Lexical coverage of spoken discourse
- 主题发起人 xiaoz
- 时间
Haiyang Ai
Administrator
Thanks for the article Richard.
Maybe it's a good topic to do a similar research on Chinese lexical coverage of spoken discource.
What do you think?
Maybe it's a good topic to do a similar research on Chinese lexical coverage of spoken discource.
What do you think?
Agree with Ocean that this is a great topic.
Afraid wordlists are not a solution in this case. In all registers and genres, function words such as the and of will sit on top of the frequency lists. The key lies in keywords and key keywords (WordSmith).
Afraid wordlists are not a solution in this case. In all registers and genres, function words such as the and of will sit on top of the frequency lists. The key lies in keywords and key keywords (WordSmith).
[本贴已被 作者 于 2005年06月21日 17时43分12秒 编辑过]
回复：Lexical coverage of spoken discourse
Afraid wordlists are not a solution in this case. In all registers and genres, function words such as the and of will sit on top of the frequency lists. The key lies in keywords
^^^^^^^
and key keywords (WordSmith).
^^^^^^^^^^^^
How such (key) key words be worked out in addition to the stop list of function words?
Afraid wordlists are not a solution in this case. In all registers and genres, function words such as the and of will sit on top of the frequency lists. The key lies in keywords
^^^^^^^
and key keywords (WordSmith).
^^^^^^^^^^^^
How such (key) key words be worked out in addition to the stop list of function words?
A stop list might help in this repsect. In theory a combination of stop list and (key) keywords is possible.
Load the stop list (Adjust settings - Lemma, match and stop lists) before making a wordlist for the resarch corpus and the reference corpus. Then extract (key) keywords as usual.
Load the stop list (Adjust settings - Lemma, match and stop lists) before making a wordlist for the resarch corpus and the reference corpus. Then extract (key) keywords as usual.