, Arey caonomis 0 anelyge the elaton yar bls peed ee eddy used om ayullple —uonomic voviahles jock || fy eae de cot be used to ctudy fel a BES fae Oi ilale ates this ol tonsmer 1Alme of eensamerepradny | | | lo Soria! serontes + SE te employed to undercland || Toc Tweion nlapedat wevable, fog, aduatn, vneome , femtly Bye) intfucace «dependent | vesiable Ces. hegprness , erime ale, atedemie exhievemea| Sy ies || \ | z ee “vasa all 4 - * | 2) tlorkek Research ox fe verearth te rhuel re agilied in mala t behave. $hulpas markelos fj ena ase the ampact of multe ae thraleytes Coy foctans of sales, curlome Satisfedtien , ca) bron toyally < foniumer 4) Eni anal Sunes ondlyge tk any ie 4h con be used 40 of attous Lodors gy, apurative ennivonmunta 2 polludron, habitat quality) 1) a dipeadent —vostable bye pee dash | | Block as Matiple (ep. war sk x feature 10 apes poe ny sf ae 5} linen ynscen [Ben abel Ne 4 = we “| the hoght Ae (oalep” i = a =OD pensases a Nosese peda tom Nequenton Ande 2 A. conpluitive cnidique (Rihwd Ades of ppeeai tom ar Matbing Learning ert Boole refnee Goielusion 5 d tenelude thet dhe thon linear rps is cployed then relettoaship: ble variables ore eeclintee Ube mullipl prawn avers near velahonships onc is uicel for tnlpretin the rmpock of multiple factory 0n a chyjeadent variable, the theta tls the 400 methods deals on the patie oF the lata bk sthe recarch goals anusing ffi lethal onde