关键字查询问题

chenkan2000

用InterMedia Text可以的，它有两个中文词法分析器：chinese_vgram_lexer和chinese_lexer。chinese_lexer比chinese_vgram_lexer聪明得多，它分析中文词组的正确性更高，但chinese_lexer只能用在utf8字符集的数据库上，chinese_vgram_lexer没有限制。然后把干扰的（的，和）加入到stoplist中。
不要忘了索引的更新！！！

chenkan2000

如果是Oracle9i，数据库字符集是utf8，可以用chinese_lexer，对中文支持比较好。

yanwang

check this link at ASKtom site

http://asktom.oracle.com/pls/ask ... 0and%20%7Bsearch%7D

ops$tkyte@ORA817DEV.US.ORACLE.COM> create table t ( data varchar2(50) );
Table created.

ops$tkyte@ORA817DEV.US.ORACLE.COM> insert into t values ( 'How#Now#Red#Cow' );
1 row created.

ops$tkyte@ORA817DEV.US.ORACLE.COM> insert into t values ( 'How Now Red Cow' );
1 row created.

ops$tkyte@ORA817DEV.US.ORACLE.COM> insert into t values ( 'How # Now Red Cow' );
1 row created.


ops$tkyte@ORA817DEV.US.ORACLE.COM> exec ctx_ddl.drop_preference('my_lexer');
PL/SQL procedure successfully completed.

ops$tkyte@ORA817DEV.US.ORACLE.COM> begin
  2      ctx_ddl.create_preference('my_lexer','BASIC_LEXER');
  3      ctx_ddl.set_attribute('my_lexer','printjoins','#');
  4  end;
  5  /

PL/SQL procedure successfully completed.

ops$tkyte@ORA817DEV.US.ORACLE.COM> create index t_ind_ctx on t(data) indextype
  2  is ctxsys.context parameters('lexer my_lexer');

Index created.


no here are all of the tokens we have parsed as "words"

ops$tkyte@ORA817DEV.US.ORACLE.COM> select token_text from DR$T_IND_CTX$I;

TOKEN_TEXT
----------------------------------------------------------------
#
RED
COW
HOW
HOW#NOW#RED#COW
NOW

6 rows selected.

ops$tkyte@ORA817DEV.US.ORACLE.COM> select * from t where contains( data, 'How' )
> 0;

DATA
--------------------------------------------------
How Now Red Cow
How # Now Red Cow

ops$tkyte@ORA817DEV.US.ORACLE.COM> select * from t where contains( data, '#' ) >
0;

DATA
--------------------------------------------------
How # Now Red Cow

dingjun123

good