Varshney 2018
Varshney 2018
$&RPSDUDWLYH$QDO\VLV
6KZHWD9DUVKQH\5DNHVK.XPDU
'HSDUWPHQWRI&RPSXWHU6FLHQFHDQG(QJLQHHULQJ
0DGDQ0RKDQ0DODYL\D8QLYHUVLW\RI7HFKQRORJ\
*RUDNKSXU8WWDU3UDGHVK,QGLD
[email protected], [email protected]
$EVWUDFW :LUHOHVV 6HQVRU 1HWZRUNV DUH UHJDUGHG DV VLJQLILFDQW WKHVHQVRUQRGHVIRUDQDO\]LQJDQGGHFLVLRQPDNLQJ7KXVWKH
GHYHORSPHQWVRIVWFHQWXU\DQGKDYHJDLQHGZRUOGZLGHDWWHQWLRQ PDMRUWDVNVRIVHQVRUVDUHWRVHQVHSURFHVVDQGFRPPXQLFDWH
GXH WR WKHLU LQWHQVLYH XVH LQ VXUYHLOODQFH V\VWHPV PHGLFDO KRPH 7KHUH DUH VHYHUDO IDFWRUV WKDW LQIOXHQFH WKH GHVLJQLQJ DQG
DQG PLOLWDU\ RSHUDWLRQV 7KH VHQVRU QRGHV DUH GHSOR\HG LQ IXQFWLRQLQJ RI VHQVRU QHWZRUNV VXFK DV HQHUJ\ FRQVXPSWLRQ
H[WUHPHO\ ODUJH QXPEHU LQ WKHVH QHWZRUNV WR HQDEOH UHOLDEOH
FRVW IDXOW WROHUDQFH FRPPXQLFDWLRQ FRQQHFWLRQ PDQDJHPHQW
VXUYHLOODQFH DQG HIIHFWLYH PRQLWRULQJ RI HQYLURQPHQW ,Q UHFHQW
\HDUV PRVW RI WKH UHVHDUFKHUV KDYH VKRZQ LQWHUHVW LQ GHVLJQLQJ DQG G\QDPLF QHWZRUN WRSRORJ\ >@ 2XW RI WKHVH WKH PRVW
HQHUJ\ HIILFLHQW SURWRFROV IRU JDWKHULQJ HVVHQWLDO GDWD IURP WKH FKDOOHQJLQJIDFWRULVWKHHQHUJ\FRQVWUDLQWDQGKDVDWWUDFWHGWKH
HQYLURQPHQW 7KLV LV EHFDXVH VHQVRU QRGHV DUH EDWWHU\ RSHUDWHG DWWHQWLRQ RI VHYHUDO UHVHDUFKHUV 7KH VHQVRU QRGHV DUH EDWWHU\
ZKLFK GHSOHWHV TXLFNO\ ZLWK HDFK RSHUDWLRQ &OXVWHUEDVHG RSHUDWHGDQGWKXVWKHLURSWLPDOXVHGHILQHVWKHQHWZRUNOLIHWLPH
SURWRFROVDUH NQRZQWR SURYLGHEHWWHUUHVXOWVWKDQRWKHUURXWLQJ 6LQFHWKHQRGHVDUHJHQHUDOO\GHSOR\HGLQKDUVKHQYLURQPHQWV
SURWRFROVLQWHUPVRIUHVLOLHQFHHQHUJ\HIILFLHQF\GDWDLQWHJUDWLRQ UHSODFLQJ RU FKDUJLQJ WKHLU EDWWHULHV LV QRW IHDVLEOH 7KXV
DQG VFDODELOLW\ ,Q WKLV SDSHU ZH GLVFXVV WKH PRVW HIIHFWLYH GHVLJQLQJHQHUJ\HIILFLHQWURXWLQJRUFRPPXQLFDWLRQDOJRULWKPV
KLHUDUFKLFDO FOXVWHULQJ SURWRFRO FDOOHG DV /($&+ /RZ (QHUJ\ EHFRPHVDQLQWHJUDOIDFWRUIRULPSURYLQJQHWZRUNOLIHWLPH
$GDSWLYH &OXVWHULQJ +LHUDUFK\ DORQJ ZLWK LWV LVVXHV DQG
&OXVWHULQJLVRQH RIWKH PRVWFRPPRQDSSURDFKHV XVHGIRU
GUDZEDFNV$QH[KDXVWLYHVHDUFKRQYDULDQWVRI/($&+SURWRFRO
KDV EHHQ FRQGXFWHG :H SUHVHQW WKH WD[RQRP\ RI YDULRXV HQHUJ\DZDUHURXWLQJLQ:61V>@,QWKHSURFHVVRIFOXVWHULQJ
GHVFHQGDQWVRI/($&+SURWRFRODQGFRPSDUHWKHLUSHUIRUPDQFHV VHQVRUQRGHVDUH FRPELQHGWRJHWKHUWRIRUP JURXSVNQRZQDV
EDVHGRQPHWULFVVXFKDVVFDODELOLW\GDWDDJJUHJDWLRQPRELOLW\HWF FOXVWHUV 7KH QRGHV WKDW IRUP WKH FOXVWHU DUH FDOOHG PHPEHU
QRGHVRUFOXVWHU PHPEHUV,QHDFKFOXVWHURXWRIWKH PHPEHU
.H\ZRUGV:LUHOHVV6HQVRU1HWZRUNVHQHUJ\HIILFLHQF\/($&+ QRGHVRQHQRGHLVFKRVHQDVKHDGRIWKHFOXVWHU0DVWHU1RGH
YDULDQWVRI/($&+FOXVWHULQJ RU &OXVWHU +HDG &+ E\ XVLQJ DSSURSULDWH DSSURDFKHV 7KH
PHPEHU QRGHV VHQG WKHLU GDWD WR WKH PDVWHU QRGH ZKLFK
, ,1752'8&7,21 SHUIRUPVGDWDIXVLRQRSHUDWLRQVWRUHPRYHGDWDUHGXQGDQF\DQG
JHWUHOHYDQWLQIRUPDWLRQ,QWKLVZD\OHVVGDWDLVWUDQVPLWWHGWR
:LUHOHVV 6HQVRU 1HWZRUNV :61 DUH DPRQJ WKH PRVW WKHEDVHVWDWLRQDQGFRQVHTXHQWO\OHVVHQHUJ\LVVSHQWLQGDWD
SRZHUIXO WHFKQRORJLHV RI VW FHQWXU\ DQG KDYH JDLQHG WUDQVPLVVLRQ$W\SLFDOQHWZRUNFOXVWHULQJLVGHSLFWHGLQWKH)LJ
ZRUOGZLGH DWWHQWLRQ $V D UHVXOW RI DGYDQFHPHQWV LQ 0(06 ,QWKLVSDSHUZHSUHVHQW/($&+SURWRFROWKHPRVWHIIHFWLYH
WHFKQRORJ\WKHUHLVHYROXWLRQRIVHQVRUQRGHVZLWKORZSRZHU KLHUDUFKLFDO URXWLQJ SURWRFRO DQG LWV YDULRXV GHVFHQGDQWV WKDW
OHVV FRVW DQG VPDOO VL]H WKDW DUH GHSOR\HG LQ KXQGUHGV RU KDYHEHHQGHYHORSHG
WKRXVDQGV RI QXPEHUV IRU HQDEOLQJ KLJKTXDOLW\ DQG IDXOW
WROHUDQW VHQVLQJ QHWZRUNV 7KHVH QHWZRUNV DUH JHQHUDOO\
DSSOLFDWLRQ VSHFLILF DQG ILQG WKHLU XVH LQ VHYHUDO DSSOLFDWLRQ
ILHOGV OLNH GHIHQVH PHGLFDO KRPH EDWWOHILHOG VXUYHLOODQFH
GLVDVWHU PDQDJHPHQW XQGHUZDWHU VXUYHLOODQFH LQGXVWULDO
SURFHVVFRQWUROIRUHVWILUHFRQWURODQGPDQ\PRUH'XHWRWKH
LQWHQVLYH XVH RI:LUHOHVV6HQVRU1HWZRUNVWKLVDJH LVULJKWO\
VDLGDVWKH³$JHRI6HQVRUV´
7KHVHQVRUQRGHVKDYHVHQVLQJGHYLFHVFDOOHGDVVHQVRUVWKDW
PHDVXUH FKDQJHV LQ WKH HQYLURQPHQW 7KHVH FKDQJHV DUH
UHFRUGHGDQGVHQWWRWKHVLQN1RWRQO\VHQVRUQRGHVFROOHFWWKH
)LJ&OXVWHUIRUPDWLRQLQ:61V
GDWD EXW DOVR UHILQH GDWD XVLQJ WKHLU OLPLWHG FRPSXWDWLRQDO
FDSDFLW\ WR PDNH GDWD PHDQLQJIXO E\ FRPPXQLFDWLQJ WKURXJK 7KH UHVW RI WKLV SDSHU LV DUUDQJHG LQ WKH IROORZLQJ ZD\ ,Q
WUDQVFHLYHU7KHEDVHVWDWLRQSURFHVVHVWKHGDWDUHFHLYHGIURP 6HFWLRQ ,, UHODWHG ZRUN LV SUHVHQWHG 6HFWLRQ ,,, UHYLHZV WKH
c
978-1-5386-1719-9/18/$31.00 2018 IEEE 199
EDVLFV RI /($&+ SURWRFRO 7KH GHVFULSWLRQ RI LWV YDULDQWV LV x'DWDDJJUHJDWLRQIXVLRQVFKHPHVIRUGDWDFRPSUHVVLRQ
H[SODLQHG LQ 6HFWLRQ ,9 7KH SHUIRUPDQFH FRPSDULVRQ RI WRUHGXFHJOREDOFRPPXQLFDWLRQ
/($&+ SURWRFRO DQG LWV YDULDQWV LV GHVFULEHG LQ 6HFWLRQ 9
/DVWO\ 6HFWLRQ 9, FRQFOXGHV WKH ZRUN DORQJ ZLWK JLYLQJ WKH
IXWXUHGLUHFWLRQV
,, 5(/$7(':25.
$VLJQLILFDQWDPRXQWRIUHVHDUFKKDVEHHQGRQHLQWKHILHOG
RI URXWLQJ LQ :61 6HYHUDO UHVHDUFKHUV KDYH HYDOXDWHG DQG
DQDO\]HG WKH URXWLQJ SURWRFROV DQG FRPSDUHG WKHLU
SHUIRUPDQFHV :KHQ LW FRPHV WR FOXVWHULQJ SURWRFROV PDQ\ )LJ/($&+3URWRFRO3URFHVV
UHVHDUFKHUV KDYH UHYLHZHG WKH H[LVWLQJ DSSURDFKHV $OVR WKH
PRVW EDVLF /($&+ 3URWRFRO KDV EHHQ H[WHQGHG WR IRUP LWV $ $:RUNLQJRI/($&+3URWRFRO
VHYHUDO YDULDQWV ZKLFK IRUPV WKH EDVLV RI RXU UHYLHZ
%
$EEDVLHWDO>@VXUYH\HGRQYDULRXVFOXVWHULQJDSSURDFKHVDQG /($&+SURWRFROUXQVLQVHYHUDOURXQGVZKHUHHYHU\URXQG
DUUDQJHG FOXVWHULQJ RQ WKH EDVLV RI QHWZRUN PRGHO FOXVWHULQJ FRPSULVHVRIWZRSKDVHVQDPHO\WKHVHWXSSKDVHDQGWKHVWHDG\
REMHFWLYHV DQG WKH RSHUDWLRQDO PRGHO 5 +DQL HW DO >@ KDYH VWDWHSKDVH>@GHVFULEHGLQGHWDLOEHORZ
JLYHQDGHVFULSWLRQDERXWWKH /($&+EDVHGURXWLQJSURWRFROV 6HWXS3KDVH
DORQJZLWKJUDSKLFDOUHSUHVHQWDWLRQVRIHQHUJ\GLVVLSDWLRQDQG ,QWKLVSKDVHVHOHFWLRQRI&+VWDNHSODFH)RUHYHU\QRGHµQ¶
QHWZRUN OLIHWLPH 9 .XPDU HW DO >@ SURYLGHG D WD[RQRP\ RI D UDQGRP YDOXH µ[¶ LV FKRVHQ ZKLFK LV FRPSDUHG DJDLQVW D
FOXVWHULQJSURWRFROVDQGGHVFULEHGHDFKRQHRIWKHP,WDOVRJDYH 7KUHVKROGIXQFWLRQµ7Q¶DVGHFODUHGLQ(T,I[KDSSHQV
WKHGHVFULSWLRQRIYDULDQWVRI/($&+DORQJZLWKWKHLUOLVWLQJLQ WREHOHVVWKDQ7QLWEHFRPHVD&+HOVHLWEHFRPHVDPHPEHU
WDEXODUIRUP QRGH
భ
݂݅݊߳ܩ
,,, %$&.*5281'2)/($&+35272&2/
7Q ቐ ଵିቆௗቀ ቁቇ
ቑ
/($&+ D KLHUDUFKLFDO FOXVWHULQJ URXWLQJ SURWRFRO KDV Ͳ݁ݏ݅ݓݎ݄݁ݐ
WKHFKDUDFWHULVWLFRIEHLQJVHOIRUJDQL]HGDQGDGDSWLYH7KH
SUHYLRXVDSSURDFKHVWRURXWLQJZHUH'LUHFW&RPPXQLFDWLRQ ZKHUHSUHIHUVWRSHUFHQWDJHRI&+VUHTXLUHGLQWKHQHWZRUNU
DQG 0LQLPXP 7UDQVPLVVLRQ (QHUJ\ ZLWK PXOWLKRS EDVHG UHIHUVWRRQJRLQJURXQGDQG*LVWKHVHWRIWKRVHQRGHVZKLFK
5RXWLQJDQGWKHQ+HLQ]HOPDQFDPHXSZLWKDQHZDSSURDFK GLGQRWEHFRPH&+VLQODVWSURXQGV
RI/($&+ZKLFKZDVRIDPXFKEHWWHUSHUIRUPDQFH7KLV 7KHQRGHVWKDWKDYHQRWEHHQVHOHFWHGDV&+VKDYHWREHFRPH
SURWRFRODLPVDWH[WHQGLQJWKHQHWZRUNOLIHWLPHE\UHGXFLQJ SDUWRIDQ\FOXVWHU7KH\GRVRE\FRPSDULQJWKHVLJQDOVWUHQJWKV
HQHUJ\ GLVVLSDWLRQ LQ GDWD WUDQVPLVVLRQ ZLWK WKH KHOS RI RIDGYHUWLVHPHQWPHVVDJHVWKH\JHWIURP&+V7KHQRGHMRLQV
FOXVWHULQJDQGGDWDIXVLRQWHFKQLTXHV FKRRVHVWKDW&+IRUZKLFKWKHODUJHVWVLJQDOVWUHQJWKLVUHFHLYHG
)LUVWO\ FOXVWHUV DUH VHW XS LQ ZKLFK RQH RI WKH QRGHV LV ,QWKLVZD\FOXVWHUVDUHIRUPHG
UDQGRPO\FKRVHQWREHFRPHWKH&OXVWHU+HDG&+DQGWKH V
UHPDLQLQJ QRGHV EHFRPH PHPEHU QRGHV RI WKH FOXVWHU 6WHDG\6WDWH3KDVH
,QVWHDG RI VHQGLQJ GDWD GLUHFWO\ WR %DVH 6WDWLRQ %6 WKH 7KH &+ IRUPV D 7'0$ VFKHGXOH IRU LWV PHPEHU QRGHV WR
PHPEHUQRGHVVHQGGDWDWRWKHLUUHVSHFWLYH&+V7KH&+V VHQGGDWD7KH PHPEHUQRGHVVHQGGDWDLQWKHLUWLPHVORWDQG
UHGXFHVLJQDOVUHFHLYHGIURPFOXVWHUPHPEHUVLQWRDVLQJOH VOHHS LQ UHVW RI WKH VORWV $IWHU UHFHLYLQJ WKH VLJQDOV WKH &+
VLJQDO ZKLFK LQFRUSRUDWH QHFHVVDU\ LQIRUPDWLRQ RI DOO WKH DSSOLHV GDWD DJJUHJDWLRQ VFKHPHV VR WKDW VHYHUDO VLJQDOV DUH
VLJQDOV7KH&+VUHTXLUHPRUHSRZHUDVFRPSDUHGWRFOXVWHU FRQYHUWHGLQWRDVLQJOHVLJQDOZKLFKLQFRUSRUDWHVUHOHYDQWDQG
PHPEHUVIRUSHUIRUPLQJGDWDIXVLRQDQGVHQGLQJGDWDWR%6 QHFHVVDU\ LQIRUPDWLRQ RI DOO WKH VLJQDOV 7KLV VLJQDO LV
7KXV WKHUH LV UDQGRP URWDWLRQ RI WKH &+V VR WKDW WKHUH LV
WUDQVPLWWHGWRWKH%6E\DOOWKH&+VLQVLQJOHKRSWUDQVPLVVLRQ
HYHQGLVWULEXWLRQRIHQHUJ\ORDGLQWKHQHWZRUNDQGDQ\RQH
VHQVRU QRGH GRHV QRW GLH GXH WR WKLV HQHUJ\LQWHQVLYH
RSHUDWLRQ ,Q WKLV ZD\ LW QRW RQO\ PLQLPL]HV WKH HQHUJ\ %'UDZEDFNVRI/($&+3URWRFRO
FRQVXPSWLRQ EXW DOVR RSWLPL]HV WKH EDQGZLGWK XWLOL]DWLRQ
E\UHGXFLQJWKHGDWDWUDQVPLWWHGWRWKH%6)LJVKRZVWKH $OWKRXJK/($&+SURWRFROKDVEHFRPHYHU\SRSXODULWKDV
SURFHVVRI/($&+SURWRFRO7KHVDOLHQWIHDWXUHVRI/($&+ VHYHUDOGLVDGYDQWDJHVGHVFULEHGEHORZ
SURWRFRODUHDVIROORZV x5DQGRPL]HG VHOHFWLRQ RI FOXVWHU KHDGV 7KH &+V DUH
x5RWDWLRQRIFOXVWHUKHDGVLQUDQGRPL]HGPDQQHULQRUGHU FKRVHQ UDQGRPO\ XVLQJ WKH VWRFKDVWLF DSSURDFK ZLWKRXW
WRHYHQO\VSUHDGHQHUJ\ORDGDPRQJWKHQRGHV WDNLQJUHPDLQLQJHQHUJ\RIWKHQRGHVLQWRDFFRXQW
x$GDSWLYHDQGVHOIRUJDQL]HGFOXVWHUIRUPDWLRQ x,UUHJXODUDQGUDQGRPFOXVWHUIRUPDWLRQ&OXVWHUKHDGVDUH
x/RFDOL]HG FRQWURO RQ FOXVWHU VHWXS DQG GDWD QRW XQLIRUPO\ GLVWULEXWHG ZKLFK OHDGV WR QRQXQLIRUP
WUDQVPLVVLRQ FOXVWHUIRUPDWLRQ
x8VHRI7'0$VFKHPHVDQG&'0$FRGHVIRUUHGXFLQJ x6LQJOHKRS FRPPXQLFDWLRQ IURP FOXVWHU KHDGV WR EDVH
LQWUDFOXVWHUDQGLQWHUFOXVWHULQWHUIHUHQFHV VWDWLRQ (QHUJ\ LV H[SHFWHG WR HQG VRRQ IRU WKH &+V IRU
200 2018 8th International Conference on Cloud Computing, Data Science & Engineering (Confluence)
ZKLFKWKHLUGLVWDQFHIURPWKH%6LVPRUH7KLVLVGXHWR DSSURDFKLVQRWPXFKHIILFLHQWDVLWGRHVQRWFRQVLGHUVPRELOLW\
VHQGLQJRIGDWDIRUP&+WR%6LQVLQJOHKRS DQGG\QDPLFQDWXUHRIVHQVRUQHWZRUNV
x&RQWLQXRXV0RQLWRULQJ/($&+LVRQO\VXLWDEOHIRUWKRVH
DSSOLFDWLRQV ZKLFK UHTXLUH FRQWLQXRXV PRQLWRULQJ )RU /($&+'&+6 'HWHUPLQLVWLF &OXVWHU +HDG 6HOHFWLRQ
HYHQWGULYHQDSSOLFDWLRQVDQG'DWD'LIIXVLRQDSSOLFDWLRQV /($&+$GHWHUPLQLVWLFDSSURDFK>@LVXVHGIRU&+VHOHFWLRQ
LWLVQRWVXLWDEOH LQWKLVSURWRFRO,QVWRFKDVWLF &+VHOHFWLRQDSSURDFKWKH&+V
x3UHGHILQHG 1XPEHU RI &OXVWHU KHDGV /($&+ PD\ DUH QRW SURSHUO\ DUUDQJHG LQ WKH QHWZRUN WKXV JUDGXDOO\
GHYLDWHIURPFRYHULQJWKHHQWLUHQHWZRUNSURSHUO\WKURXJK GLVWXUELQJ WKH HQHUJ\ EDODQFH /($&+'&+6 PRGLILHV WKH
WKHSUHGHILQHGQXPEHURI&+VDV WKH &+ORFDWLRQ LV QRW WKUHVKROG YDOXH IRU D QRGH ZLWK WKH KHOS RI GHWHUPLQLVWLF
HYHQO\GLVWULEXWHG$OVRWKHQXPEHURI&+VPD\GHYLDWH FRPSRQHQWDQGEULQJVEHWWHUUHVXOWVWKDQ/($&+
IURPWKLVSUHGHILQHGQXPEHU
x9DULDEOHVL]HRIFOXVWHUV7KHUHLVSRVVLELOLW\WKDWWKHVL]H /($&+% %DODQFHG /($&+ ,W LV D GHFHQWUDOL]HG
RI VRPH FOXVWHUV LV YHU\ ODUJH DQG IRU VRPH FOXVWHUV WKH DOJRULWKP LQ ZKLFK QRGHV NQRZ DERXW WKHLU SRVLWLRQ DQG WKH
VL]HLVYHU\VPDOO SRVLWLRQRIWKHH[WHUQDOUHFHLYHU,WDVVXPHVWKDWWKHUHZLOOEHQR
GDWDDJJUHJDWLRQDW&+V,Q/($&+%WKHUHDOGLVVLSDWLRQRI
,9 9$5,$1762)/($&+35272&2/ HQHUJ\ E\ WKH QRGHV LH WKUHVKROG HQHUJ\ OHYHO 7 GHFLGHV
G\QDPLFDOO\ WKH QXPEHU RI FOXVWHUV µ1¶ 7KXV WKHUH LV EHWWHU
6HYHUDO GHVFHQGDQWV RI /($&+ SURWRFRO WKDW H[LVW LQ GLVWULEXWLRQRIHQHUJHWLFH[SHQVLYHUROHRI&+
OLWHUDWXUH FDQ EH FODVVLILHG RQ VHYHUDO SDUDPHWHUV VXFK DV
PRELOLW\ FHQWUDOL]DWLRQ KRPRJHQHLW\ VWDELOLW\ VHFXULW\ $/($&+ $GYDQFHG /($&+ 7KLV SURWRFRO DLPV DW
UHOLDELOLW\HWFEXWWKHPRVWDSSURSULDWHFODVVLILFDWLRQZRXOGEH GLVWULEXWLQJ HQHUJ\ GLVVLSDWLRQ DPRQJ QRGHV LQ D EDODQFHG
RQWKHEDVLVRIWKHWKUHHVWDJHVRI/($&+QDPHO\VHOHFWLRQRI PDQQHU E\ PDNLQJ PRGLILFDWLRQV LQ WKH WKUHVKROG IXQFWLRQ 7
&+V IRUPDWLRQ RI FOXVWHUV DQG GDWD WUDQVPLVVLRQ >@ $ QXVHGLQWKHFOXVWHUKHDGVHOHFWLRQDOJRULWKP7KHWKUHVKROG
PLVFHOODQHRXVFDWHJRU\DOVRH[LVWVWKDWLQFOXGHVVHFXULW\DZDUH IXQFWLRQLQFOXGHVWZRIDFWRUV*SDQG&6S&6SUHIHUVWR&XUUHQW
/($&+LPSURYHGYHUVLRQVDQGWKRVHGHVFHQGDQWVWKDWPDNHXVH 6WDWH3UREDELOLW\ZKHUHDV*S UHIHUVWR*HQHUDO3UREDELOLW\>@
RIVRIWFRPSXWLQJDSSURDFKHVIRUHQHUJ\HIILFLHQF\ 7KH7KUHVKROGIXQFWLRQLVJLYHQE\(T
7Q *S&6S
ா̴௨௧
ಿ ቀ כቁ
ேିቀௗ ቁ ா̴௫ ே
಼
+HUH NUHIHUVWR QXPEHURI &+VU UHIHUVWRSUHVHQWURXQG1
UHIHUV WR WRWDO QRGHV LQ WKH QHWZRUN (QBFXUUHQW UHIHUV WR QRGH¶V
FXUUHQWHQHUJ\DQG(QBPD[UHIHUVWRQRGH¶VLQLWLDOHQHUJ\
9 /($&+ 9LFH &OXVWHU +HDG /($&+ 7KLV SURWRFRO
EULQJVWKHFRQFHSWRIDVHFRQGDU\&+9LFH&+>@WKDWZRUNV
DV&+ZKHQWKH&+GLHV,QWKLVZD\GDWDIURPPHPEHUQRGHV
LV VWLOO DEOH WR UHDFK WKH %6 WKHUHE\ QRW PDNLQJ WKH FOXVWHU
XVHOHVV)LJGHVFULEHV9/($&+
)LJ&ODVVLILFDWLRQRI/($&+9DULDQWV
$ &OXVWHU+HDG6HOHFWLRQEDVHG3URWRFROV
/($&+& &HQWUDOL]HG /($&+ ,W XVHV D FHQWUDOL]HG
DSSURDFK IRU EHWWHU GLVWULEXWLQJ WKH &+V LQ WKH QHWZRUN 7KH
LQIRUPDWLRQDERXWORFDWLRQRIWKHQRGHVDQGWKHLUHQHUJ\OHYHOLV
VHQWWRWKH%67KHQWKH%6VHOHFWVWKRVHQRGHVIRUZKLFKWKH
UHVLGXDO HQHUJ\ LV PRUH WKDQ WKH DYHUDJH QRGH HQHUJ\
6LPXODWHG DQQHDOLQJ SURFHVV LV XVHG WR ILQG WKH RSWLPDO N
FOXVWHUV DQG FOXVWHU PHPEHUV MRLQ WKH FOXVWHUV GHSHQGLQJ RQ
PLQLPXPVTXDUHGGLVWDQFH )LJ9/($&+3URWRFRO
/($&+) )L[HG 1XPEHU RI &OXVWHUV /($&+ $QRWKHU /($&+7/&+7ZR/HYHO&OXVWHU+HDG/($&++HUHD
FHQWUDOL]HGDQGORFDWLRQEDVHGSURWRFROLV /($&+)LQZKLFK VHFRQGDU\&+LVIRUPHGEDVHGRQUHVLGXDOHQHUJ\DQGGLVWDQFH
IL[HGFOXVWHUVDUHIRUPHGWKURXJKVLPXODWHGDQQHDOLQJ7KHEDVH EHWZHHQ%6DQG&+,WWDNHVWKHUHVSRQVLELOLW\RI&+WRFROOHFW
VWDWLRQVHOHFWVWKHFOXVWHUVDQGWKH&+VDQGEURDGFDVWVFOXVWHU DQG IXVH GDWD VHQG WKLV IXVHG GDWD WR &+ ZKLFK XOWLPDWHO\
,'RIHDFKQRGH7KHILUVWQRGHLQWKHQRGH,'OLVWLVVHOHFWHGDV IRUZDUGVLWWR%67KXV7/&+DYRLGVWKHSUHPDWXUHGHDWKRI
WKHILUVW&+WKHVHFRQGQRGHDVWKH QH[W&+DQGVRRQ7KLV WKH&+VDQGH[WHQGWKHLUOLIHWLPH
2018 8th International Conference on Cloud Computing, Data Science & Engineering (Confluence) 201
/($&+0$& ,W LV D &+ VHOHFWLRQ SURFHGXUH WKDW PDNHV WKHVWHSVDUHVDPHDVWKDWRI/($&++HQFHWKHVWDELOLW\RIWKH
XVHRI&+DGYHUWLVHPHQWPHVVDJHVIRUGHFLGLQJFOXVWHUKHDGV QHWZRUNLVHQKDQFHGE\WKLVSURWRFRO
,QWKLVSURWRFROIRUHYHU\QRGHDXQLIRUPUDQGRPWLPHLQWHUYDO
5W LV VHOHFWHG ,I WKH QXPEHU RI &+ DGYHUWLVHPHQW PHVVDJHV & 'DWDWUDQVPLVVLRQEDVHG3URWRFROV
UHFHLYHG LQ WKLV WLPH LQWHUYDO LV OHVV WKDQ RSWLPDO QXPEHU RI 0XOWLKRS/($&+,Q/($&+WKHGLVWDQFHIURPWKH&+V
FOXVWHUKHDGVWKHQRGHEHFRPHVWKH&+HOVHWKHQRGHGHFODUHV WR WKH %6 LV QRW WDNHQ LQWR FRQVLGHUDWLRQ 7KXV WKHUH LV PRUH
LWVHOIDVDQRUPDOQRGH,WSURYHVWRJHWORQJHUQHWZRUNOLIHWLPH HQHUJ\GLVVLSDWLRQZKHQWKHGLVWDQFHLVPRUH7KXVPXOWLKRS
WKDQ/($&+ /($&+ DOORZV FRPPXQLFDWLRQ EHWZHHQ &+V DQG %6 YLD
PXOWLSOHKRSVE\ILQGLQJWKHRSWLPDOSDWKEHWZHHQWKH&+VDQG
% &OXVWHU)RUPDWLRQEDVHG3URWRFROV %6>@7KHSURFHVVRIPXOWLKRS/($&+LVVKRZQLQ)LJ
/($&+( (QHUJ\ /($&+ >@ ,Q WKLV SURWRFRO LI WKH
UHPDLQLQJHQHUJ\RIDQRGHLVJUHDWHUWKDQWKHVFKHGXOHGHQHUJ\
WKHQRGHEHFRPHVD&+HOVHLWEHFRPHVDFRPPRQQRGH7KHUHVW
RIWKHVWHSVDUHIROORZHGLQWKHVDPHPDQQHUDVLQ/($&+
)=/($&+ )DU=RQH /($&+ 7KLV SURWRFRO XVHV WKH
FRQFHSWRI0LQLPXP5HDFKDELOLW\3RZHU053IRUFRPSXWLQJ
WKRVH QRGHV IRU ZKLFK 053 LV PRUH WKDQ $YHUDJH 0LQLPXP
5HDFKDELOLW\ 3RZHU $053 6XFK QRGHV DSSHDU LQ )DU=RQH
)=7KHIRUPDWLRQRI)=WDNHVSODFHDIWHUWKHFOXVWHUIRUPDWLRQ
SKDVH (DFK )= KDV D &+ FDOOHG =RQH +HDG =+ 7KH ]RQH
PHPEHUV VHQG GDWD WR =+ DQG =+ WUDQVPLWV LW WR &+ 7KH )LJ0XOWLKRS/($&+SURWRFRO
UHPDLQLQJQRGHVVHQGGDWDWR&+V,QWKLVZD\HQHUJ\GHSOHWLRQ
IRUIDUGLVWDQFHVLVPLQLPL]HG 7//($&+ 7ZR /HYHO /($&+ 7KLV SURWRFRO EULQJV LQ
WKHFRQFHSWRIWZROHYHOKLHUDUFK\IRUWUDQVPLWWLQJGDWDWRWKH
&HOO /($&+ +HUH ZLWKLQ HDFK FOXVWHU WKHUH DUH VHYHQ %6 7KHUH DUH WRS OHYHO &+V SULPDU\ &+ VHFRQG OHYHO &+
VHFWRUV NQRZQ DV FHOOV >@ ZLWK HDFK FHOO KDYLQJ D FHOO KHDG VHFRQGDU\ &+ DQG ILQDOO\ VLPSOH QRGHV >@ 3DUWLDO DQG
VHOHFWHG UDQGRPO\ LQ WKH EHJLQQLQJ ,Q WKH VXEVHTXHQW SKDVHV FRPSOHWH ORFDO FRPSXWDWLRQ LV GRQH E\ VHFRQGDU\ &+ DQG
HDFK ROG FHOO KHDG VHOHFWV D QHZ FHOO KHDG G\QDPLFDOO\ XVLQJ SULPDU\&+UHVSHFWLYHO\7KXVGXHWRVPDOOGLVWDQFHVEHWZHHQ
UHVLGXDOHQHUJ\RIQRGHVDQGUHSODFHVLW7KHFHOOKHDGFROOHFWV WKHEDVHVWDWLRQDQGWKHFOXVWHUKHDGVWKHUHLVEDODQFHGHQHUJ\
DQG DJJUHJDWHV LQIRUPDWLRQ IURP FHOO PHPEHUV DQG ORDG7KHSURFHVVRI7//($&+LVVKRZQLQWKH)LJ
FRPPXQLFDWHV GLUHFWO\ ZLWK WKH FOXVWHU KHDG 7KH IROORZLQJ
ILJXUH)LJVKRZVKRZ&HOO/($&+ZRUNV
)LJ7//($&+
06/($&+ 0XOWLKRS DQG 6LQJOHKRS /($&+ 7KLV
SURWRFROEULQJVWRJHWKHUWKHFRQFHSWRIVLQJOHKRSDQGPXOWLSOH
KRS WUDQVPLVVLRQV ,I SUHVHQW FOXVWHU DUHD VL]H LV OHVV WKDQ WKH
)LJ)RUPDWLRQRIFHOOVLQ&HOO/($&+ WKUHVKROGFOXVWHUDUHDVL]HWKHUHZLOOEHQRDFWLRQLHWKHFOXVWHU
PHPEHUV ZLOO FRPPXQLFDWH LQ VLQJOH KRSV (OVH 'LMNVWUD
./($&+.PHGRLGVDOJRULWKPLVXVHGLQWKHILUVWURXQG DOJRULWKPLVXVHGWRFRPSXWHWKHPLQLPXPHQHUJ\FRQVXPSWLRQ
IRU FOXVWHU IRUPDWLRQ 7KLV HQVXUHV JRRG VHOHFWLRQ RI FOXVWHU SDWKWUHHZKLFKLVXVHGIRUPXOWLKRSFRPPXQLFDWLRQEHWZHHQ
KHDGVDQGXQLIRUPFOXVWHULQJ)URPWKHVXEVHTXHQWURXQGVWKH &+DQGPHPEHUQRGHV7KXVQRGHVZLOOQRWGHSOHWHWKHLUHQHUJ\
(XFOLGHDQGLVWDQFHDSSURDFKGHFLGHVWKHQH[W&+$IWHUJHWWLQJ GXHWRORQJGLVWDQFHVLQWKHFOXVWHU
GXSOLFDWH &+V G\QDPLF UHFOXVWHULQJ RFFXUV WKDW VHOHFWV &+V
EDVHGRQDOLYHQRGHV
' 0LVFHOODQHRXV3URWRFROV
4/($&+4XDGUDWXUH/($&+7KLVSURWRFROZRUNVE\
6RIW&RPSXWLQJEDVHG3URWRFROV
GLYLGLQJ WKH QHWZRUN LQWR IRXU VHFWRUV FDOOHG DV TXDGUDQWV
D/($&+)/)X]]\/RJLF/($&++HUHWKH&+VHOHFWLRQ
,QLWLDOO\WKHORFDWLRQDERXWWKHQRGHVLVVXEPLWWHGWR%6ZKLFK
PHWKRGLVGRQHXVLQJIX]]\ORJLFWKDWPDNHVXVHRIWKUHHIDFWRUV
RUJDQL]HVWKHQRGHVLQWRIRXUTXDGUDQWV2QWKHEDVLVRIWKUHVKROG
GLVWDQFH QRGH GHQVLW\ DQG EDWWHU\ OHYHO ,W FRQVLVWV RI WKUHH
YDOXHVRPHQRGHVDUHHOHFWHGDV&+VLQHDFKGLYLVLRQ7KHQRGHV
PRGXOHV IRU WKLV SURFHGXUH WKH )X]]LILFDWLRQ 0RGXOH WKDW
FKRRVHWKHLU&+EDVHGRQWKHSULQFLSOHRUSUR[LPLW\7KHUHVWRI
WUDQVIRUPV WKH WKUHH LQSXW SDUDPHWHUV LQWR IX]]\ VWDWHV WKH
202 2018 8th International Conference on Cloud Computing, Data Science & Engineering (Confluence)
,QIHUHQFH(QJLQH.QRZOHGJH%DVHWKDWFRQVLVWVRIUXOHVDQG ILQGSDFNHWZKLFKGHQRWHVKRZUHOLDEOHLWLVWRIRUZDUGSDFNHWV
WKH'HIX]]LILFDWLRQ0RGXOHWRJHWEDFNWKHFULVSYDOXH,WVKRZV WKURXJKDSDUWLFXODUQHLJKERUQRGH)RUVHOHFWLQJ&+VZHLJKWV
EHWWHUSHUIRUPDQFHWKDQ/($&+ DUH FRPSXWHG WKDW VLJQLI\ UHVLGXDO HQHUJ\ 7KH QRGHV ZKLFK
KDYHOHVVZHLJKWVDUHUHIUDLQHGIURPEHLQJHOHFWHGDV&+V
E%1 /($&+ %D\HVLDQ 1HWZRUNV /($&+ 7KLV LV D &+
VHOHFWLRQ SURFHGXUH WKDW XVHV %D\HVLDQ 1HWZRUNV DQG LW XVHV 9 3(5)250$1&(&203$5,621%(7:((1
WKUHH SDUDPHWHUV GLVWDQFH UHPDLQLQJ HQHUJ\ DQG GHQVLW\ RI /($&+35272&2/$1',769$5,$176
QRGHV8QLIRUPGLVWULEXWLRQRI&+VLVGRQHE\G\QDPLF]RQLQJ
DSSURDFKDQGJUHHG\PHWKRG7KHQRGHVDUHGLVWULEXWHGLQ]RQHV 7KHYDULDQWVRI/($&+IRFXVHGRQVHYHUDOSDUDPHWHUVVXFK
EDVHGRQWKHLUGLVWDQFHVIURPWKH%6,QHYHU\]RQHWKH QRGH DV H[WHQGLQJ WKH QHWZRUN OLIHWLPH EDODQFLQJ WKH HQHUJ\ ORDG
ZLWKKLJKHVWSUREDELOLW\DFFRUGLQJWRWKHFKDLQUXOHRI%D\HVLDQ GLVWULEXWLRQ LPSURYLQJ WKH VWDELOLW\ SHULRG HWF $ EULHI
1HWZRUNV EHFRPHV &+ LQ WKH FXUUHQW URXQG ,W KDV D ORQJHU FRPSDULVRQDPRQJWKHYDULRXVYDULDQWVRI/($&+LVJLYHQLQ
QHWZRUNOLIHWLPHWKDQ/($&+ 7DEOH , $OO WKHVH GHVFHQGDQWV SHUIRUPHG EHWWHU WKDQ WKH
FODVVLFDO/($&+SURWRFRO
F/($&+*$ *HQHWLF $OJRULWKP EDVHG /($&+ 7KLV
SURWRFROKDVDQDGGLWLRQDOSKDVHFDOOHGSUHSDUDWLRQSKDVH7KLV
SKDVH XVHV JHQHWLF DOJRULWKP IRU GHWHUPLQLQJ WKH RSWLPXP 9, &21&/86,21$1')8785(6&23(
SUREDELOLW\ RI WKH QRGHV IRU EHFRPLQJ &+V 7KH %6 UHFHLYHV
LQIRUPDWLRQ DERXW WKH QRGHV QRGH ,' DQG ORFDWLRQ DQG WKH 2QH RI WKH SULPDU\ GHVLJQ LVVXHV RI :61V LV HQHUJ\
RSWLPXPSUREDELOLW\RIEHLQJ&+VLVFDOFXODWHGDQGEURDGFDVWHG FRQVWUDLQW ZKLFK LV WDFNOHG E\ WKH UHVHDUFKHUV XVLQJ VHYHUDO
WR WKH QRGHV 7KLV UHGXFHV WKH HQHUJ\ FRQVXPSWLRQ IRU HQHUJ\ FRQVHUYLQJ DSSURDFKHV $PRQJ DOO RI WKHP WKH PRVW
FRPSOHWLQJRQHURXQGDQGKHQFHLQFUHDVHVWKHQHWZRUNOLIHWLPH HIIHFWLYH DSSURDFK LV FOXVWHULQJ /($&+ DQG LWV VHYHUDO
GHVFHQGDQWV WKDW KDYH HYROYHG ZLWK WLPH KDYH EHHQ GHVFULEHG
6HFXULW\$ZDUH3URWRFROV DQGUHYLHZHGLQWKLVSDSHU&ODVVLILFDWLRQRIWKHVHSURWRFROVRQ
WKHEDVLVRISKDVHVRI/($&+KDVEHHQSUHVHQWHG
D6/($&+ 6HFXUH /($&+ )RU EULQJLQJ VHFXULW\ LQ $ V\VWHPDWLF SHUIRUPDQFH FRPSDULVRQ EDVHG RQ VHYHUDO
/($&+ WKLV SURWRFRO PDNHV XVH RI RQH ZD\ KDVK FKDLQV SDUDPHWHUVVXFKDVKRSFRXQWVFDODELOLW\ORFDWLRQLQIRUPDWLRQ
,QVWHDGRIFRPSOH[DV\PPHWULFFU\SWRJUDSKLFRSHUDWLRQVLWXVHV HWFKDVEHHQFRQGXFWHG%XWVWLOOWKHUHLVVFRSHRILPSURYHPHQW
VLPSOHV\PPHWULFRSHUDWLRQVWRSUHYHQWVL[NLQGVRIDWWDFNVRQ LQWKLVEDVLFDQGYHU\LPSRUWDQWSURWRFROLH/($&+$OLWWOH
VHQVRUQHWZRUNVWKDWKDYHEHHQGHVFULEHGLQ6/($&+ ZRUNKDVEHHQGRQHWRLPSURYHWKHGDWDWUDQVPLVVLRQDQGGDWD
DJJUHJDWLRQ VFKHPHV ZKLFK FDQ IXUWKHU H[WHQG WKH QHWZRUN
E/($&+707KLVSURWRFROZRUNVRQPDLQWDLQLQJUHOLDELOLW\ OLIHWLPHRI:61V
RIWKHQHWZRUNE\D YDOXH7SDWKDYDOXHREWDLQHGIURP 5RXWH
7$%/(,
3(5)250$1&(&203$5,621$021*/($&+9$5,$176
3URWRFRO <HDU 'DWD7UDQ 6HOI2UJD 'LVWULE %6)L[HG 6FDODELOLW\ 'DWD$JJ 1HHGRI +RPR
VPLVVLRQ QL]DWLRQ 8WHG 0RELOH UHJDWLRQ /RFDWLRQ JHQHRXV
,QIRUPDWLRQ
/($&+ 6LQJOHKRS <HV <HV )L[HG /RZ <HV 1R <HV
/($&+ 6LQJOHKRS <HV <HV )L[HG 0HGLXP <HV 1R <HV
'&+6
/($&+& 6LQJOHKRS <HV 1R )L[HG 0HGLXP <HV <HV <HV
/($&+) 6LQJOHKRS 1R 1R )L[HG /RZ <HV <HV <HV
/($&+% 6LQJOHKRS <HV <HV )L[HG 0HGLXP 1R <HV <HV
7//($&+ 0XOWLKRS <HV <HV )L[HG +LJK <HV 1R <HV
6/($&+ 6LQJOHKRS <HV <HV )L[HG 0HGLXP <HV 1R <HV
(7/($&+ 6LQJOHKRS <HV <HV )L[HG 0HGLXP <HV 1R <HV
/($&+( 6LQJOHKRS <HV <HV )L[HG 0HGLXP <HV 1R <HV
0+/($&+ 0XOWLKRS <HV <HV )L[HG 0HGLXP <HV 1R <HV
$/($&+ 6LQJOHKRS <HV <HV )L[HG +LJK <HV 1R <HV
2018 8th International Conference on Cloud Computing, Data Science & Engineering (Confluence) 203
7%/($&+ 6LQJOHKRS <HV <HV )L[HG 0HGLXP <HV 1R <HV
/($&+0 6LQJOHKRS <HV <HV 0RELOH%6 +LJK <HV <HV <HV
DQGQRGHV
/($&+70 0XOWLKRS <HV <HV )L[HG 0HGLXP <HV <HV <HV
/($&++ 6LQJOHKRS <HV <HV )L[HG 0HGLXP <HV <HV <HV
06/($&+ 0XOWLKRS <HV <HV )L[HG +LJK <HV <HV <HV
9/($&+ 6LQJOHKRS <HV <HV )L[HG 0HGLXP <HV <HV <HV
/($&+)/ 6LQJOHKRS <HV 1R )L[HG /RZ <HV <HV <HV
05/($&+ 0XOWLKRS <HV <HV )L[HG 0HGLXP <HV <HV <HV
)=/($&+ 0XOWLKRS <HV <HV )L[HG 0HGLXP <HV <HV <HV
/($&+*$ 6LQJOHKRS <HV <HV )L[HG 0HGLXP <HV <HV <HV
&HOO/($&+ 0XOWLKRS 1R <HV )L[HG 0HGLXP <HV <HV <HV
4/($&+ 6LQJOHKRS 1R <HV )L[HG 0HGLXP <HV <HV <HV
./($&+ 6LQJOHKRS <HV <HV )L[HG 0HGLXP <HV <HV <HV
/($&+ 6LQJOHKRS <HV <HV )L[HG 0HGLXP <HV 1R <HV
7/&+
/($&+%1 6LQJOHKRS <HV <HV )L[HG 0HGLXP <HV <HV <HV
/($&+0$& 6LQJOHKRS <HV <HV )L[HG +LJK <HV 1R <HV
9,, 5()(5(1&(6
>@$EEDVL$<RXQLV0$VXUYH\RQFOXVWHULQJDOJRULWKPVIRU
ZLUHOHVVVHQVRUQHWZRUNV&RPSXWHU&RPPSS
>@ $N\LOGL] ,) 6X : 6DQNDUDVXEUDPDQLDP < &D\LUFL (
:LUHOHVV6HQVRU1HWZRUNV$VXUYH\&RPSXWHU1HWZRUNVSS
>@ +HLQ]HOPDQ : % &KDQGUDNDVDQ $ %DODNULVKQDQ + $Q
$SSOLFDWLRQ VSHFLILF SURWRFRO DUFKLWHFWXUH IRU ZLUHOHVV PLFURVHQVRU
QHWZRUNV7UDQV2Q:LUHOHVV&RPPSS,(((
>@ +DQL5 ,MMHK $ $ VXUYH\ RQ /($&+ EDVHG HQHUJ\ DZDUH
SURWRFROVIRUZLUHOHVVVHQVRUQHWZRUNV-&RPPXQSS
>@ .XPDU 9 -DLQ 6 7LZDUL 6 (QHUJ\ HIILFLHQW FOXVWHULQJ
DOJRULWKPVLQZLUHOHVVVHQVRUQHWZRUNVDVXUYH\,-&6,SS
>@+DQG\0+DDVH07LPPHUPDQQ'/RZ(QHUJ\$GDSWLYH
&OXVWHULQJ +HLUDUFK\ ZLWK 'HWHUPLQLVWLF &OXVWHU +HDG 6HOHFWLRQ
0RELOHDQG:LUHOHVV&RPP1HWZRUNWK,QW¶O:RUNVKRSRQSS
,(((
>@ $OL 06 'H\ 7 %LVZDV 5 'HF $/($&+ $GYDQFHG
/($&+ URXWLQJ SURWRFRO IRU ZLUHOHVV PLFURVHQVRU QHWZRUNV ,Q
,&(&(,QW¶O&RQIRQSS,(((
>@ <DVVHLQ 0% .KDPD\VHK < 0DUGLQL : -XQH
,PSURYHPHQW RQ /($&+ SURWRFRO RI ZLUHOHVV VHQVRU QHWZRUNV
9/($&+,Q,QW¶O-'LJLW&RQWHQW7HFKQRORJ\$SSOLFDWLRQV
>@ <HNWDSDUDVW $ 1DEDYL )+ 6DUPDVW $ )HE $Q
,PSURYHPHQWRQ/($&+3URWRFRO&HOO/($&+,Q,&$&7WK,QW¶O
&RQIRQSS,(((
>@ /RVHUL 9 0RUHELWR * 0DUDQR6 6HSW $ WZR OHYHOV
KLHUDUFK\IRUORZHQHUJ\DGDSWLYHFOXVWHULQJKHLUDUDFK\7//($&+
,Q 9HKLFXODU 7HFKQRORJ\ &RQIHUHQFH SS
,(((
>@ /L ; = =KDQJ $/ /LDQJ <= 6HSW ,PSURYHPHQW RI
/HDFKSURWRFROIRUZLUHOHVVVHQVRUQHWZRUNV,Q,0&&&UG,QW¶O
&RQIRQSS,(((
204 2018 8th International Conference on Cloud Computing, Data Science & Engineering (Confluence)