xây dựng thử chương trình tách thuật ngữ tiếng việt theo phương pháp cổ điển
Bạn đang xem bản rút gọn của tài liệu. Xem và tải ngay bản đầy đủ của tài liệu tại đây (375.5 KB, 42 trang )
Website: Email : Tel ( : 0918.775.368
MỤC LỤC
: !""#$%!
&'()*+(,,-(,,./0
12)232,245#
26789:"
22;<=>?@!
ABCDE!
ABFC>*EG!
ABFC>HIDEG
J&KBLMHIDENOG
ABCLMHI
BP;EBLMHIJ
QRSET9UL8V9:J
)I>WLXKL8V9:J
JQYZ[KYLCVBN\E]$
J^P;EBL?_>?@9`*NG0
$&KB>WaE2bNNNc2dG#
ABC>WaE#
BP;EB>WaE"
BP;EB>e"
af&"
JafgaEEc;hNchQG
JBP;EBYNi>CLM+2cGJJ
JAj>WME(;cN(>NNgGJ$
0k]aESLiHI@E+J"
222BEIBBSEhDE$
67NPEVHI@EhC$
B9;`V@EhC$
@EK979UV@E$
@EKEBN`EFL$
J@EKEBN`EFEl$
$@EKEBN`EFB$
JTKB979UVT$
JTK9;`m89U97L>$
JJTnElKoK8>Mp9`$
$L>KB979UVL>$
$L>n=ElKo$
$JL>n8>M9ME$
BP;ECEFBV@EC$J
NEM8>MT$J
JNEM8>ML>$J
T@EC$$
T9;cTEq$$
JTM$$
$*ZET8>MEF$0
L>@EC$#
L>9;$#
JL>Eq$%
JL>EqEE$%
JJL>Eq_>M$%
J$BKrL>$%
B979UiIKHI@EhC$"
JBEIBs9BEBC>_>Is9:NEIBLiHI@E
hC$"
&KBLMHI@EhC$"
&KB>WaETHI@EhC$!
2htL?YERP;ENfB>WEF@EhCP;EBp
9U$
P;ENfKKB9P[EI_>?@$
JA@_>IM?P;ENf$
u2245,+A,'.0
^,0
BaE:BB0
BM?P;ENf0
J
v62w(,5x()y0J
$
BẢNG DANH MỤC HÌNH HOẠ
,fkiS:L?_>?@9`
,fJaf>WaE>e
,f$6z`U>{Bc;VKB
K
,f06z`U>{_>CEFN>?8_>N?GKBKC>*s*J
,f#ME;cNBR+(*GKBR.dG
,f%ME;cNXXeBR(.d
,f"afU>{ME;cN
,f!MaESD+?
,f MaESD+?|EN}2NDNN
0
LỜI GIỚI THIỆU
(EK??s;<FC>9~N<KkrgaEU@>VD~kK
EP•NEgoE>?}aEK?sBaE9P[P>NFKDR=C>_>Ir>
@KaE_>;<FC>Q>Er#HBNUs;<FC>9~nFEPX
@aZE_>N\ENE`RaEECaETaf;<FC>
_>C9:D>8TFEH%sB€ESEaEECaE9~Y
Y@CP>NFFC>N<KP>NFaEaE_>BaES_>I=K
DR=;<FC>(EK??s>r>P>NFKDR=aEn7<g•\;
‚8€kp€KsX8gƒkaf?_>?aK„E9:>nFE
>r>:P>NFKgBaEABCaE<9L?EzIaE
:kMVp€KaE:aNP•EKp€M9kE
hCE}€>=>?@:;<F}>9~N<KkEKg\€E
SE*FE@kP[WNEE}€>=>?@„EPK97Y@sB
C_>IN`;<FC>9~N<K:IEsKr]?@>NEM9kEVBp
€(•|EKBp€M9kEC>_>I;hC€ESE;<FC>9~
E|KEIN8:>aE€9kEVEP•K•9nC>>89kE
V\;,C_>IN`;<FC>EK??gaE…9;>roKk;
8>qP>NF]C>K…g†9nKBaESsCi?BP;E
B>W9U>?U9p]C>KaEW8IBaESEP•ZE
BNU7BK>E8r:>EN9USSS9iM
9kEVp€s9P[]P>FE?}>r>ECSVp€9P[E\KB
€ESE‡N[DR=BEC;FsgB>r>RSEaE<€
8;9U‡N[B>r>LiVBK~9MsBKW@P[
NEkp€skM€ESEXN9•SSBS9iK?X}
E\ˆCLiKDR=NY>?@‰‚B€ESEK?saE9P[P>NFsDR=
Kg@D>8BS9iSUPXMEPXEV9:(•BaE<ME
K?KBLisBK~nU9PNB_>?@9`M9kEkBC>
_>I8
ABafFC>BNU<€9k;sBaEP>NFPX
MEFC>E|9ME;sEP•WN…N8:>N€…:e
NEFC>KB€LiNPX9ngaEBCN=V89:K?K
BLiNPX9noXPXES9iSUVEP•BS9iK?
K]9`KBLiK?KKEP•9PNNEKISU
ABaEIBaNP•E?9pfEP•gaEWN9U9:>
oBLiK9PNBLiXBN€9nnUKPXEg
sBYB`NP•Es„EnUK]_>CEFBNP•E?k>E
FC>KEP•gaEf>EN9P[g@KafBBC]E
hf@sEKE}€>:^BCN€NE;<FC>AE*N?
*GN9•XKBABFC>*EGKN>ELE}
€>BPP<EE}€>KB>WB:Ni>CLMK,C>?}E9~
9P[BSEK>9P[FEg@_>IN8_>N\EPL?_>?@9`sME;c
N
,r>@B>WBE}€>*EKWN>EN}BE>z]
C>n8>N|N>>NG(PErXaEK|EP>NFKN
#
9pŠEEK?M9P[P>NFPXBMEFC>B8>N|cN>>NG
78>N|cN>>NGhiSPNEBKD>8IsC]EB
NEN}ksWBaEHsE8?•sBBsPi9CRNEk
aE?WinUW8?NŠENEkC_>IN`;<FC>;KF
C>9P[P>NFn8>N|GfFC>gU>D‹@kŒC*9nk
89:97NKK@K9UnUfg@KgBN€TE>zFC>P
W?Bgj>W9UEI_>?@89:K?9P[E\Kgj>W•DE•?A
BFC>HI&KBABFC>HIgaEoWN>EKk?
knBaE9P[P>NFPXMEHIs89:97NKK@Kn
UAB9P[BaE`RsT_>Bg€PXEY9BP;E(FE
N€P<ENTEPaiNE_>Bg€PEnU9P[BC9URSE
BS9i>K?
k]KB_>N\ENEABFC>HI?9P[Dq9@PK
BKBˆDb‰sˆDQ>NO‰sKˆDENO‰
N}@EX9~nN8:>KaENE9:KLXHIPB
E}€>V~E2&sNEB…EiEC<2?<BCE}€>
VBNP•E9M\<js^Bs(W&Is>?}sBKaE9nV
?@>WN>EK89:E}€>:BHI@E+s@E^B(FEEa
EFK?KBEaEFP;E9]>W[gDR=
,C?sPnkaESK9P[KC>_>INElYgB
HI@EhC(:aEECaEVPX9P[BNU@€M
Ž*>r>kWs>r>BNUg@sHBst~kEK?KEHEsB
aE9P[DR=aE_>HI9CRs_>s_>BNUX]9k
nE7T9ns>r>E}€>KDL?YEBaESABFC>HI
@EhC9E9P[@€N\E
NE9:KYWK?sDNfK?BE}€>pE_>V:
DEKB€ESEVn:K
S9iV9:KKPXEXBNUBaESLM
HI@EhC<BE}€>>NE9:K>WH]EC
DLKI;r?(E>?{(E\&f9~E|N8:>NE_>B
NfPXE‹E}€>:9:KDI;P>+>89~E|
k]9`PXENE_>BNfE}€>9:K
%
I. Đặt vấn đề
(P|E9~@sr>@BaE9P[N9pC?ŠPXME
KC>HIBaE9nnUKBKBsBKC>gsBaE
g@sBKE}€>g\*ZBSE;<FC>KNEM9kE
Vp€KN8p@K9M:>[igP>NFKDR=sPEgaE
U_>}9P[NŠE…N8:>MEaEgB9P[P>NFPXMEHI
WiE?INEBaE9P[P>NEB;<FC>frXNE
]|E„E9P[p€PXMEHI,C?sBp€9~BSEaE
ECaEK_>I=C]EaEHE8?•siSBC]ERSE>
(>?}9noYYKB_>I=>zEFC>HIs>E8BaE
Sg€s…FC>‹YYŠPXMEHI|EPnBEI
>WLMsfg@KC>sBaESNi\aEŠS9i]Eg}s
BCN€sN_>?@9`NY@N}BE>zFC>gU>K?
hXY@9ns89:97NKK@K|EnUgB9P[FE
aEF>iTBE>zKC>HIn>EBE>zFC>K?I
9P[DR=P@K9UEP•ZEnUnFEaESY9kEBN[E|
NECBCN€KgBaEd•NKEs|EIU>N•I
8VFC>HIsU>N•B97NPEVBFC>MK?9UnUn9P[
FEP;EB>Wr@
hCgBaETBE>zFC>HINEBp€hC(
••IYKFEg@_>IE}€>:HIn>Es:FC>H
IKBgj>WDR=9~9P[BNUN}@EX>?}sFEHI@E
hCMnFE97NPEN}EVnnUW8?9P[E?YgBC:
7giBs|BKEFB@EhCNEBHIXBEaEFp
@N}@EXP@E+s@E^BhW?fFE97NPEK?IP<E@
K9@Bgj>WgBFC>HIsrInFEg=>WXK9U
nUWSE9P[FEP>@V@EhC„EPEI_>?@9P[FE€
MNE@EhC
6UNI•9P[FEL>mK?s9zBŽ9TFEPXE}€>:A
BFC>HIsfU>FE97NPEV@EhCsT9n9:NP;EPXE
E|EI_>?@KBLMHI@EhC€M<BE}€>;
Bg@_>IVE}€>NE9:KYWK?ŽKPEPX@9r>}
>WH]ECVX9:K
"
II. Cơ sở lý thuyết
1. Khái niệm Text Mining
!"#$%&
hCRSE;<FC>KM9kEVkp€9~9P[BNUNE
…E%HN<M9L?hXFC>9P[>WNE>]_>BNfM9kEV
kp€sk>r>9P[97NKfg@KgBN€TFEFC>
9n6niKD>8B9UVKB^BCN€T;<FC>(EP•
W8?NŠEnN8:>N€K|EgaEP•ENPX9E…:e
NEFC>sCSV|EKBCsgBBBN€9nsSS
FE>r>RSEaE;siSPNEBC>?}E?C‡
N[_>?@9`
ABFC>KE9MV?@>V_>BNf^BCN€T;<F
C>•>BNfgBN€9P[YC>B_>BNf>WK\F
C>snElKofB‹>N€NGn=ElN}WFC>n?\E
€gaEIKN}KkQ*PBP;EB]Eg}NPX9L?
hfW?gBFC>EzCK
7Z‹>fnU
f9P[T8gfkQ*KPEoFE‹>Z[XS9ifg@
X9P[E\KN€ŽnFEK]9U9BEBB}>i‹>P
!"##$%
6kXV‹>fS>kKg>EM_>?@>snU9]XC
]E79]XEP•ZEhiSXFC>VkaE?s_>BNfABF
C>fN9P[k>WP[€>9P[EIKZ><ZEi&•s9]
XC]EfN8XsNPXgP:nPE8€kBkWg@MK
„EWN9P[9:>K?_>BBBKi
iF>SEV‹>nU99P[_>Y}_>9@S9ifg@
hXkBkSNBINfB?i<aE?f>WN}gaEnEBN`s7Z
KX9]X
nU_>aE9MgBN€nN8:>‹>9P[8?NPEgaE
I‹>K„EnEBN`snUKXsF>iPEMrP•Es97CKg
BSEBgj>WYN}]Eg}*9n>aInB}>iKBK9B
B‹>9BEDDqsgaErP•E
nMsABFC>YNnUKk_>BNfDB9`‹>TB
*N>sRSEBgj>W‘nP\B?sWMEs]Eg}sL
MKBgj>W9P[BNU<EKE}€>Ni>CLMPME;c
!
NLM>NNgGsB>WBN>?:ENENGs_>?M>W
N>N>G
nUDq9@k]KBi9]XE}€>:ABFC>
c bGfkBDMLMGTk‹>F
C>KkNEBXNPX
c NENGfkBDMz_>?Tk‹>FC>K
k@Y9BnEBN`Y
c !">NEGKCaI>E9UfNBWDB9`
F>MBn?BM9UaIFC>
c #$%>NOGKC9fg@kaI>En•
kWFC>
!"#'()%&
ABFC>HI?BCN€TB;<FC>HID>
G9:W9@@NfNi\B‹>faENG?N€
gEG9BE_>L7nEBN`cNGTBKC>HI8>
N|•>BNfK?nU9P[KC<NkEgj>WABFC>N>?:
]EsfP|E9~8?9~9P[9:W<N}Ggj>WABFC>N>?:
]E*EGPXEXCBCN€TB;<FC>n8>N|
aE9P[P>NFPXMEE>?};8iKHIWin
U8?NŠEFC>zMPXMEHI…ng]P[EX;N8:>
XBFC>n8>N|gBY@sFEE}€>Er9L?9~8?NŠEn
9@!’aEVkp€ŠPXMEHI6nnUKBaEH
E8?•sBU>‹>9:>NsB@>97KEsB?}>r>g@>MsEI_>?@
_>?:[sBPi9CRGsBaEN}BP;EMA
BE}€>:;<FC>N9•KFEH%sEP•P<ENŠEnUP>
\MaEPXMEFC>n8>N|(PEN}Y@>Er#HB
NUsEP•‹ZEBC]EP>NF<MEHIKWNi…nD>PXE
ZEP•ED>?};T9nEP•nUNŠEBIeABFC>
HInUnEBN`P;EM;N8:>rXBIeAB
FC>N>?:]EgB>?}„EnU8?E?NŠEBgj>WAB
FC>HI€M;:>XBgj>WABFC>N>?:]E<
fIYCN}FC>HI]9~<ME8>N|Kni•b>OO?G
kiSKBgBFC>HIsgLiBKBE}
€>g\snBaE>
c ˆNKkC}_>9@9>9r>‰
c ˆND>8CnU@>E}NEB>‰
c ˆDnUEHIk]€E9>9r>‰
c ˆE}KkE>?}]9:>KDY}NEB>‰
Q>gLiBaE_>N\EK?sC]ErI9PNB>?
>LSUEiBME
c ˆ@>SE}nUEL?Nk]C9>9r>‰
d•NKE<9L?nYLi>?>W<€9k6U9M9P[gKHEP
W?rInFEaENfE}€>:Ni>CLM}@;
&KBABFC>HIKkKBE}€>9lYsEz
N8 :> gj >W „E P B PXE E} €> gB > > W aE
bNNNGsLiHID?Gs@D>8aEbN
DNGsW9M>NEGsLMHIENOGsU`NY_>
>OGsaEEC;<FC>s\B?NEGKILBgj
>WABFC>
NE9:KK?V?@>9:W9@KBSUs9nKKB
D ENOG K K B bN
NNGBE}€>XoTEM<PXfU>sgIBsBK:9:
BE}€>SU>K?KS9iNPX•KSS>WH]
EC
hXkC]EABHIP•EEzPXi
c &PX:DR=‚PXK?sC]EŽ>?UHITME8>
N|:MEn8>N|hiSsXHI&#'(' )"*C]E
Ž]E•ELiK&#'('+ ++)",BT9P[P>N}ENŽ
kBn8>N|9UCCDR=
c MmBaEgaEr@‚PXK?skLifB
MmBaEaiTHI&PXK?S>kN8:>K
EaEF9E9P[LiKgj>WŽ9P[ZE9ULim
PX@hiSs@>gj>WLiHIoYKDBD>8
D>8CTgBsg9nnUMmBTSP%s-s%
.s. “
c ABFC>9~9P[EIP[XBgj>WgBFC>
EGN>?:]E
nN8:>gj>WKP;EB]9P[RSEDE9UfN
Bg@N|XsB‹>XsKB}g@XBPX:DR=KBgj>W
N8€MŠLikLX97CKB>ki97Cs>9n
@KBSEBP;EBgBFC>g9U€KLi]Eg}
KLiB}g@BPX…MŽgBIHI9r?9VTWBH
IsiSPLXHI
S}>>]ZEVDEP•EK9P•E]C>_>IsKCsK
97NPE9UNf{Kfg@BW[NkEXVBHI*9nsBgj
>WiVDEnU9P[LLNKBCSK|EY
CgDR=gBHIMaEK|EnUNiNKMLi
9P[YC<|E
BMaE9P[NiNnUK
c &'()IRs9P[}g@X‡HIKWB~B
BgBN€9P[YCN}B~V‡HI(n
>EsnUEIRNŠEB~P;E€EXBTgBs‡kT
gBn_>CXkV9:SUK9n
c &(‚9L?EIRNŠEkHI9P[EB~XTETD>8
CNEHI9n
c &)!(‚9L?X‡HIf8?B>‡Ts>‡T9n
>k:klYK9nK9nCfgBHI9P[
YCN}BgC9P[EB~‡HI”>9U
VP;EBK?KB>WEF9P[BNiKnD>PXEW
N>EKBaE_>N\EVHI;P;EBNPX
9L?
BMg@[
c *!"+!(k]>WBNPX9L?EIRNŠEFC>
E>?}‹>9P[MW|L9UN[E|Bgj>WDR=Ea
EFY}B8>N|n|‹N}Y@nU9P[RSE
Pk;<CDR=gBN€
c &,-)!(‚9L?‡HI9P[9iXB>WEF
8?NTkL8B>WEFQ>9nskC]EŽLiY
L]k>EVB>WEFW>>CVTE>WEFC_>
9@BW>>CgBBL]}g@KBq9gBŠ
gBB_>CXEF|EM}g@K?nU„E9P[
RSE9U\KpE[V9:VB€
c *./01234(AaEE]EPM}g@aEP•E
YCBZ_>BEN}B|‹VHIsgj>WK?R
SE[@VkSEE>?}‹>VBHIAj>WK?9P[
E\KˆNiHIE>?}‹>‰
2. Bài toán phân loại văn bản (Text categorization)
*
^LMHIDENOGKDR=nBKC>KBX
gB>?BLnENG6L?KkBSLX}_>9@C
N_>?@9`DR=hX‡DR=Lnsg9PNkKC>sk_>?@9`
9P[9PNnn>kkXK?gaE(@>n>kkLXK9nf
IoNLXKn>kKhiSs9PNkV9::UsrI9P
N_>?@9`NŠEV9:9n>kBLX'/0*12*'1*34?
8€kaUKgBBC]ELMHIP•EKCX
k56'NEENG>WB9n9P[>E8kW‹>9U
SSCM?\W‹>K?EzkWBYUnEB~9P[
LXNPXnMEDs?G<9nDKYU9P[LXs?K~?LXG
9P[EBnhX;8>;<PW?sgkYU9P[>E8C
]EsnŽ]E•E>?NkKB\TW9KM‹>KBDMYUX
9nKkLX^LXHIKKB?K9EnFEPXBNU
@€_>N\EKE>?}LV?@>YBNUMŽEr9L?VB
aEE>?}NY>?@
+
, /0123415
kP;EB]Eg}LXHI9P[9:>gU<kT9UV
9:nL89P[9:D>8^P;EBK?RSEkT9UXk8>N|9;
EIT9UK?nUM?9P[{KEN}kW[KC>9P[LXŠE
?KnU`9P[Y9kEE:>EaEFgB>
|EDDqCSLMHI<V9:VKC>iSsk]
KC>:FE9kEWsKk]gBn:89:aEEC|EEIR
NŠEBV9:KXPE]9`)I>WV|EgaE>9P[
FEV9:TLVKC>PE?K9nsn}CKC>XkNE
FEV9:9P[Cg}NET9UC]EA@_>IKq9:rNHGY
P;E€EVKC>X‡TEV9:n‘
nk8:D>8CK9k]P>s?9k[=s9k@LMP
W?hiSsgLM€N}NXkEP•9\ˆfP•E‰sFE
LMPBK9kEW7EKaEECfgBZ[sNEgL
XBV9::9kEW\E]EPk>]T9UPW?Ž9PNkL>
NI•>E>ENŠE8IBV9:9n9:>n:9kEW,?nBgBsX
kEP•9\€N}NfP•EsWgaEi[ZE9ULM
FEKC>XFEV9:@;PFE9kEWn|s9kEWn
DP;E]Es9kEWLC
NEKE}€>K?s|EŽK>W:8>N|VT9UV9:sB
\YKBRSEBN\E]VB|N}E•NEL8sKk]gi
Y@:C}M9UV9:
,,,7'8'9:;
NEKE}€>VBBEI)>OBK+NKH "K !s
aE9:D>8CRSEkT9UnL89UDB9`FE9:KiV
kKC>–—h:7gj>WsT9UEzrBnTgn9MC
BV9:N}ECsKkU>{L8VBV9:K?
knTgnKkBBT7BU>€}_>9@f
NME@><}VV9:hiSsV9:aEBCg}BTPK•s
r?>s@sgBsr>E>?CsEP•KP;Es“|=NŠEFETK?
gaE9P[}g@X9r>SaEB?}g@X><8gƒ_>CEF
El}>>eKPgU>srs“
L?V9:9P[p€KkL8s?n>EKp€Kk
MEg9nk]V9:nU>kkK|VL?L8G
)I>Wfg@V9:N}T9U„EEznrfg@V9:9;
V9:BGKYN>?:N\E]VV9:N}L?Y@snNI•sL>m
>X€9kKfKC>K?ŽZXV9:9~˜kL>mPW?9P[
J
NI•‡V9:N}ECNENP•E[9;EI8sN\E]VkV
9:K]r>8GBTP;E€EsNEBTs9P[f8?NEKC>–—
^r€VEI>WnNBCN>?:Br>8f8?N}L?
–—hXrEI>WK?s|EnUoNNŠEkKC>9:W9@V:
FE9kEWn|sFE9kEWL:sFE9kEWEBB<|BsZ
[XV9::FE9kEWsBW]EKY}gaE<|B
,,,<5=2>5'90''?6>@
?fBBT9;EIsk]N\E]nU9P[RSE<EI
>W9U9`ElGq99`P[EYZ[VBTXBV9:KJG
9€_>N\EVB|V>kL?L8–—
MN\E]9r>}s|EE\KBN\E]YZ[sn}CXB
}g@EFBTKBV9:KB}g@EFB|N}L?hiSs@>KC>
9:W9@Tˆ4'%A0$BCfn9En:aaKZ[BTD4'%
A0$BC7DCFEV9::aas9kMNEB_>CK?P
@K˜h:NYEBs9nEEnVTD4'%A0$BCKV9:aaX;Y9nE
EnVTDC™PW?s]}g@EFˆCKV9:aa9P[EB
kN\E]m;
nU8?NŠEsN\E]
g
Vk}g@PW?EFkTgKkV
9:šs?EFkV9:gKV9:šVnN}L?GnU9P[9`ElP
9k Z [ N>E f V 9:V FE K C> 9P[ 9P N€ T K?
∑
∑
∈
∈
=
$
$
E
E
$
?
F
–—‚9L?qipE9P[YC8IBKC>n‘*s
E
?
K
q9YZ[VKC>XV9:šsK
$
K]rD>8CVT?V9:
gNEKC>
AaE?s|EgaEKM8€EI>W9BEW?K9UfN
q99kZ[
E
?
VBKC>BlYkB9kW?K9ns
kq9PW?9P[9BEBŠE?<>?}EsK>9nC]E9P[
>8>?CN}[BKC>B>?}EnUIP•ED>?}EBFE
N\E]i[ŠE?BKC>
IB@WK??}>r>NŠE9P[KHE?6UNB9:>9nsXk
qBEr9|EsXFE9:K9V›snUEI@NŠEFEHIN}:
V9:K?ErPgaEE•D>8CNEFEHIaEP•EA9n
U>€VBN\E]nU9P[9;EIn
∑
∈
=
$
E
$
F
–—
œ}>r>iM€VBN\E]cYLCBN\E]cKgI
HELCEF|EkV9:rIP;E€EXkW9BEgUG
FEKC>7gBsFEV9:KP;E€EXErP8IBKC>NE
;<FC>f|EKaif|EgaEq9PN8gƒg@>WZ[
KXBKC>P;E€E
$
(PW?sN\E]
E
F
Vk|šN}L?nU9P[9BEBP9k@9p
V9kZ[F
E
V9:_>FEKC>NE;<FC>kB9;EI9U
iBkgIHELCK9nkBN•NM
∑
∈
−=
E
E
G?F
J
G
s<9n
∑
∈
=
E
?G
KEBN`N>EfV
E
?
_>;<FC>C•*sK
E
?
9P[DB
9`<EI>WK?KgaEi9@EBN`F
E
NEk?}>r>iDB;s=
>?@aEnU9P[BSEqiBN\E]‚9L?|EgaE
K>W:=P<EK?
hXB@WK?sXk;<FC>WsN\E]VBV9:P
B9kEWsBW]Es}}Ž8f8IBKC>9:WŠE>
:BV9:K?7gBsnYNkNEB•BN\E]V|EŽ
s:>KC>NE9ngaEP;E€EXFEV9:K?sPE‹En
r9:W9BEgU9@BV9:K?
678"719#$(:;((&
^P;EBLXHI+78"719Nc*G9P[
9PNKH %–J—N}L?EzB|NE9P[EB~<B>WEFs
BBL?€|9P[EB~ŠEBN\E]V>WEFP;E€E9]X
KC>‹>sKBBL?9P[E•~<BLXkC]ELXP
W?ŽLMkKC>
E
<qR9C_>?BN\E]KB>WEF9P[
EB~B|NEXc;
→
E
9@gXXk|BA9ns~
V|K?9P[EB
E
6]BPXEBLMPW?RSEU>{
|f |f
aENM aENM EM EM
P;EEC
P;EEC D>8ge>
D>8ge>
aEEC
8 8
aEEC
|f|f
|f
|f
|f
|f |f
|f
,fkiS:L?_>?@9`
0
HI<ME`LsKPW?BL?„E9P[U>{PXME`L
kiS:L?_>?@9`9P[MNE,f
kP;EBgIZE9U>8>?CkL?_>?@9`LM'
Š
<@P[D'0?HC–J—@P[K?ŽgUNDC>8IBgBC
>8>?CnZE~Xn7
'
7
→
'
G™@>gaEsY\kgBC
$
sLL?KBXKC>nZEEBN`
$
K†K‡XPW?k
L?N}EC•>BNf9C_>?7MN}BL?9@g‡BVL?
B€BgBC>8>?}EBZEMNZ'
sg9nn9P[\
PK~VB9n&PX_>?@9`KC\>WEF
$
<9nŽDI?NB
'0skP;EBY\K\[iaE?N>?}s
kL?•_>BX}•nU`Ws@>PBBL?_>B97CXFC>>8
>?C
6]BPXEBM?L?_>?@9`PW?EzkP;EB
}L?KkP;EBDqXL?9UMmFEB_>B97C–J—
3. Bài toán thu thập thông tin (Information retrieval - IR)
*"<=
>WaE2bNdNGKkNEFEKBgBF
C>HI&KBK?V?@>WN>EKCfNBKC>NEkW[
BKC>n‘k9:>gCK9nB9:>gCK?nUKkN>?8
?kHI
A9:>gC9PKKkN>?8sKBŽ9PNB>?>W9UfN97
NPEVL>N>?89ns>9nBXB97NPEVBKC>n‘9Uf
NBKC>Z[8XL>N>?89nNEKBK?safVKB
ErXKBQNE>?}sKB>WaEKKB9P[
BNU<€9k;6]XKBQNEsL>N>?89PKKW
[BC(PEXKB>WaEsL>N>?89PKnUKk
L>HnEFEl,C]EŽfBLiEFElVL>N>?89Uf
N97NPEVn
#
A>WFC>s|EP•E]E•Efg@BFC>iDB
NEBNP•E[gBs|EgUN9UDkaEnNEkC
?gaEA>WaEsg@_>IiDBP•E9P[_>LsPEaE
P•E|E>]fg@kBP;E9]iDBXkaE97C
9P[9PKQ>9n|EŽY\aEZ[8TBg@_>IVq
DR=NPX9n(@>|EBnXBgU>C]EgB>s|EŽ
8?NŠENEk>EBN>?8;<FC>skqfg@Y8K9U
KI~kN>?8sKL>m9UfNL>NI•9P[@9@XgBC
NiD>8aEG97CKXkL>m97CNE>WaEsk
qfg@ŠfNkKC>KEP•ZE9ErBC]E>W
aE2d?G9P[RSE9U>WBKC>}_>9@B?}>r>N•
NKEh89:X>WaEKCDR=BHInk>E}_>kM
9@BHI9P[RSENPX9n,fJ9PNkafP;EB>W
aE>e,U}sC>WaEK_>BNfDR=7MsXDR=
9r>KK9r>NEz…E7iBM?}>r>
BK?>?U9pN>?8k@P[n‘ŠHEiZ
[VKC>9~W9P[
hC>WaEnU9P[9`El8€kMaEKi
SPgU>HIsfIsL>?}s<9L?|Eo9:W9@C
5!2
53-
678!
9):1
;!:1
]
˜
<!=
;>!
&!=
?.
,fJaf>WaE>e
%
>WHI<HIKkMaEKP;E€YCKgj>W
DR=9;EI;nU8MNŠEBgj>WK?„EnU9P[BSE
>WaE9P;EC
Bgj>W>WaEnU9P[NKM
c Bgj>W>e
c Bgj>WnBSENi>CLM
(n9r>}EzBgj>WYN}BP;E€>WBKB
\N>?:]E(n€]E•E>WN€ŠEBgj>WBSENi
>CLM9UEK9P[Bg@_>I];
+"<=
(EK??sBaE9E9P[BNUMŽ:]P[EKV?@>KT
2N2N9~N<K;P>NFs_>I=K97CK;>WaE
nEKC[[iN>ELKBaE>W9P[Z[X
>r>EP•ZE6nK=VBE}€>>?}L>NEBlYP
gBFC>*EGsNiD>8aE2bNDNGs>W
aE2bNdNG
d8:>BP;EB>WaE9P[BNUKg@_>IK|E
9MgB]NE9nnN8:>P;EBzM<ME>eBP;E
BK?P•EYBP;EBB\p9Uk]P;EBgB
9P[BNUPXEYNi>CLMQ>9L?s|EŽfU>L>;:
BP;EB>WaE
,+">
^rXBgj>W>e9P[BNUTFEH %9@FEH
"sKrXNE]|EYN}B>WBKaE€B\N>?:
]ENEKE}€>K?o9:W9@Baf"-I0JI0
"IKs"-$0I'L3JI'?>0'I"IK
,,,G-I0
&KafE}€>@P[Es9;EI8sK9P[UC9U9P
N=P<E;IB@P[ED;–0—,r>@9zE=NŠE8IB@
P[E}€>YN}CBEFL>N>?8KBKC>9P[P>NFa
f&E}€>@P[>WBKC>9P[EBEBN`D?IC€EX
N>?89n)IRKC>
š
9P[U>{KWB>WEF
{ }
$E
sss
J
=
s<
9n
Kk>WEFD>8CNEKC>
E
kN>?89P[U>{ŠEk
U>€EVB>WEFEzBBR+(*s.dsK(.
hiSXN>?8
MNJ,OPJPQ&6KKQR?
‚9L?qfg@&ŽW9P[8IBKC>n}g@XA
PEgaE}g@XAJ7BKC>n}g@XA$
SU;sXkL>N>?8
"
MNJ'()%OPJJ@A;*B(;(KOPJPQ&+(;CKK
,C]EŽ]E•EfN8IBKC>>kV9:D&ISGCsKSU
;KBP;EB>WaE€gaEIKBP;EBLX
HI
,,,,T'">>
}g@EFN>?8KKC>nU9P[U>ElkKBB
KK?P•EN89;EIkN@P[9P[RSEE\K'%.2':3
U>>,
@P[K?9P[RSENEk@9pVqfg@&s<9n
onBBRE+(*•P<EiV@P[K?9P[9PNgDDq
]P[EVB>WEF>ENEL>N>?8KNEKC>Q]K?9P[E\K
€9zE•D@KnU9P[RSEPkKB
hiSsBTgBAsAJsA$9P[}g@XBKC>*s*Js*$s*0
B>
A}g@X*s*Js*$s*0
AJ}g@;*s*J
A$}g@X*Js*$
K•žA+(*AJ+(*A$
hXN>?8•s|EŽnB€9zE•D@P>
$*J
J*s*$
*0
,,,,&-"$%"15
Aj>Lfg@>rYK;<Vaf&>?}EK??nN8
P•ED>?}9P[RSE7ZngBW(PEX8€BKsn
8?BKBKB9P[RSE–0—
6PNkWBKC>KkN>?8
{ }
P
sss
J
KkN>?8•s
|E9i(EBN`VKBGJM*
K6UW9P[BKC>}_>s
|Er•D@BKC>EIrVKBKm98IBKC>€E
XKBm;k.V')NPX(EPŸEK?nU9P[9`El
PkEBN`KBG7KkEiN`BXkHIK9nB
€X8Vgj>WK?Kf9P[B\EBN`EPŸE•Z[
6UYCaffg@&s|EnURSEk]gj>W
C>_>I>?}sB>WB9ngaE9P[9:WNEKE}€>K?
,,,W,&5'
‡kKC>r9P[9BoSDG<k]>WEFs‡>W
EFK?}>Ik>EVK}>B>WEFK?P•E9P[E\KB>W
EF9~E•oS?BTgB6UC>W9P[YCnEs
|E}•D@BTK?BTgB9P[P>NFNE'X"Y'sKX
!
‡TgB>kkTYEŽnBBKC>€TgBK?6UI~
kN>?8s|EŽYCfg@N}boSK?
Aj>WK?9P[RSE<:>C]EP;EMXB9k]P>gB
>VCoSfg@iS&cNG
BP[9UVgj>WK?K
c P>NF_>B:>nUrgaEEP>NF}9@$’Xgi
PX9r>G
c )BKWWKp€MoS
c )BK[BBKC>@>|E_>BK
>?}s|E„EnBP>9UN}E
c YC{KE
c ]9k
c *{KE‡N[BT9zEEl
,,,G-$0I'L3JZI'?>0'I"ILZ<GK
afgaEEc;9P[<NkETaf&NECU
CB>WEFVKC>–0—)]EPaf&s|EEB~B
KC><WB>WEF(PEN}Y@s9UgB>9P[eNEC
U>{K}>KC>*9P[U>{<kI'L3"L';XBaE]
€EX‡:>KN\E]€EXTE>WEFSUNENP•E[K?s
KpEa>WEF9P[9El9UDB9`k>EVKC>N\E]9P[
i<DB>8D>8CK9k_>N\EVTgB
NJF
*F
*,,,*F
P
K
hiSsgLiKC>
K
KKE}€>s}_>9@
C9L>9r>snc;9P[MN}9z`Jc:>P>
BN\E]N}‡c;U>{DB>8D>8CVB>WEFNE
‡KBKC>
s>WEF[0:1sG0\D>8CXDB>8rP[K
E}
6>9r>
@%A
@%A
J],^*],_^K
J],`*],K
,f$6z`U>{Bc;VKB
K
"#sJ#KC>
s>WEF[0:1sG0\D>8CXDB>8rP[KJs
%
NEafK?skN>?89P[9]DRPkKC>–0—Df0G
,?nBgBs|EŽU>L>N>?8<kc;N\E]VB>W
EFQ>gYCCLiL>N>?8Ž>9P[kc;hCY
CL>N>?8K?Y8KCBc;VL>N>?8XBc;9M
CBKC>k}>>eK9nA@_>IŽ>9P[kB
BKC>n_>Cˆ1CXL>N>?89~9PN8}sBKC>9nŽ
9P[•D@NfYEIrKŽ`•<k.VK9n
6Uic;U>{kKC>sBTN}ECNEKC>9P[p[M
N}Y@sCYC9P[YCB>
c BTS9P[B9
c ^LCBT<gIEN•E
6]X+EF7^BEFs‡T9P[BC<BgIEN•E
(PEEaEF@EhCMI?89:T9;KTEq6L?„EKk8
9:gngHgLBTNE@EhChiSsXT'"0 NE@E+s
€EXnKT' NE@EhC*89::TEq}E?:>U>r
NE@EhCB89:9nE\KYW•NE@EhChiSsXL>4'
:H0snUnBLB4':H0+K4'+:H0
(PW?s9]X@EhCs|ErnBP;EBBT97C;
,,,,&%'.3('&abca
N\E]Vk>WEFnU9P[DB9`:>BB@W
>EKRSEP;E€dbds<9nN\E]9P[pE[<?@>]
c tB>8>WEFNbN_>?cbGc97NPEDB>8D>8C
>WEFNEKC>
c (E`9IDB>8VKC>N>bN_>?cbGc97
NPEDB>8V>WEFNEKkW[BKC>,?n
@%A
@%A
E}
6>9r>
*
*J
_>N?
,f06z`U>{_>CEFN>?8_>N?GKB
KC>*s*J
J
BgBsk>WEF@gD>8CNEBKC>fdŽs
…@>nD>8CP•ED>?}NEBKC>fdŽ8
hiSaE€PX9L?9P[9:D>8nU9P[ZE9UiBEBN`9~n
<N}–0—
E
d
d
d
D
## +=
–0—
D>?
E=
d
<9nd
KDB>8D>8C>WEFS
NEKC>^L]NEd9P[i
BŠEP;EBEIiXgIHED>8CS
NEKC>K?
,,,,[4.3:J>"0? K
ABN\E]B>WEF9P[DB9`s|ErkK•D@9U
9`EB9kP;E9zEEFBc;N>?8KKC>k]q99kP;E
9zE9P[UCPX9L?‚9n•K*rP[KBW>WEFNEN>?8K
NEHI
M ∩
aE€9;EI8
M
M
+
∩
J
C]V*
M
M
∪
∩
C] N
JJ
M
M
+
∩
C]
Gs M
M ∩
C]MzE
k9BEB9kP;E9zEaEP•Es9P[@9@P9BEB–0—s
DB9`EnEFc;KC>Kc;N>?8<qiBPkg@_>I
kM67Cs9BEBK?P•E9P[iX9kKVc;6kP;E9zE
9P[DB9`aE€PX9L?–0—
∑ ∑
∑
= =
=
==
"
"
"
F
F
M
M
M>"
JJ
Gs
)IRIN>?8KKC>9P[>eB<9kKV|EsaE€ŽN<
}9;EI;
J
∑
=
==
"
FMM>"
Gs
Q>g8IBKC>9P[BXN>?8s|EŽ9P[•D@EI
r9kP;E9zEsg@_>IKkB9~9P[•D@VBKC>*
BK?nU9P[DR=ŠEBRSEBgj>WgB>
,,,W,&5'
afgaEEc;N8]aEgYCs9nNEY@k]
qD8Do9;EI9P[RSE,U}KU>CVBc;ozM
gBCEFNEY@sBc;@g9P[P>NF9r?9VKiP
V|EhiSsn8I$>WEFsKC>*o9:W9@#>WEFsPW?
gaEr@IP>NF8IBaE]€EXc;P;E€EXKC>K?
k"-$0I'L3:1 :9nU9P[RSE[C9UKEI9k
€MV>WB–0—•P<EVafKP>NFc;NEkC9~
9P[>?U9pCK?NI:kBBKC>XBTgB97C
ZEXaE:DB>8&}MCN>?D>8oSsC>?U9p
„EICB97i•EVCBBc;Aj>WK?Nk
qiB8W9P[XFEN>?8P;E9]ms…XFEN>?8
XsqiL]>eBŽYgf]gq(P[9U€Vgj>WK?
KriBBBL]>e>gnY?9pVd6:>9nN8nU
DI?NNEY@siSg}7DB9kKC>NEp[
6UPXP[EC>_>IVq>eBs|ERSEfP;E]
P[EB>WEFNEkKC>PL]>eBhXBNP•E[KC>
E•fqiD8DogaE9P[iDBs>?}gj>WK?„Enk]
P>9U>
c 'P<EVgiPXKC>N<}gaEn=ElX8€M
>eK
c 6k€MiBm;N8:>XBgj>WNPX9L?
c nUiBNPX
(PW?s9kP;E9zEnU9P[YC<aE€>
*NEg
Gs
∑
=
=
"
F
M>"
6+ D;"#E@F:( *( &
BP;EBNi>CLMP•EYNi>CLMWN>EKBEI
>W>8>?CB?\,?nN•;srInk_>BNf>8>?CB?
\LMHINPXgRSEn•>BNf>8>?CK?N8_>N\E
(@>B"e8 [=s$%>9P[Žn8P[EN8](PEEP[
Ms@>_>BNf>8>?CgaE[=fnU‹9@S9pKkC]E
BP;EBK?P•EI9]7Xk]89:>
JJ
c )I>W>?>W
c ^P;EBP>NFaE[=
c NBYS9p>k•EKM9kE
,r>@BEI>WYNi>CLMP•EE•I_>BNfY\NEg
RSEœ@>]K?_>?@9`:4"VC]E(PE>k•EK
M9aEsnUC]EŽLKfNMES9p?Bi>j_>B:>s_>B
NfY\`{>saEP>NF_>B:>8IB=N}9:>KEIM
9kEVC]E*9nsBP;EBK?rnYY9:>oNEM9kE
&}MEI>Wi>j?B„ErnEI>WSU?('Kf
Q>9L?|EŽE}€>SU;:BP;EBK?
,,gh"P3L?JPI?0IF?$K
nUnP;E9]MNŠEsBE}€>Er9L?:2dgBKaE
NEBgj>W9P[9:D>89UDiCk>EVKC>KN>?8s?nB
gBKYC9P[BLiEFElhXS}>K?sC]EnUB
SEBlYN€BDR=9Ufg@K>WaEKaE
K?n9P[El9M9P[gIHE\KgIHEpE_>BBVME(;c
N(>NNgG
hXCRSEME;cNs|EnUU>{krP[ENPEN
€NElYVKBsKnU9P[RSEKaENEC]E>
WaE
,,,,�"3L?
6UnUU>K@KME;cNnUBSEDR=>WaE
s|EŽ9`Elk]gBC9P[RSENE=>?}ME;cN
tL?YEBg]VafiBME;cNKB9;`E\K
j"JI?IKEN8:>B97iVN;cN\–0—s?n9|E
;KB|MEK?9P[amEB;cNV9kEW
‚B|ME<,f#UCBqBE+(*6r>NV|ME
Ž>@>B9r>K9:>>(n9P[YC<qBXEBN`EPŸE
J&K K\9r>N9:>n,U}KCYCqE.dŽnEBN`
EPŸEEI…#D,f#GBEBN`NEE79P[E\KBN\E]s
9`El9kMV}g@NEafiBVME;cNsN\E]
P•E9P[9`ElKEBN`ŠNEgIE–cs—
NENP•E[€M;siSgYCqB(.ds|Er
:>;k9;`sB9;`9nE\K9;`k
afiBME;cN9P[U>{<B>WEF:g@]Vn
B‹>g@]GKNEB>WEF:BK|E9P[9KMB>WRB
N\E]G
J$
,,,,G-? ;.2'0
af9P[9:D>8KkafX
c XB>WEFN>?8B|ME9r>KG¡•?N
c XBKC>B|ME9r>NG¡*?N
c XBoSB|eG¡?N
NEf"s|EnBg=C>>
c
coS>WEF
c
*
cKC>
c
•
c>WEFN>?8VEP•ZE
c
š
cN\E]g@]EFEHVME>WEFKkEHVME
KC>
c
_
cN\E]}g@EF>WEFVN>?8K>WEF
c
š
cEBN`}g@EF>WEF
K
š
c
š
cN\E]}g@EFKC>*
KKC>*
š
G
G
ž
G
G
ž
,f#ME;cNBR+(*GKBR.dG
B.C
B0C
A%D
@%D
@ @EF
@ @
,f%ME;cNXXeBR(.d
J0
X>WEFN>?8U>{B?}>r>EP•ZE‡k|Kk>W
EFNEN>?8XKC>U>{WBKC>‡|_>?@>9@kK
C>B|NEXK?nB}g@:>nN\E]sUCYP;E9zE
EFBKC>)BN`P;E9zEK?9P[iB<N\E]EFElVB
>WEFNE‡KC>XB>WEFKX9kE‡|U>{k>W
EF9P[9BoSB}g@nEBN`EFB|KBg@]EFB>W
EFNEN>?8
B}g@nN\E]EFBEH>kX>WEFno‹KBEH
>kX>WEF
E
U>{gIHE?=ElV>WEF
NEKC>
E
}g@nN\E]
9`El9k_>N\EV>WEF
NEKkp[B
KC>BEBN`g<9r>VBN\E]K?nU9P[N\EE‹>}7X
8€BDB9`K(@>EBN`E‹>}9P[RSEsnUŽI9]7
XB89:>
c •E\K
c An9M9P[YkS
6UfNN\E]F
E
}g@EF
K
š
Gs|EEIRNŠE9k}g@EF
>WEFHEg9zED>8CNEkKC>sKEBN`K?oEIgnk
rD>8CNEkKC>
}g@EFKC>9P[U>{<aE€>
∑ ∑
∑
−+
=
E $
$E$$E$
$
$E$
llll
ll
RJ*EK
™
G
,,,W,T(''90"
€HEMEEz
J#
,f"afU>{ME;cN
