请问如何设置后，TOSCA——LOB tagger可以按段落标注？

[armstrong]

下面的段落有两个句子：
Life has its ups and downs, its peaks and its valleys. No one is up all the time, nor are they down all the time.
经过TOSCA——LOB tagger标注后，经过处理成为如下的文本。
<s>
Life_NN has_HVZ its_PPG ups_NNS and_CC downs_NNS ,_SCOM its_PPG peaks_NNS and_CC its_PPG valleys._SPER
</s>
<s>No_NP one_CD1 is_BEZ up_RP all_ABN the_ATI time_NN ,_SCOM nor_CC are_BER they_PP3AS down_RP all_ABN the_ATI time_NN ._SPER
</s>
显然是将段落分成句子了。
请问如何设置后，TOSCA——LOB tagger可以按段落标注？
谢谢！

 

[刘语料]

回复: 请问如何设置后，TOSCA——LOB tagger可以按段落标注？

偶也有类似的问题,希望专家给予帮助.
谢谢!

 

[xiaoz]

回复: 请问如何设置后，TOSCA——LOB tagger可以按段落标注？

You can try marking the paragraph boundaries before tagging it.

 

[armstrong]

回复: 请问如何设置后，TOSCA——LOB tagger可以按段落标注？

You can try marking the paragraph boundaries before tagging it.

I add <p> and</p> at the beginning and the end of the paragraph repectively,but the result is not with the boundary of the paragraph marks <p> and</p>.

 

[xiaoz]

回复: 请问如何设置后，TOSCA——LOB tagger可以按段落标注？

I haven't tried this tagger, it appears that the tagger ignores all existing tags, i.e. everything in <>. Then how about trying cheating the tagger by using some non-word strings to indicate para boundies (e.g. ppppp as para opening and qqqqq as para end) and after tagging, change ppppp into <p> and qqqqq into </p>? - unless you have found a way to adjust the tagger setting.

 

[armstrong]

回复: 请问如何设置后，TOSCA——LOB tagger可以按段落标注？

I'll have a try as your intructions, thanks a lot

 

[armstrong]

回复: 请问如何设置后，TOSCA——LOB tagger可以按段落标注？

thanks, Dr.Xiao. I tried, but it didn't work.