Big Data Analytics

 

Big Data Analytics-

MsVk u;k vfo"dkj ugha gS] daI;wVj vkSj MsVkcsl ls igys ds le;] dkxt ij gekjs ysunsu fjdkMZ] xzkgd fjdkMZ vkSj laxzg Qkbysa vkfn lHkh MsVk gSaA daI;wVj fo'ks"k :i ls LçsM'khV vkSj MsVkcsl us cM+s iSekus ij MsVk LVksj djus vkSj O;ofLFkr djus dk ,d rjhdk fn;kA ,d ekml ds fDyd ij lEiw.kZ tkudkjh miyC/k gks tkrh gSA

gj nks fnu esa ge mruk MsVk fØ,V djrs gSa ftruk geus 2000 dh 'kq:vkr rd fd;k FkkA gj nks fnu esa gekjs }kjk cuk, tk jgs MsVk dh ek=k yxHkx 5 zettabytes ¼orZeku esa½ ls c<+dj 50 zettabyte gks tk,xhA nqfu;k Hkj ds IykaV vkSj dkj[kkuksa esa vkS|ksfxd e'khujh c<+ jgh gSa tks MsVk bdëk vkSj lapkfjr djus okys lsalj ls ySl gSaA fcx MsVk dk rkRi;Z bl lEiw.kZ MsVk ds laxzg ls vkSj fofHkUu {ks=ksa tSls O;kikj vkfn es vius ykHk ds fy, {kerkiwoZd blds mi;ksx djus ls gSA cM+k MsVk dafiu;ksa dks vius O;olk; dks csgrj <ax ls le>us esa l{ke cukrk gS vkSj mUgsa fu;fer vk/kkj ij ,d= fd, x, vlajfpr vkSj dPps MsVk ls lkFkZd tkudkjh çkIr djus esa enn djrk gSA fcx MsVk daifu;ksa dks MsVk dk mi;ksx djds csgrj O;kolkf;d fu.kZ; ysus dh vuqefr nsrk gSA

fcx MsVk ik¡p dk v fuEukuqlkj gS&

•   Volume– MsVk dh ek=k tks mPp nj ls c<+ jgh gS- isVkckbV~l esa MsVk dh ek=k ekih tkrh gSA

•   Velocity– osx og nj gS ftl ij MsVk dh ek=k c<+rh gSA lks'ky ehfM;k c<+rs MsVk ds osx esa çeq[k Hkwfedk fuHkkrk gSA

•   Variety– oSjk;Vh ls rkRi;Z MsVk dh ek=k çdkjksa ls gS ;kuh fofHkUu MsVk QkesZV tSls VsDLV] vkWfM;ks] ohfM;ks vkfnA

•   Veracity– osjflVh miyC/k MsVk dh vfuf'prrk dks lanfHkZr djrk gSA MsVk dh mPp ek=k ds dj.k ;g mRiUu gksrh gS A

•   Value– ,Dlsl fd, x, fcx MsVk dks ewY;ksa esa cnydj] O;olk; jktLo mRiUu dj ldrs gSaA

fcx MsVk fo'ys"k.k dks mi;ksx] tkudkjh dh [kkst] fu"d"kZ dh lwpuk nsus vkSj fu.kZ; ysus ds y{; ds lkFk MsVk ds fujh{k.k] lQkbZ] :ikarj.k vkSj ekMfyax dh çfØ;k ds :i esa Hkh ifjHkkf"kr fd;k tk ldrk gSA fcx MsVk ,ukfyfVDl esa cM+s iSekus ij fofHkUu lzksrksa ls MsVk ,d= djuk] foy; djuk vkSj bls ,d rjg ls çLrqr djuk 'kkfey gS rkfd fo'ys"kdksa ds fy, vafre mRikn MsVk forfjr djuk vklku gks tk, tks laxBu@O;olk; ds fy, cgqr mi;ksxh gSA

MsVk fo'ys"k.k çfØ;k esa lfEefyr pj.k bl çdkj gSa&

·     Data Exploration– O;kolkf;d leL;k dk irk yxkus ds ckn] MsVk fo'ys"kd leL;k ds ewy dkj.k dk irk djrs gSA

·     Data Preparation–bl pj.k esa MsVk folaxfr;ksa dks MsVk ds varxZr feflax oSY;wt dh rjg ns[krs gSaA

·     Data Modelling– MsVk rS;kj gksus ds ckn ekMfyax pj.k 'kq: gksrk gSA ekMfyax ,d iqujko`fÙk çfØ;k gS ftlesa lq/kkj ds fy, ekWMy dks ckj&ckj pyk;k tkrk gSA ;g O;kolkf;d leL;k ds fy, lcls vPNk laHko ifj.kke lqfuf'pr djrk gSA

·     Validation– bl pj.k esa] DykbaV }kjk çnÙk ekWMy vkSj MsVk fo'ys"kd }kjk fodflr ekWMy bl ckr dks osfyMsV djrs gSa fd fodflr ekWMy O;kolkf;d vko';drkvksa dks iwjk djsxk ;k ughaA

·     Implementation of the Model and Tracking–bl vafre pj.k es] dk;kZUo;u çfØ;k dks iwjk fd;k tkrk gS vkSj blds ckn ;g lqfuf'pr fd;k tkrk gS fd ekMy lgh rjhds ls ykxw fd;k x;k gS ;k ugha \

,d MsVk fo'ys"kd dh fuEufyf[kr ftEesnkfj;k¡ gksrh gSa&

o   xzkgdksa ds fy, O;kikj ls tqM+s eqíksa dks gy djsa vkSj MsVk vkWfMV vkijs'ku dks ijQkWeZ djsA

o   lakf[;dh; rduhdksa dk mi;ksx djds MsVk dh O;k[;k djsaA

o   lq/kkj ds voljksa ds fy, {ks=ksa dh igpku djsaA

o   tfVy MsVk lsVksa esa :>kuksa ;k iSVuZ dk fo'ys"k.k] igpku vkSj O;k[;k djsaA

o   çkFkfed ;k f}rh;d MsVk lzksrksa ls MsVk çkIr djsaA

o   MsVkcsl@MsVk flLVe cuk, j[ksaA

o   ijQkWjesal baMhdsVlZ dk mi;ksx djds leL;kvksa dk irk yxk,¡ vkSj lgh djsaA

o   ,Dlsl flLVe fodflr djds MsVkcsl dks lqjf{kr djukA

Data cleansing- fn, x, MsVklsV ls] MsVk fo'ys"k.k ds fy, vko';d tkudkjh dks lkWVZ djukA MsVk Dyhaftax ,d egRoiw.kZ pj.k gS ftlesa MsVk dk fujh{k.k folaxfr;ksa dks [kkstus] nksgjko vksj xyr tkudkjh dks nwj djus ds fy, fd;k tkrk gSA

;g cM+s MsVk lsV~l dks lalkf/kr djus] mUgsa lclsV esa foHkkftr djus] çR;sd lclsV dks ,d vyx loZj ij lalkf/kr djus vkSj fQj çR;sd ds fy, çkIr fd, x, ifj.kkeksa dks fefJr djus ds fy, ,d :ijs[kk gSA

MsVk ,ukfyfVDl ds fy, loksZÙke midj.k : Hadoop, Apache Spark, RapidMiner, MongoDB, Storm, Tableau, Cassandra, Microsoft Azure, Talend, Zoho Analytics, SAS, Cloudera, KNIME, R, Xplenty, Big challenges with big data, Lumify, QlikView, Qubole, Splunk, Databricks, Apache SAMOA, Splice Machine, Adverity

Importance of Big Data Analytics-  fcx MsVk ,ukfyfVDl okLro esa lwpuk çkS|ksfxdh ds {ks= esa ,d Økafr gSA daifu;ksa }kjk MsVk ,ukfyfVDl dk mi;ksx gj lky c<+ jgk gSA daiuh dk çkFkfed Qksdl xzkgdksa ij gSA blfy, ;g {ks= fctusl Vw daT;wej Qy&Qwy jgk gSA fcx MsVk ,ukfyfVDl ds rhu foHkkx gSa&

1& fçflØfIVo ,ukfyfVDl       

 2& fçfMfDVo ,ukfyfVDl         

 3& fMfLØfIVo ,ukfyfVDlA

pkj n`f"Vdks.kksa fuEu gS\

•   MkVk lkbal ilZisfDVo

•   fctusl ilZisfDVo

•   fj;y&Vkbe ;qtsfcfyVh ilZisfDVo

•   tkWc ekdsZV ilZisfDVo

Big Data Analytics and Data Science ¼fcx MsVk ,ukfyfVDl vkSj MsVk foKku½

,ukfyfVDl esa fofHkUu lzksrksa ls çkIr fofHkUu vkdkj ds MsVk ij mUur rduhdksa vkSj ,ukfyfVDl ds Vwy dk mi;ksx 'kkfey gSA cM+s MsVk esa mPp fofo/krk] ek=k vkSj osx ds xq.k gksrs gSA MsVk lsV fofHkUu vkWuykbu usVodZ] osc ist] vkWfM;ks vkSj fofM;ks fMokbl] lks'ky ehfM;k] ykWx vkSj dbZ vU; lzksrksa ls vkrs gSA

fcx MsVk ,ukfyfVDl esa e'khu yfuZx] MsVk ekbfuax] uspqjy ySaXost çkslsflax vkSj LVSfVfLVDl tSlh ,ukfyfVdy rduhdksa dk mi;ksx 'kkfey gSA O;olk;ksa ds fy, fo'ys"k.k çnku djus ds fy, MsVk dks fudkyk tkrk gS] rS;kj fd;k tkrk gS vkSj lgh :i çnku fd;k tkrk gSA MsVk oSKkfud O;olk; esa fcx MsVk ls fuiVus ds fy, ftEesnkj gSaA vkips gMwi] LVkWeZ] dSlkaæk] eksUxks Mhch vkSj dbZ vU; MsVk ,ukfyfVDl Vwy 'kkfey gSaA

•  Business and Big Data Analytics ¼O;olk; vkSj fcx MsVk½& gMwi tSls cMs+ MsVk ,ukfyfVDl Vwy LVksjst dh yxr dks de djus esa enn djrs gSaA ;g O;olk; dh n{krk dks vkSj c<+krk gSA uohure ,ukfyfVDl Vwy ds lkFk] MsVk dk fo'ys"k.k vklku vkSj rst gks tkrk gSA ;g le; vkSj ÅtkZ dh cpr djrs gq, rsth ls fu.kZ; ysrk gSA

•   Real & time Benefits of Big Data Analytics ¼fcx MsVk ,ukfyfVDl ds okLrfod le; ds ykHk½& çkS|ksfxdh ds ykHk ds lkFk fcx MsVk ,ukfyfVDl ds {ks= esa Hkkjh o`f) gqbZ gSA f'k{kk {ks=] cSafdax dks fcx MsVk ,ukfyfVDl ds vf/kdre mi;ksx ds {ks= gSA blds dkj.k dbZ m|ksxksa esa cM+s MsVk dk mi;ksx gqvk gSA dqN bl çdkj gSa&

•   cSafdax

•   gsYFkds;j

•   ÅtkZ

•   çkS|ksfxdh

•   miHkksäk

•   fofuekZ.k

• Job Opportunities and Big Data Analytics ¼ukSdjh ds volj vkSj fcx MsVk ,ukfyfVDl½& fcx MsVk ,ukfyfVDl esa fuiq.k is'ksojksa dh Hkkjh ek¡x gSA vkWxsZukbts'ku ;ksX; is'ksojksa ds fy, vkd"kZd çksRlkgu vkSj iSdst dk Hkqxrku djrs gSaA