??? ????? ?????? ????????????? / ??????????? ?????????? (????, ???????????? ????????????????

1
10.2. ?????????? ?????????????? ??????????
??????? ? ?????????????? ????????? ???
??????????????? ??????????????? ??????????????
2
???????? ??????????????
3
?????????????? ???????

• ??? ????? ?????? ????????????? /
  ??????????? ?????????? (????, ????????????
  ???????????????? )
• ????? ???????????
• ???????? ???????
• ?????? ???????? ???????

4
????? ?????? ???? ???????????

• ???????? ?????????? ???????
• ??????????? ?????? ???????
• ???????? ???????
• ?????? ???????? ???????

5
???????????? ?????????
?????????? ??? ?????? ?????????? ????? ???????
?????????? 60
6
?????????????? ??????????????
7
?????? ?????????????? ??????????
v      ?????????? ??????
??? ??????, ?????????? ?? ??????? v     ????????
????????

• ????????? Reuters-21548
• 135 ??????,
• ???????? ?????????? ?????????
• ???????? ? ??????? gt 80

SIGIR 2001, SIGIR 2002 Operational Text
Categorization Workshop
8
?????? ????????? ???????? ? ??????? ??????????
9
?????????????/????????????? ???????
10
??? ????? ????????
11
????????? ?? ??????? ???????


12
????????? ?? ????????? ???????(kNN)
13
????????? ?? ????????? ??????(SVM)
d
c
14
??????????? ???????? ?????????SVM (Support
Vector Machines)
???????????? ?????????? ????? ?????
????????????? ??????????????? ???????????
15
?????????? ???????????
16
Reuters-21578, SVM, (PR)/2 ? max(J)
(ModApte split)
17
??????? ?????????? ?? Reuters-21548
CHINA'S HEILONGJIANG PROVINCE BOOSTS GOLD
OUTPUT Gold output in the northeast China
province of Heilongjiang rose 22.7 pct in 1986
from 1985's level, the New China News Agency
said. It gave no figures. It said the province,
China's second largest gold producer after
Shandong, plans to double gold output by 1990
from the 1986 level. China does not publish gold
production figures. However, industry sources
estimate output at about 65 tonnes a year, with
exports put between 11 and 31 tonnes. China is
selling more gold abroad to offset large trade
deficits in recent years, western diplomats said.
REUTER CONSOLIDATED TVX TO BUY BRAZIL GOLD MINE
STAKES ltConsolidated TVX Mining Corpgt said it
agreed to issue 7.8 mln treasury shares to
acquire interests in three gold mining companies
in Brazil and an option to increase the company's
interest in a platinum property. The company said
the transactions will bring immediate production
and earnings to Consolidated TVX, enhance its
precious metal potential and is expected to
improve cash flow and earnings on a per share
basis. The company did not give specific figures.
Consolidated TVX said it will acquire 29 pct of
CMP, a public gold mining company in which TVX
already holds a 15 pct interest, making TVX the
largest single shareholder. The company also
agreed to acquire a 19 pct stake in Novo Astro, a
private company, and a 16 pct interest in Teles
Pires Mining, increasing the TVX's ownership to
51 pct. In addition, Consolidated TVX said it
will acquire the right to add a 10 pct interest
to a platinum property in which it already owns a
29.4 pct stake. CMP earned 11 mln Canadian dlrs
in 1986 and expects to produce 42,000 ounces of
gold in 1987 at a cost of 160 U.S. dlrs an ounce,
Consolidated TVX
18
??????? ??????
 
??????????? ???????? ????????
19
??????? ?????? ?????????????? ?????????? ???????

• ?????? ??????????? ?????? 300-500 ??????,
  ?????? ?? ??????? ?????????
• ?????? ?????????? ??????????? ?? ????????
  ? ?????????? ????????? ?????????,
  ???????????? ??????? ??????????????
  (????????? ?????????) ??????????? ??????????
• ????????? ????? ????? ????????? ??????

20
????????? ???????? ??????????? ? ?? ????? ????
??????? ? ???????? ?????

• ?????????? ??????????????? ?????
  (www.socialpolicy.ru)
• ?????? ?????????? ?????? ???????????
• 350 ??????, 4 ?????? ????????
• ????? ?????? gt ?????????? ????????

21
????????? ???????? ??????????, ?? ?????????????
?????????? ? ????????

• ????????????? ??????? ?????????? RePec
  (www.repec.org), SocioNet (www.socionet.ru)
• ????? ????????????????? ?????????? ??
  ????????? ? ??????????
• ?????????? Journal of Economic Literature
  Classification System (JEL )
• ????? 700 ??????
• ????? ??? ??????????? ??????? ? ????? ??????

22
????????? ???????? ????????????? ? ????????????
??? ??????????? ??????(????? ???????
??????????????)

• ?????????? ???????? ?????????
• ????????????? ????????????? (???? ?511
  15.03.2000) - 1168 ??????
• ????????? ???????? 10,000 ??????????
  ?????????????????? ???????
• ?????? ??? 47 ?????? ????? ??? 100 ???.,
  ?????? ??? 200 ?????? ????? ??? 20 ???.
• Inconsistency ???? ???????????? ?????????
  ????? ?????? ?????? ??????

23
????????? ???????? ??? ???????? ?? ??????
?????????

• ??????? ????????? ???????????? ??????
• ???????? ?????????????? 600,000
  ???????????? ??????????
• ??? ?? ??????????
• ??????? ?????????, ??????? ????????
• ??!!!
• ??????????? ????? SVM-light, ????????? ??
  ??????????? ?????????? ?? ??????????? ?? ?????
  ??????? ??? 50 ??????????

24
?????????? ????????? ??? ??????? ??????? ?????
??????????
25
??????? ?????????????? ??????????
??????????? ????????? ??????? ??????????????
?????????? ?? ?????? ?????????????
????????? ?????????????? ???????????
????????????? ???????? ?? (450 ??????, 4
??????) ?? ???????? ???????? ??????
????????? ????????????????? ?????? ????????? ???
(80 ??????) ?? ?????????
??????????? ?????? ?????????????? ????????????
(180 ??????, 3 ??????)
????? ??????? ?????????????? ??????????

• ?????????????? ?????????? ?? ??????????????
  ???????? ????? ?? (???? ?????????? ?? N511
  ?? 15 ????? 2000 ?., 1169 ??????)
• ?????????????? ?????????? ?? ?????????????? ???
  ?????? (3200 ??????)

26
?????????? ??????????????? ??????????????

• ????? ?? ??????, ????????? ? ???????????-?????????
  ??? ?????????
• ????????????? ??????? ? ???? ?????????? ???????
  ??? ?????????? ????? ??????? ?????????, ?????
  ?????????????? ?????????? ? ??????????????
  ????????????? ????????? ?????????
• ??????????? ?? ??????????? ??????????? (?????????
  ??????? ???????????) ?????????????? ????????????
  ?????? ?????? ????????? ? ?????? ?????????????
  ??????? ????????, ????????? ????????,
  ??????????????? ???????? ???? ? ??????? ??????????

27
????? ???????? ???????
???????
????????????2
????????????1
???
?11
?12
?13
?21
???????22
?
?
?
???
???


-

28
???????? ?? ???????????-???????????? ?????
27 ????? ???????, 64 ?????? ???????? 105 ?????
?????????

• ?????????? ??????? ????????
  ???????????? ????????
• ?????? - ???????????? ????? ????????
  ??? ?????????, ???????, ???????,
  ????????????????, ??????? ????????, ?????,
  ?????????, ??????? ????????? ? ??.
• ???? ?????????????? ???????
  ??????????? ?????????, ?????????????
  ????????, ??????, ???????? ??????, ?????????
  ?????????

29
??????????? ??????????? ??? ???????? ?????????
??????? ??????? ?????????

• ???????? ?????? ??????? ???
  ????????? ??????? ??? ????????? ?????????
• ????????? ??????????? ????????? ???????
  ?? ?????? ???????? ????????? ? gt
  ?????????? ??????? ?? ?????? ?
• ?????????? ??????????? ????????? ?
  ??????, L ?? ?????, N ??? ??????????
• (??????????????? ???????) L
  (?????????? ?????????) N
• ???????? ??????????? ??????????? ???????
  ??? ??????????? ?????????
• ??????????? ???????? ???????????? ???????
  ?????????? ????????? ? ????????????

30
????????????? ?????? ??????? ???????? ?????????
31
??????????? ????????????? ??????? ?????????
?????????
32
?????????? ??????. ????????????? ???????

• 040000000 ???????
• 040050000 ???????? ???? /???????/
• ???
• ???????
• ? ?????
• ? ????,??????????????????????????
• ???
• ???? ?????L //????,????,????,???????,
• ? ???????????????????????????????
• ???
• ?????
• ? ???????????????????????????????

33
?????????? ??????. ????????????? ???????

• 380000000 ????????????? ???????
• 380020000 ????????? ? ??????? /?????????????
  ???????/
• ?????????E????????????? ???????E
• ? ?? ??????? ?????????E
  ??????? ?????????E???????E
• ? ???????????????????? ???????E
• ? ??????????????????????????????L
  ?????????????????????????????????
  ????????????????????????????????
• ??????? gt ??????????, ???????, ?????????,
  ??????, ???????????????, ????????

34
????????? ??????? ???????. ??? ??? ?

• ????????? ???????-???????
• ??? ?????????????
• ? ??????????????
•  
• ???1
• ??????? ?????????????? - ???????????? (???
  ?????????????) ??????? (? ??????????????)
•  
• ??????? ? ?????????????? ??????????? ??? ??????
  ??????? ?????? ?????, ????? ? ?????? ???? ???? ??
  ???? ???????, ??????????? ? ???? ??????? ???
  ?????????????
• ????????????? ????? ???? ???????? ?? ??????
  ????????? ??? ??????????? ???????

35
???? ???????????? ????? (?????????????
????????????? ?? ?? 26 ???? 1995 ?. N 604)
36
????????? ????????????? ?????????????
???????? ???????????? ????
????????? ???????????? ????
37
?????????? ???? ??????? ?? ?????? ?????????????
????????????? ??????
a 0.7 ???? ???? ???????? ????????? ?
???????????? ????????????? ? ???????????   ???
????????? ????????????? ?????????????   0.9 - ???
?????? ????????? ????????????? ????, 0.7 - ???
???????? ????????? ????????????? ????,
0.75 - ??? ?????? ?????????? ????????????? ????
38
?????? ???? ??????????
??? ?????????? ???????????? ????????? ??
?????? ????? ????? ???????????? ??? ??????????,
?? ? ???? ???????? ?????????? ? ??????
????? ???? ????????? ?????? ????? ?????????
?????? ????????? ? ????????? ???????,
???????? ?? ???????? ???????????? ?????????
????? ????? ?????? ????? ?????????
??????. ???? ???? ????? ?????? ??????? ??? ??????
????????? ?????????? ? ????????? ?????
????????, ???? ??????? ????????? ??????????
??????????? ? ?????? ?????? ??????
39
?????????? ???? ???????
??? ??????????
??? dijk - ???????, ?? ????????? ?????????????,
pijm - ???????, ????????? ?????????????, ? - ??
??????? ?????? ???????, ????
??????? ???????, ?? ?????????
?????????????, ? ???? ?????
40
????????????? ?????????? ??????????????????????
(?? 26 ??????? 2001 ?. N 905)
? ?????????? ?????????? ????? ??????????????
?????????? ????????? ? ??????????????
????????????? ? ?????????????? ? ???????
???????????? ? ????????????? ????????????
???????????? ? ?????? ?????
090090000 ??????????? ???????????? 64 T130817
???? ?????? 12 64 T4200
????????? 8 63 T106444 ???????? 8 63 T6783
??????-???????? 6 62 T955
???????????? 6 62 T956 ???????????
??????? 4 61 T131907 ?????? ???????????
????????? 3 61 T1023 ???????????
??????? 2 61 T1022 ???????????
????? 1 60 T137619 ???????????
???????????? 1 60 T9340 ???????????
???????? 1 60 T5962 ??????????? 1 60 T24
97 ????????????? ??????? ????? 1 15
41
???????? ??????, ??????????? ????????? (??????????
??? ????????????? ?? ?? 26 ???? 1995 ?. N 604)

• 150100020 ?????????? ? ??????? ???????????
  ??????? ?????? 95
• T791 ??????? ??????
  29 95
• T792 ??????????????
  17 76
• T6999 ???????? ?????? ?? ?????????
  2 62
•  
• 090040000 ?????????????
  81
• T55 ?????????????
  12 81
• T482 ???????? ?????????????
  4 64
• T6217 ???????-???????????? ??????????
  2 62
•  
• 020010040 ??????????? ?????? ??????????????
  ?????? 77
• T2576 ???. ????? ?????????????? ??????
  7 77
• T5109 ???????????? ???????
  3 63
•  
• 150040000 ????? ??????????? ??? ??????????
  ????????? 76
• T792 ??????????????
  17 76
•  
• 070070020 ??????????????? ? ???? ??????????
  ??????? 75
• T124039 ???????????
  15 75

42
??????? ?????????????? ??????????
??????????? ????????? ??????? ??????????????
?????????? ?? ?????? ????????????? ?????????
?????????????? ??????????? ?????????????
???????? ?? (450 ??????, 4 ??????) ?? ????????
???????? ?????? ????????? ?????????????????
?????? ????????? ??? (80 ??????) ?? ?????????
??????????? ?????? ?????????????? ????????????
(180 ??????, 3 ??????)
????? ??????? ?????????????? ??????????

• ?????????????? ?????????? ??
• ?????????????? ???????? ????? ??
• (???? ?????????? ?? N511 ?? 15 ????? 2000 ?.,
  1169 ??????)

43
????????????? ??? legal-?????????

• ?????????????? ????????????? ???????????
  ?????????? ???????????????? ?? ?? ?? ??? ??????
• 183 ?????? -- ???????????? ????????
  ?????????????? ??????????? ??????????? ??????????
• ??? ???????? ????????? ????????????? ????????????
  ????????? ?? 4496 ??????????, ?????????????????
  ?? ??????? ?????????????? ?????????? ????????
  ??????
• ??? ???????????? ????????????? 55519 ??????????,
  ??? ??????? ?????????? ????????????? ??????????
  ???????, ? ??????? ??? ????????? ?????????.
• ??? ????????? ?????? ??? ?????????? ? ?????????
  ????????, ????? ?????? ? ????????? ???????????
  ?????????? ??? ???????? 170

44
????????????? ??? legal-?????????
?????? 1 SVM ?? ?????? ?????? 2 SVM ??
?????????????? ?????/???????, ?????????????
?????, ??? ? ??????? ??????????, ???? ???????.
21746 ????????? ???? ? 1203087 ???
?????-???????? ??? ????????? ??????? ?? 4496
??????????. 29918 ????????? ????/???????? ?
1569958 ??? ?????/??????-????????.
45
????????????? ??? legal-?????????

• ?????? 3 ????? ????????? ????????,
  ?????????? ?? ????????????? ?????? ???????????
• ???????? ??????? ? ???? ????????? ???????
  ??????? ? ??? ?????????? ?????? ????????
  ??????? ????????? ???.
• ???????? ?????? ??????? ????
• ??????????, ???????????? ???????, ?????
  ????? ?? 1 ?? 3.
• ????????? -- ??????? ???????? ?????????
  ????????, ??????? ?? ??????????? ?????
  ???????, ???????????? ?????????, ?? ???????????
  ??????????????. ??????????? ???????????
  ??? ??????? ????????? ???? ??????????
  ?????? ???????? ???????, ??????? ????? ?????
  ????????????????.

46
????????????? ??? legal-?????????
??????? ?????????????, ????????? ?? ??????,
????????????? ?????????? ?? ?????? ??? ??????
?? ???? ????? ?????????? ideal50.
47
????????????? ??? legal-?????????
?????????? ???????? ?????????? ??? ???????
????????????? ideal50
48
????????????? ??? legal-?????????
??????????? F-???? ?? ?????????? ???????? ???
???????? (? ??????? ??? ??????, ???????????
??????? ???????? ? ????????? ????????)
49
??????? ? ??????

• ??????????? ?????? ?????????????? ??????????.
• ?? ????? ???????? ????????? ????????? ? ???????
  ?????????????? ?????????? ????????
• ????? ??????????? ??? ????????? ??????? ??
  ??????????????.