nghiên cứu và đề xuất một bộ lọc web bằng phân loại văn bản cho một proxy server nhỏ
Bạn đang xem bản rút gọn của tài liệu. Xem và tải ngay bản đầy đủ của tài liệu tại đây (1.92 MB, 105 trang )
Mục lục
!"#$
%&'()*+,
-.%/#,
0.1#(%12#,
34%567(%12#,
38(9#:%#(7;7<!;,
30-%57<!;=
>(%?:@%A:%B
CDA%B!;.E7EF%G.HI#HE.AG
J4%K.L%%/.M5(3
J%N(F%"6;6O(?%7P.M1.%Q(-3
J0R%%R%.MI@%:HSBTUHV/.:6
W'&X7;@%:6Y7ZH[\.U].6(^J
$W'&X7;F%KA57ZH[\.U].:__`:.A^
,SD#(!;.E$
%&'(0aWb>+c,
0d%G/6HSBTUH,
0SBTUH\eUH`.UM^,
005_:A".%Q. 6S.HSBSD#(f,
00C.%#gQ.DO(.MA((%12#0=
00d%:6YDN/#0=
000d%:6Y7ZH[0=
0
0034%KA57ZH[00
00S._hF%&'(F%GFF%KA57ZH[0
00QFi%#j.MA(F%KA57ZH[0J
00J*#G.MR%F%KA57ZH[0J
00WA_G%%:7ZH[0J
00$k(D(HSF%KA57ZH[7EA7/BeUH3
%&'(3lk+mno3,
3S._h.QFi7<!;BTUH3,
3:%_G%!U7ED:%_G%.Mp(\:@_.7Ee%.U_.^3,
30%q.r@% :\@UgTAMDHA@(^3,
33/.%h(!G%(G\s:.(_g_.U6_^=
3BGg1#"#A6::6UWg_.U6\W^
3SB9#:+s3
3JB4
30tKgDP(([.%Q.$
30n;]#<.%A6S.F%&'(F%GFBeUH$
300%#i..AG,
303 6&XGH&-u:.%#i..AG=
30v%R%.%#i..AG0
33P:%BF%&'(F%GF(%12#0
33Gg%BE(Rf0
330%N(.%#iXu:G%.QFi.%UAD5(6Gg%B (G6_G.3
333nq!I6H1.MA(G%.QFi.%UAD5(6Gg%B (G6_G.3
33tKgDP(HSF%KA57ZH[\U].:__`UM^
%&'(twcxkyz{yn||
8%2DN/#
<#.M}DN/#.%UA.%#i..AG%#j
3
0<#.M}DN/#!;]#<.%AiF.MR%J=
3%#jH~DN/#J0
0v%R%.%•(%/6J
0%•(%/6.%UA2(D(\dI6.M:%A5.!S(u:.%#i..AG^J
00%•(%/6.M165(J,
3[.%#i.[.Q7E&#!€
3&#!€.r(H&-
30G&#!€%A.%#i..AG
E!q.$
E!q.4MA]g$
0v.[%.Q.GH&-.%#i..AG$
3•%&'(.MR%$
S._h[.Q.MA(%&'(.MR%$
%•(%/67E!G%(G$
v.M&(.%•(%/6‚<#%R%GD~%7$
04%&'(F%GF.%•(%/6ƒS._h.%•(%/6$
3nG%(G62!S%/#9#[,
%&'(d+„…†‡4|sˆ=0
dQ.#i=0
dQ.#i=0
0d%[Z(2(D(=0
35%Q=3
0&-(F%G..MI=3
0dQ(%~%&-(F%G..MI=3
00%[A#i=
…{+d‰=
DANH MỤC BẢNG:
[(3S._h_[F%j6BTUH.%UAF%&'(.%2+s
[(30dQ.9#[!G%(Gu:U.4MAŠU.‹Œ
[(dQ.9#[==".%•==.M:(%B(•#%1.%#S-F<6,
[(0dQ.9#[==".%•==.M:(%B(•#%1@%v(.%#S-F<6,0
[(3%%i@Q.9#[.%•(%/69#:65(,
[(%h(@1_h&X(.M:(@% :!}(.%UA.Ž/,
[(%h(@1_h&X(.M:(@% :!}(.%UA.Ž/,J
[(J%h(@1_P.%:g!8(&•(τ[%%&V(!Q%/#_#<.BTUH==
DANH MỤC HÌNH:
R%~.MLu:HSBSD#(.M1FMA]gu:65(3
R%0E%R%.MR%D#g/.<6.M#giF,
R%00%%B:6h9#:%/.MA(M}..ML%.%v(.0
R%03*#g.MR%F%KA57ZH[0J
R%0I#D•%N(!I6yy.MA(6S.6q.F%‘(%:%;#\[
9#g;%R%M’M::‹3Œ^0,
R%0G!&(.%‘(HI#D•@%A[(G%aƒ\’#DU:_.:U_^(N:
G!I67E7-!I6\[9#g;%R%M’M::‹3Œ^3
R%0JG7U.AM7E\[9#g;%R%M’M::‹3Œ^3
R%33v%R%.%#i..AGF%KA57ZH[0
R%*#:%/(N:GH[(.MA('_VDN/#.iF%#<#g/7E.iF.%•
\_:6FU^J
R%0#gQ.L%% :.E/#H:A(€6A5HY!G%D<#\:^7E.%“% :\H^
A@U”:.A_`AATUDHg_.AFTAMD_.G%B\^y_.U66(\D^(G.MB(_h\U^
\[9#g;M’DUM::‹3Œ^JJ
R%3E%R%2(D(.%•(%/6.M1(:AD/J
R%W'!€65(.%•(%/6E!q.FMA]g
R%&#!€H&-0
R%J&#!€H&-03
R%&#!€H&-3
R%$&#!€H&-_A_G%%/_h.M:(7E(&•(J
R%,&#!€E!q.2(D(
R%=v%R%65(.%•(%/6HSBFMA]gF%KA57ZH[$J
R%n;#%Ž%!~:%ŽFMA]g.M16GgU.$J
R%0I#!€HI#D•G(G.M~%/_h.&'(.Pu:_•_A7-_$$
R%3E%R%.%•(%/6.%#i..AG,=
R%v%R%@I6.%•4MA]geUH–.UM.M165(,3
J
R%E%R%@I6.M:%/#9#[.M165(,
R%JI#!€_A_G%_h&X(.%UA.Ž/.M:(H~@% :,J
R%I#!€HI#D•_h&X(.M:(9#:!&X(&•(τ,
R%$E%R%@I6.%•_P.%:g!8(&•(τ==
R%,n€.%~HI#D•%/#9#[@%.%:g!8(&•(τ=
DANH MỤC TỪ VIẾT TẮT:
W _g66U.M(.:W#H_MHUMU
’ A.U.::(U6U.D7UM_:M:’7MA6U._
‰W A_UW6:M.g
W A6::6UWUM7U
ed UFM:7UDeUHdUM
–4 –UM:_`UM4MA.AA
4 gFUM.U].M:_`UM4MA.AA
4 .UMU.4MA.AA\(%.%265(^
s `AM6:.AsU.MU7U
W4 .UMU.WUM7U4MA7DUM
W W6:M.g
W.AFTAMD D:%6G.r@%v([%%&V(!QSD#(7ZH[
4 M:_6__AA.MA4MA.AA
+s +`AM6sU_A#MUA:.AM
$
Lời nói đầu
t#<.F%G..r6S.%/.M5(_•D(.UMU.(EgE(F%8HQyGD~%7
.M#giF.UMU.F%G..MI65%6—O(7-g1#"#9#[C%q.%—7/_•
D(D%7TUHu:%;#(&7E.MG%_•D(%N(.M:(TUH˜!U™E6
HZ(%A5!5A!2]•%S%<.E!h7-(-.M“
-%N(g1#"#H2.%Q.! y(&(%12#.Q%E%.%P%/!;.E
Egy.M1'_Vv((%/F%KA57ZH[y!;.EEg6#h!5. 6S.([F%GF
B%N(.M:(TUH˜!U™%:g%N(.M:(TUH˜<6™6S.G%%/#9#[
#g%1y6S.7<6E!;.EEg%&:!5. EB%N(%R%[%.MA(6S.
.M:(TUHy!KgE7<!;"!&X!"#.&.%16%š6%AE.%/6.1#%L%!
Ev(7/BTUH
*#:!Kg.v]%K.%E%[6' Tiến Sĩ Nguyễn Văn Hiệp ‚[(71
M&(n5BG%d%A:F!•.i.R%%&-(D•y.M#g;!5.@Q.%2
!Q.vyG6'GAnh/ChịF%›(Sau đại học Trường Đại Học Công Nghệ Thông
Tin (thuộc Đại Học Quốc Gia TP. HCM)yG6'GH5Hœ!•(}F!•7E!S(
71.v%AE.%E%#i7ZEg
Người viết
Dương Đình Dũng
,
Chương 1: TỔNG QUAN
1.1. Giới thiệu:
n;.E!&X%BDP:.M1%/.M5(_•D(65(.UMU.%/:gy.5(:
!R%•(%&.5GD~%7v.M&(.UMU.F%G..MI65%6—%&(
•(.;6j%N(6h(#g%I6.MA(! y%&%N(%R%[%@%1#DK6yH5A
P7E%;#7<!;@%v(E%65%@%Gyž_— .G!S(]<#!Q(&DO(
.UMU.%<.E(-.M“ƒ%N((&%&:C.%2!"g!u7;%N((#g%5! R
.%Qy7<!;BTUH!&X%;#(&!"#.&7-%;#G%E6@%G%:#y%&(
O(%&-(!Q6S.6.1#E(Z%q%N(.M:(TUH!S%5
1.2. Mục tiêu nghiên cứu:
<!;(%12#%L%.MA(!;.EEgE(Z%qG.M:(TUH7E@Q
.%2_•D(!I]KgDP(6S.HS TUHEv((%/.M.%2y.%IEF%KA5
7ZH[\U].:__`:.A^n;.E"F%[!5.!&X%N(SD#(_:#
(%12#%N(!I665%u:@Ÿ.%#i.F%KA57ZH[7E@Q
.%2".%Q.%A@Ÿ.%#i.Eg%š6GFD( .h.%<.7EA!;.E
(%12#
tG!~%%N(!I6H<.iF.r%N(%&'(.MR%2(D(BTUH
%/ 7E%N(!I665%gQ#u:%N(F%&'(F%GF]KgDP(HS
BTUH
n;]#<.F%&'(F%GFBTUH7E]KgDP(6v%R%yE!q.6S.HSB
TUHDO(@Ÿ.%#i.F%KA57ZH[nq..%#i..AGBTUH!•E!q.
1FMA]g!I.%•(%/67E!G%(G.%#i..AG
%ŽM:%N([.Q".%Q.%A!;.E
1.3. Phạm vi nghiên cứu:
1.3.1. Tổng quan chung về vấn đề:
=
A.L%%<.u:65(.UMU.E.M#g;.[.%v(.MS(M•7E!:D5(y1
7/@%:.%G •((qF%;#MpMh<!;@%v(F%[EDO( @% @%Z
6E(-%57/_•D( %A.r(!h.&X(yDO(!}((#€.%v(.65(
7E%<.E%N(7<!; 19#:!Q!5A!2y.&.&V(y.%#"F%A(6Ÿ.
u:(&DO(7E7<!;:%u:6S.9#h(:
(:g.r@%.UMU.7ETUHF%G..MI7<!;H[A6i.•(%&A5HY%N(
.M:(TUH%2:SD#(@%v(E%65%@%v(%A(&DO(.M#giF!Q!•
!&X!q.M:*#:6S..%(:DEF%G..MIy.%UA_PF%G..MIu:v((%/7E.ML
.%2u:A(&6 .%@¡(p;7-6S.@Ÿ.%#i.(Z%q.M:(TUH
G%&'(.MR%BTUH.%&(DO(Gv((%/%&D:%_G%.Mp(7E
D:%_G%!U\H:@_.7ET%.U_.^y(Z.r@% :\@UgTAMD^y%q!~:%Ž1@Q.
\+s%:g4^y@I6_AG.SD#(\TUHA.U.^y77
nq!I6%#(GF%"6;6Eg .%IE.M1FMA]g_UM7UM%:gU.%Aq
[%:S._h%&'(.MR%%Ž%5g.M16Gg!'y(}FF%%#g%@I6_AG.7/
DO(65(.UMU.u:AU66R%.h.%'nh7-%N('9#:y.8%2y
DA:%(%/F.%R7/DO(%&'(.MR%BTUH@I6_AG.(:g!"#7EA!I9#[
C.AEHSG.M#giF.rH1.MA(%:gG%E%7!S.%iF.rH1(AEE7v
O(9#:.MB(-6S.&#&X(-%N(g1#"#!9#:E6%A7/]•CV
!"#@I6_AG..MV1HiMS%'7E#v!›%Y.h!S]•CF%[:A%'!I!GF
2(%:%% (Gg1#"#yQ#@%v(HSB7v.R%.5A1!I6.%p.8%:E
%i6%/.%h(65(
1.3.2. Giới hạn vấn đề:
MA(6v.M&(.M#g;.[.%v(.9#:65(.UMU.(&.:_•D(M<.
%;#2(D(!€%B:7E6#.6UD:7EA.MA(v(.M#g;.[.%v(.\F%"
-EG.M:(TUH^A! y%N(SD#(!&X(&@%:.%G.QF.%#M<.!:
D5(MA(!;.EEgy(&7Q.%B6S.F%"SD#(.MA(6S..M:(TUH!I
(%12#! E7ZH[H1.%I%/.MA(6S..M:(TUH
ZH[E%N(.r(ND•.[%A%N(SD#(!&XA(&DO(
.MA(.M#g;.%v(ZH[DP:.M1;.[(%L%E(v(N6ES(!€(A
(&_•D(!I.M:A!87-%:#/F%G..G.M:(TUH•(7igy(&.:
DO(%;#(v(N.M1.%Q(-!I.%I%/SD#(-!;.EEgy(&7Q.
%B(v(N.Q(%!I(%12#7R!KgE(v(N!&X_•D(%;#
.M1.%Q(-
MA(!;.EEg(&7Q.DO(@Ÿ.%#i.BTUHHš(v((%/@%:6Y7Z
H[\.U].6(^.%IEF%K-F7ZH[\.U].:__`:.A^-F%&'(F%GF
BTUHEg.%P%/@%G6F%GSD#(7EF%K.L%+s7-v((%/@%G6
F%G7ZH[%&(SD#(%u!5A2(D(.MA(#i7ZEgE@Ÿ.%#i._A
_G%%:7ZH[
1.4. Ý nghĩa khoa học:
t¢.7;6q.@%A:%By!;.EEgE6S.H8_#(C.&V(2(D(v((%/.M
.%27EA?%7P(%12#:.AE65(
t¢.7;6q.@Ÿ.%#i.yE6S.!;.E%&-(2(D(!Q6.1#.P!S(% :
%š6E6([6H _P:.%/Fu:A(&.MA(9#G.MR%7i%E%%/.%h(
BTUH
t¢.7;.L%]•%Sy9#:!;.EEg.v%g7B(( F6S.F%"7EAv(7/H[A
7/(G.M~!5A!2]•%Sy.%#"F%A(6Ÿ.7E.&.&V(
1.5. Lý do chọn đề tài và phát biểu bài toán:
(Eg.%G(Z60==y%L%F%u:%E%˜%v(.&1.~%7;9#[C
!5 C .UMU.™ _h =0£0==£ƒƒƒƒdn (N:Hh HS &#
%L%•.%v(¤Z% :.%v(.¤v(¤dQ%A5%7En"#.&p.!"#
%/#P7EA!"#.%G($ƒ0==
/#GD~%7%A.%#1.UMU.v(S( .%P%/(%16%Ž%.%v(
.&Egf/# (Z%q!&XG.M:(TUH˜!U™fE6.%QEA!I9#[CD~%
0
7.UMU. %/#9#[fn E%N(K#%Y!q.M:%A%N((&E6v(
(%/.%v(.
tKgDP(6S.HSBeUH%š6F%7%A7<!;:.AE.MA(7/.M#g
iF65(.UMU.E6S.g1#"# .%i.%;#%E_[]#<.F%"6;6!•.#(M:
.%~.M&(6S._h%&'(.MR%BTUHF%7%A6GgG%K%AqG–MUT:
u:GW47E•( %;#v((%/]KgDP(%&'(.MR%BTUH<.[!;#
%#(6S..%P.M5(EE6%i6!&(.M#g;DA_•D(GF%¢F@I6.M:7E
_A_G%1.y6S.gQ#!I6@%GE@%v(.P!S(iF%i.G%E%7_•D(
TUHu:(&DO(
%L%7R%N(CDA7E!;#@/.P%1! (&7Q.%B!;.E“Ứng
dụng phân loại văn bản xây dựng bộ lọc Web”!I]KgDP(6S.HSB'!S(
7E%/#9#[-g1#"#@%pF%%N(%5%Qu:%N(%&'(.MR%O(
A5%&_:#
h!SE67/%:%yL.E6(%—&#.%v(65(
/#9#[(Z%q!&X%N(.M:(TUH %5
n'([%5%Q_P:.%/Fu:A(&Hš(G%.P!S(iF
%i.
'!S(.MA(7i%E%E67/!&X7-%;#'_VDN/#%#<
#g/%AHSB@%G%:#.%UA.r(?%7P
<!;BTUHHš(F%&'(F%GFF%KA57ZH[Eg!•!&X% 6(%1
2#VCentre for Communication Security School of Information Technology and
Computer Science University of Wollongong, Australia% 6Eg(€6G.G([
sA(HA#ysU%:U%W:`:7ƒ:7EegW#_A!•!&:M:!;.E“Web Filtering
Using Text Classification” ‹,Œ 7E !; .E Eg!&X % .MX HV Smart Internet
Technology Cooperative Research Centre, Australia7EA.%G(Z60==3
3
P:7EAHEHGA!•.MR%HEgV.M1.v(%12# [.Q7E7Q..%E%#i
7Zu:6R%%N([.Q.MA(#i7ZEg.v%g7B(_—[.%/.h!SE6
7/y!S%L%]G7E%/#9#[.%P.%
¥4%G.HI#HE.AGMA(6S.65(6Gg.L%!&X@Q.h7-6v.M&(
.UMU.H1(AEy%AF%¢F(&DO(.M#giF!Q%N(.M:(TUH.5G6Gg
%u.M165(/@I6_AG.&#.%v(65(E%š6@I6_AG.9#G.MR%_•D(
65(u:(&DO(MA(HE.AGEg.vDO(F%&'(F%GFF%KA57ZH[
]KgDP(6S.HSBTUH!q..5(¦M:7EA65(%š6@I6_AG.7E(Z@%v(
%A.M#giF%N(.M:(TUH]<#§
¥W'!€HE.AG%&_:#
R%~.MLu:HSBSD#(.M1FMA]gu:65(
R%V.M1%A.%<g7~.MLu:HSB!h7-7/@I6_AG..M#giFu:
6S.FMA]g!:(%A5.!S(.M165(WPBTUH][gM:_:#@%6S..M#giF!&X
%AF%¢FS..%E%F%"BTUHDP:.M17ZH[]•CSD#(.MA(6S.!~%
D5(7ZH[7E7R.%QGF%&'(.%2B%&EF%&'(.%2_A.r\@UgTAMDƒ
6:.%(^%:gGF%&'(.%2.%7@%G19#:!QF%K.L%7ZH[ .%I
!&X_•D(
1.6. Phân tích hiện trạng
1.6.1. Những phần mềm cùng lĩnh vực trên thế giới:
/:gy?%7PEg!&X%;#%E_[]#<.F%"6;69#:.K6y7R%N(
g1#"#(Z%qG.M:(TUH]<#(EgE(%;#%E_[]#<.#v!86-
5(
(AE
%E%
F%"
4MA]g
(#g1
.%ug
(&
DO(
H1
.MA(
%E%F%"BSD#(
v((%/%š6!U6!Q%/#9#[:A%<..MA(7/(Z%q.M:(TUH]<#7E
(N:.AE%A65(
<!;BeUH!•!&X%;#%EF%G..MIF%"6;6.M1.%Q(-]KgDP(
.M1%N(([F%GFBTUHS._hF%"6;6!I%R%%&
W#M`A.MA‚’.UMFM_U%MU:.4MA.U.A!KgEF%"6;6u:%•(
W#M`A.MAyF%"6;6Eg.%Q.@Q.%UAG%.QFiBTUH7E(Z
%q.rFMA]g9#:+s7E.r@% :y @%A[(0=A5@I#(Z%q
.UMU.–.UMƒeUH–.UM_DA%•(ƒ4M_6.UMU.`.UM_¨TUH
`.UM_F%G..MIyEF%"6;6.%P%/(G6_G.7E(Z%q4%"
6;6Eg!&X9#[(GADO(@Ÿ.%#i.BTUH!S(@I6_AG.SD#(
.M:(TUHV(¦7EA#g%1y.%UA%&-(D•9#[.M~u:%E_[
]#<..%RF%"6;6Eg•( H (DG(u:@Ÿ.%#i.DO(F%&'(
F%GFB%q.r@% :
.UMU.WU#M.gWg_.U6_ƒ4MA7U.:eUH–.UM(Z%q%N(.M:(
TUH@%v(6A(6#h%;#%'G_[F%j6BSD#(@%GE
F%"6;6!&XE!q.VD5(U.£_UM7UMy%E_[]#<.#(<F_©
6S.D:%_G%%;#!~:%Ž47E+s19#:!Q%N(.M:(TUH
]<#(AEM: •(%A(&9#[.M~!&XF%¢FH8_#(.%16%N(
!~:%Ž\4%:g+s^6-]#<.%/7EAV_VDN/#%/ u:
WWUM7UMeUH–.UM!KgED~%7.L%%XF.MA(_[F%j6W
\.UMU.WU#M.g:DUUM:.AM_^u:%•(MA_A`.4%"6;6W
%5gVD5(.&(•:7r:@I6_AG..M#giF65(y7r:E6ug%/6
\FMA]g^•((h(%&GF%"6;6.M1W@œ6D~%7BTUH
7-@Ÿ.%#i.F%K.L%1@Q.y(Z!~:%Ž\47E+s^y@I6_AG.S
D#(Hš(.r@% :Gg1#"#BTUH!&X(&9#[.M~.%Q.iF
.%v(9#:G#i.BTUH
1.6.2. Tình hình triển khai bộ lọc web ở Việt nam:
MA(%N(Z6("!KgD~%7.UMU.F%G..MI65%6—DA%L%_G%
#(<F.%v(.u:%E&-•(%&v((%/WM:!yO(7-_P
5%.M:%u:G%E#(<FD~%7.UMU.17<!;.QFi7-.UMU.
u:(&DKD•DE(%'rv(_Vy.M&(%ByGF%›(U.v(S(y9#G
:`¢.UMU.y!Q(:!R%(&DO(.M#giF%;#%'7E7/@I6_AG..M#giF
•(%&9#g!~%_•D(65(•(!&X!;iF!Q#g%1yV.r(<F!S
_•D(7•› !;#(R! %&:!&X%}.MB(p6nI%R%%<.E%N(
D~%7%:g9#G:`¢.UMU.y.5%N('Eg!:_h%ŽDO(HS!~%.#gQ
W!I@Q.h6E@%v(DO(6S.HSBEA(Z%qy —DA.h!S.M#giF
u:@%G%%E(H~[%%&V(\f^
4%"-GD~%7%A.%#1.UMU.V.%E%F%h€%L% M1(7E
[&- %#(%Q6@%_•D(GHSBTUHy%ugQ#DO(F%"6;6W
7-%2Z(4MA]g_UM7UME%L%
A%N(!q!I6%&.M1y?%7PF%G..MIHSBTUH•(@%v(F%G.
.MI65%6—yF%"6;6!I%R%%<..MA(?%7PEgu:/.:6Eed
\UFM:7UDeUHdUM.G([•&'(š(v(.gn/%A5nv(%E%
F%h.%P%/^4%"6;6Eg7r:!&XiF%i.F%1H[0-G%2Z(
6-%&dI6.M:SD#(u:.M:(eUH•.iF.HGAGA !~:%Ž6:
!&X%Ž!~%(Z%qG.M:(eUHEF%"6;6(G!/Fy9#:g_h9#h.Qy
9#[(GAP!S(iF%i.D:%_G%G.r@% :y.M:(TUH<6.r65(y.%v(
HGA@% F%1H[ed6-#g%1yed0%&:.%#gQ.F%!&X(&
DO(65(yDA.L%%/#9#[y.h!SE67/y•(%&%u!~%u:A(&§
%P%/%u.M&'(u:S&#%L%7•.%v(7;9#[C!5C.UMU.y
v(.gn/.AG7EM#g;_h/#!•F%h%XF7-GH&#!/.Ž%£.%E%
F%h7EGv(.g7•.%v(.MI@%:E!q.F%"6;69#[C.UMU.v(
S(%AG!5C.%#S65(TE6S.([F%GFF%"6;69#[
C!5C.UMU.%5g.M1;%/!;#%E%eDAT_S._h.L%Z(%L%u:
J
T(€6(Z%qG.M:(TUH SD#(]<#y.P!S(iF%i.D:%
_G%GTUH!U"F%[%qy.MX(}FG!5C.UMU..MA(7/(%.%v(.
@%G%%E(_•D(D~%7y.L%.;_•D(D~%7#g%1y_:#6S..%
(:(p.MI@%:.%RF%"6;6EgH~F%[2(u:GF%›(U.M<.65%6—y
'!•(•HY7-O(CDA%&%N(F%"6;6BTUH@I.M1
1.7. Sơ lược về khai mỏ văn bản (text mining):
d%:6Y7ZH[\U].(^E_P@%G6F%G%N(.%v(.6-%:g.M&-
! @%v(HQ.!QHš(6Gg.L%.%v(9#:7/.ML%]#<..%v(..P!S(.r
%N((#€.E/#@%G%:#S.gQ#.h.%U%h.E_P1@Q.7-%:#u:
.%v(.!&X.ML%]#<..%UAD5(_P@/6-%:g([.%#gQ.6-!IF%G.%/M:
%;#%'Hš(F%&'(.%2.%•(%/6.%v(.%&(
d%:6Y7ZH[ _P@%GH/.7-%N((R.:HQ..MA(HS.R6@Q6.M1
TUHMA(! (&DO(!'.%#".R6!&X%N((R!•HQ..M&-7E!&X!&:
M:HV%N((&@%G<!;@%:6Y7ZH[E_P.G%H5%M¦ME((N:
%N(.&/#%/ @%v(19#:%A_P".%Q.u:(&DO(7-6!L%
.R6M:%N(.%v(.19#:
MA(@%:6Y7ZH[y6.1#E@%G6F%G.%v(.%&:HQ..M&-!Kgy
%N(!;#6E(&.:%&:M¦}<g1•(@%v(.%I7Q.M:
d%:6Y7ZH[E6S._PHQ.%I.MA(?%7P@%:6YDN/#y6E@%:6Y
DN/#%š6.R6M:%N(6v%R%9#:.K6.r%N('_VDN/#-S.7LD
!I%R%.MA(@%:6YDN/#E7/DO(G% :!'6#:%E(u:@%G%%E(
!IDP!AG%N(_[F%j6EA!q.("%:#.M1@/%E(LDQ#@%G%%E(
6#:6S.!œ`:_%(%.y(&HG#v6#hHG.%1671F%A!œ`:_%(%.
7R.%Q%}(F%[!&X!q.("%:#S.2(D( 19#:E7/F%G.%/
_P(:i.MA(_•D(.%“.LD(G%EF%K.L%.R6@Q6.M16S.&X(
-.%“.LD(.%#.%iF!I.R6M:_P%1%/%.r%N(.%“.1#]EHR%.%&(
d%GH/.'H[(N:@%:6YDN/#.%v(.%&(7E@%:6Y7ZH[E
.MA(@%:6Y7ZH[%N(6•#!&X.ML%]#<..r7ZH[(v(N.P%1
.%:g%A7/DO(%N('_VDN/# <#.M}u:%N(_P@/'_VDN
/#!&X.%Q.@Q.r%N(%&'(.MR%!I]•C6S.G%.P!S(¤ZH[!&X
7Q.HVA(&!I!B
#g%1y 6S.?%7P!&X(BE]•C(v(NHš(6Gg.L%\›(B
E]•C(v(N.P%1^6E9#G.MR%Eg.5AM:%E(A5.]•C.MA(@%.%P
%/%N(.G7%Y.MA(F%K.L%7ZH[%‘(%5y(&.:7Q.M:6S.
%&'(.MR%.ML%]#<.%N(6.ry.r%N(HEHGA%:g9#gI_G%.&'(!h
D•DE(y@%%I.%~%A(&!BH[. 6&XSD#(7ZH[
%N(%&'(.MR%%A!S%L%]G%<F%i!&Xy!I%R%%&
%&'(.MR%M}..ML%.%v(.G%K.r7ZH[F%<#.M}.M[7;.%v(.G
<#.M}y7LD%&'(.MR%!B.iF.7ZH[6v.[.%v(.G%K.M[7;%B
.1y!~:%Žy@ŸZ((%;(%/Fu:(&! -!S%L%]G .%I1!Q
$=ª
S.7<!;.%Q..%P%2:%«%<.E7/2(D(@%:6Y7ZH[7EA?%
7P_%%BS.7LDD•%iM:%<.E.MA(v(.MR%u:AWT:_A!&:
M:([.%Q.(#g1%Ku:ZH/%%Q6 Hš(G%.R6@Q6.MA(G6h
1@Q.(G.QF.MA(G.iFAu:%N(.&/#7;_%%B
S.7LD@%Gy2(D(@%:6Y7ZH[!I([9#gQ.K#%Y-.MA(G
HS(UE%N(FMA.UEA.&'(.G7-%N(FMA.U@%G
%N((-%5'H[u:@%:6Y7ZH[E
d%v(.%I7Q.M:%N(%&'(.MR%D~%7ZH[.MB7«.MA(6S.
.%(:DE
%v(.(&.:".%Q..%&(@%v(!&X(%5.MA(%N(6•#
(#g1H[
1.8. Sơ lược về phân loại văn bản (text classification):
$
*#:%;#Z6&X(.E/#_h% :F%G..MI7&' 6S.@L%.%&-@%8(
€%&6S..<.gQ#y@%[Z(_pF]QF7EF%KA5.E/#6S.G%.P!S(.MV
.%E%7<!;9#:.MB(%E(!"#
%:7<!;@%G%:#.MA(F%KA57ZH[(A667ZH[\.U].
#_.UM(^7EF%KA57ZH[\.U].:.U(AM”:.A^<!;(A6619#:
!Q7/.R6@Q66S.% 6<#.M}.;6j.MA(.iFG.E/#MA(@%!
7/F%KA5› .1(B@%GEF%KA57ZH[\U].:__`:.A^ .%I
!&X]U6%&.G7u:7/<#.M}@%A%2:.E/#.%UA% 6<#.M}
4%KA5.E/#]#<.%/.MA(%;#2(D(%&BUƒ6:y!~%%&-(
6:yB.%&MG\_F:6^y(G6_G..y%Ž6.P!S(GHEHGA@%A:%Byž
4%KA57ZH[.P!S(M<.%<FD•7R7/.8%27ZH[.%uv( .%I%Q6
%F%L9#G!p.
QFi7&X..MS.MA(F%KA57ZH[!&XDP:.M1%N(@Ÿ.%#i.6Gg
%B: .%I%iM:H:(:!A5@%G%:#.MA(7/.%Q.@Q%/.%h(F%KA5
7ZH[HI#D•.E/#y]KgDP(HSF%KA5y&X((GHSF%KA5
1.9. Nội dung đề tài:
MA(#i7ZEg"!5.!Q6S..iF.E/#(%12#C.%#gQ.7E.%Q.
@Q%&'(.MR%
S.H[6%%B:\DU6A^E!q.%A.%#i..AG7E.%I(%/6.M165(
<#.M}DP@Qu:#i7Z%&_:#
%&'(8(9#:
%&'(0'_VC.%#gQ.
%&'(3(%12#7<!;
%&'(tKgDP(2(D(‚.%•(%/6‚!G%(G
%&'(dQ.#i7E%&-(F%G..MI
,
Chương 2: CƠ SỞ LÝ THUYẾT
2.1. Khái niệm bộ lọc web:
2.1.1. Bộ lọc web (Web filter):
SBTUHEF%"6;6 %2Z(B7EA5SD#(%I.%~.M16S.
.MR%D#g/.TUH%:g@% :7E7~.MLTUH\TUH_.U^6E(&DO(h(p(.M#giF
!QSB@I6.M:SD#(u:6S..M:(TUH%:g6S.!~:%ŽTUHDP:7EA6S.
.iFG#i.7E.%:g.%QH<.@¡SD#(@%v(6A(6#h7-6S..M:(TUH.%:g
.%Qy.%&(.M:(EgSD#( D›(˜U__UUD™\<6.M#g]#<.^
R%0E%R%.MR%D#g/.<6.M#giF
%E9#[.M~%/.%h(E%N((&.%&(p69#g;@I6_AG.7E<#%R%
A5SD#(!9#:HSBGHSBTUH.%&(!&X_•D(.MA(G.M&(
%By.%&7/y./6:`¢.UMU.yGD~%7.UMU.v(S(y(:g[.5%E
•( .%IGFD((N:.AE%A.M“A.MG%%N(SD#(@%v(E%65%
.r.UMU.
2.1.2. Tại sao cần thiết có một bộ lọc nội dung?
nKgEHSF%i9#:.MB(!I@I6_AG.SD#(.M165(7E9#g!~%G%
DO(.E(#g1.M165(%&.%QEA
0=
(&DO(.%&( @%#g%%&-(E67/M1(.&%:g%N(!;#F%
F%GF.MA(%E((E67/.MA(%:g(AE(E67/.5'9#:n;#Eg_—
.1#F%L%E((E67/ (G.M~7E .%I @%[Z(E6%A%/.%h(65(
M'7EA.M5(.%G.MR.M/DA65(!:(H~56D(!I.[]#h(%N(.E/#F%
F%GF
2.2. Lý thuyết dùng trong nghiên cứu
2.2.1. Khai mỏ dữ liệu:
E.%:A.GM}..ML%.%v(.%N#L%y%&:HQ..;6j.MA(6S.@%hDN/#
-%v(.M}..ML%!&X›(BE.M.%2.r.MA(@%hDN/#y .%I([
.%L%DN/#.M1.iFDN/#! .r! #(<F.%v(.% .MXM:9#gQ.!~%yDP
HGA%:g@%G9#G.DN/#d%:6YDN/#!&X_•D(MS(M•.MA(G(E%
F%K.L%.%~.M&(y9#[CF%K.L%MuMAy9#[C7EF%K.L%_:%Y(y@%:
.%GTUHy@%:.%G7ZH[yž
2.2.2. Khai mỏ văn bản:
2.2.2.1. Khái niệm
E.G7@%:.%G.%v(..r%;#.iF.7ZH[y(#€.E/#Eg .%I
.rHEHGAyHE7Q.(%12#y_G%y.%&7/!/.•y.%&!/.•y.M:(TUHyž
v(D(u:@%:.%G7ZH[F%7%A7/M}..ML%.%v(.y(A6% 6
7ZH[yF%KA57ZH[yž
R6G.%v(.%N#L%.M16S..iFG7ZH[d%:6Y7ZH[!I.R6
.%v(.\s^•((h(%&.M#g7<W
W7ZH[E.iF%XFG7ZH[.r%;#(#€@%G%:#%&HEHGAy
HE(%12#y_G%y.%&7/!/.•y.%&!/.•y.M:(TUHyžG(#€.&/#
Eg(EgE(%;#A(&@%v(.%I!B%:g.QF%i.<.[G.%v(.
.MA(!
s}..ML%.%v(.\s^%& .M1_h&X(7ZH[E(%;#%&(.%v(
.F%K.G.MA(%;#.E/#@%G%:#
0
E.AGM}..ML%.%v(.]G!~%.E/#EAEA %2:.%v(.
6E(&!B6A(6#h.%v(9#:6S..r@% :EA!
%&7igv(7/M}..ML%.%v(.(h(%&.M#g7<W
nS%L%]G
nS%L%]G\4MU_A^E.%E%F%"F%".MZ6u:G.E/#!&X
.R6M:.M1.%P.Q 19#:!Q.M#g7<\K#.M[%L%]G^
nS.R6!&X\sU:^.%E%F%"F%".MZ6u:%N(.E/# 1
9#:!QK#.M#g7<
t¢.6v%R%6%%B:_:#
R%00%%B:6h9#:%/.MA(M}..ML%.%v(.
4%"sUU7:.\19#:^E.iFG.%v(..R6!&X 19#:!Q
?%7P:"9#:.K6
4%"sU.MU7UD\.R6!&X^E.iFG.%v(..R6!&X
4%"(:A(N:%:F%".M1E(-E(.h.y%A.:HQ.62!S.R6
!&X7E.%v(.19#:%;#%:gL.4%"Eg#vV.•_hu:0
v(.%2.M1
¬-
¬-¬-
retrieved
retrievedrelevant
precision
∩
=
¬-
¬-¬-
relevant
retrievedrelevant
recall
∩
=
All
Relevant
Retrieved
Relevant &
Retrieved
00
2.2.2.2. Một số loại khai mỏ
Phân tích kết hợp dựa trên từ khóa:S..E/# .%I]U6%&6S.
%# @C.P7E .%I]G!~%Hš(.iFG.r@% :/F%K.L%G
.E/#DP:.M1.r@% :!I.R6M:6S.@Q.#i7;.E/#!
Phân tích tài liệu tự động:h(%&6S.(&.MXCy% .MX!pP
.MA(7/F%KA5.E/#Hš(G%˜!B™.<.[G(#€.E/#!Q
7E]QF .%UA.r(A56S.G%.P!S(
Đo độ tương đồng giữa các tài liệu: nA!S.&'(!€(E7/]U6]¢.
.E/#! ]U6 .%#S7;6S.D›(7Z%BEA%:g.%#S7;6S.
.G([EA! Aq•( .%IDO(!I]QFA57ZH[.%#S7;?%
7PEA
Phân tích trình tự:nAG_P@/yDPHGA]#%&-(%&!• H1.M1y
7ZH[E6S.%# G@C.PD•!5.6S.C%;#.E/#(V!Qy
%;#<F!SD•!5.7;6S.7<!;rG7<!;Eg%/.%h( .%I
!&:M:DP!AG7;GD•HQu:%/.&X(%:g%N(!;#_—][gM:
.QF.%UA
Xác định các hiện tượng không bình thường:/.&X(@%v(HR%
.%&(E6S.7ZH[!Q _P@%GH/.%:g˜G.L%™9#G@%G_A7-
O(A5 !Q.M&-! !I%A6S.@Q.#i7;_PH<..%&(u:A5.
7ZH[
2.2.3. Phân loại văn bản
2.2.3.1. Định nghĩa phân loại văn bản
4%KA57ZH[E.G7@%V(G6S.%:g%;#A57ZH[(v(N.P
%1!&X]G!~%.M&-7EAG% 6.E/#EA! !•!~%(%?:.M&-%‘(
%5%&y6S.%/.%h(U6:%i.%v(!/F!Q.%R.%v(!/F! .%I!&X(G
E˜Š#@™%:g˜AƒŠ#@™y —!I.%#iX%A7/M:9#gQ.!~%1%:g@%v(
03
1] :.%v(!/F! 6S.G%.P!S(S.7LD@%Gy]QFA5GHEHGA7EA
G.%IA5%&9#h.Qy.%I.%:Ay.%&'(65yž
ML%]#<..%v(.\s^E.G7.R6@Q6%N(!A519#:u:G.E
/#6E%N(.E/#! DP:.M17E.%v(.".%Q.%:g@%G/6!&X!~%
(%?:.M&-<g6S.7LDy6S.%/.%h(s .%I%iD/%N(!A57ZH[
.MR%HEg.MA(!;6HGAyv(.gy!~:!I6y7E%.Q.&'(.r%N(6•#
9#[(GA(%;(%/F
4%KA57ZH[7E.ML%]#<..%v(.E%:D5(u:]•C7ZH[v(
5y%&( _P.&'(.G%:g%XFP(N:%N(@Ÿ.%#i.%Q66S.F%"@%16
.h%;#%/.%h(sF%G..MI6S.7E@Ÿ.%#i.F%KA57ZH[!I![6H[A
Mš(%N(7ZH[! !&X]•C!I!&:!Q6S..iF7ZH[%2:DN/#6A(
6#h
(%?.%UA6S.G%@%Gy.: .%I6&(.&X(Mš(6S.%/.%h(F%KA5
7ZH[6E%/.%h(! (G%N(A5DP:.M1%N(F%K!A5!&X.ML%]#<.
_#h..MA(H&-@%V(Gs4%G.HI#.%UA.AG%ByF%KA57ZH[E.G7
]<F]Ž%E6!L%\.:M(U.`#.A^%&:HQ.M¦Ψ]Φ-y–¬y.MA(! C(%?:
u:%E6Ψ]Φ-y–¬!&X(BEHSF%K-Fy6EΨ7EΦ˜E(.MO(@%-F
E(.h.™MA(!
®-
y
0
yž
6
¬E6S..iF(€6GA5!&X!~%(%?:.M&-\ m.%IA5
"F%KH/.^
E6S.?%7Pu:%N(.E/#
2.2.3.2. Đặc điểm của phương pháp gán nhãn:
‚:([_•Mš(%N(A5%ŽEG%•@C%/#yC(%?:u:%•@C%/#
%ŽEG@C%/#@%v(6:((N(%?:([.%L%!I(}F]KgDP(%N(HS
F%K-Fy. 657ZH[š6.MA(%•@%v( (%?:
‚WP9#g@Q.G7ZH[7EA.MA(GA5F%[!&X.%P%/DP:.M1'_V
SD#(7ZH[%2@%v(DP:.M1.%v(.7;DN/#6E.%v(.Eg .%I
_©.r6S.(#€DN/#H1(AE
0
‚%AMš(ySD#(u:7ZH[E6S.@%G/66:(.L%%u9#:y!;#
Eg 1Mš(.%E%71u:6S.7ZH[.MA(6S..%IA5@%v(.%I!&X
9#gQ.!~%%p%p
‚4%.%#S7EA2(D(y_PF%K-F .%I
%•!' 7ZH[ .%I!&X(G%L%]G7EA6S.A5y.MA(
.M&(%XF_hA56®0\.iF^.: %•%~F%K
n:%• .E/# .%I!&X(G7EA6S._hA5%•
‚c1#"#!q.M:%AHSF%K-F7ZH[
WPF%K-F2(\:MD:__`:.A^nh7-D5(Eg% .MX6S.(G
.M~.MA(.iF-y–¬% %ŽM:%AHQ.7ZH[E.%E%71%:g
@%v(F%[.%E%71u:D
Š
7E
4%&'(F%GFEg%N#D(%AHS
F%K-F!SiF
WPF%K-F6;6\WA`.:__`:.A^% .MX(G.M~.MA(‹=yŒ %A
HQ.<F!S.igu:%/.%h(7EA.%E%71u:D
Š
7E
G%Eg%N#
L%%AF%&'(F%GFF%KA5.&'(.G
2.2.4. Một số phương pháp phân loại văn bản:
4%KA57ZH[E6S..G7%B (G6_G.y.G7! (G%N(%•!~%
.M&-%AG.E/#6-y.r! !U6_A_G%7-.iF%#<#g/(€6G.E/#
!&X(G%•G%/.%h(F%KA57ZH[.P!S(.M#g;.%h(!'.%#"E
GFD(%AG7ZH[!'([7R.%Q!U62(D(%A.M:(TUH7-G_1#
1@Q.F%[!&X]U6]¢.j.%iS._hF%&'(F%GF.QFi%L%
SF%KA5:¯7U:gU_\^!&XDO(MS(M•yHV.L%!'([7E
%/#9#[.L%.AG_•D(9#:%/."_#<.u:.r.MA(.E/#%&
%N(.r .%I7E_•D(G.r .%IEg!I(G6S.A5!h7-6S.
A5.E/#
dƒU:MU_.U(%HAM\d^EG%.QFi.%h(@1y!KgEF%&'(
F%GFF%KA57ZH[%L%]G%<.n&:M:6S..E/#yd%B@
0
.E/#.&'(.P.r.iF%#<#g/7E_•D(%N(A5u:G.E/#
Eg!IF%G.%/M:A5u:%N(.E/#!:(!&XF%K-FE/#
!&XHI#D•HV%N(7U.AMu:%N(.r7E_P(h(%:#(N:%:
.E/#!&X!AHš(G%_•D(@%A[(G%aƒ\’#DU:^%:g
%N(%E6@%G(N:%:7U.AMEg
Kg9#gQ.!~%EF%&'(F%GF6Gg%B.P!S(9#g5FGKgF%K
-FDP:.M1DN/#%#<#g/ }.H1.MA(u:Kg9#gQ.!~%
!&X@Q.%XF7-6S.@I6.%•.M16S..%#S.L%7E%G%M:(AEu:
6S.}..&'(2(7-@Q.9#[@I6.%•S.G!&X@Q.%XF7-6S.
A5WPF%KA56S..E/#Hp.!"#.r}.(h7E_:#! .%Z6G}.
H1.MA(!Q@%6S.}.G!&X.R6!Q56 }.y@I6.%•@Q.%XF
7-}.!•!&X.%P.%!I]G!~%}..QF.%UAyA5u:.E/#EA5
u:}.G#hO(
W#FFAM.7U.AM6:%U_\W^W•D(6S.H;6q.9#gQ.!~%!I%:
%N(!I6DN/#7EA.MA(G-FW•(!&X2(D(!IF%K
A57ZH[-6•#!'([%<.u: yG.E/#%#<#g/!&X
DO(%&%N(7U.AMy7E.%#i..AGF%G.%/_1#H;6q.\%gFUMF:U_^
F%K%:.%E%%N(-F@%G%:#u:G.E/#%#<#g/%N(
.E/#@I6.M:!&XF%K-FDP:.%UA%N(7~.MLu:%}(!h7-
_1#H;6q.
4%KA5DN/#7ZH[1@Q.\%gFUM.U].D:.:^W•D(6S..iFDN
/#F%8HQ!I_A_G%_P%/#9#[u:%:.%#i..AG7Ed%A
7/F%K-F.M:(TUH%}(]U6]¢.j.%i.L%%N#L%u:G
_1#1@Q.ySD#(u:%N(.E/#1@Q.7EDN/#.MA(_PF%K
-F7E.R6M:DN/#HQ!8 .%I(:.Z(!S%L%]Gu:_PF%K
-FHV6S.%/_h-
4%KA57ZH[%A7/BTUHk(D(65('ƒMA%K.5A!I
B%N(.M:( SD#(@%1#DK6%}(DO(6S._P.%#.%iF
