On how to search a verb and come out with different forms

[Li He]

Dear Anthony,

I have a question on the KWIC searching. How can I realize input a search term like go*_V*, I will get the results: go, goes, went, going, gone. You know there are many verbs have some special past tense and past particles. I dont't know how to realized this in Antconc. So sometime I have to manually input the special forms in the searching list. 

When I am using Antconc to do the searching, I input go*_V*, I will just get the resutls with go, not the extended result: go, goes, went, going, gone. If there is a setting can help me get  the extended, it would be helpful. 

And there is another question: how to avoid the duplicated results. Let me still take "go" as an example. When I am creating the searching list, I just want to keep the word "go". If I save it as “go*_V*", the searching results will give me: go, govern. And if I met "govern" later I may add "govern*_V*" in my searching list. Oops, after searching the my list, I may get the concordance lines of govern twice. So, how can I avoid such kind of problem when I am using antconc?

And there is another question comes to me: Is there a function that can help me exclude the words that has collected in my searching list? What I am doing right now is extracting some new items. In each round I will exclude the existed items that I have identified with a specific function: showing politeness. When I am trying to extract some new verbs with the function of politeness, I want to exclude these verbs collected in my searching list. And I will just tag other verbs not included in my list. Then I will get some new items. 

Best. 

Helios

[Laurence Anthony]

Hi Helios,

>When I am using Antconc to do the searching, I input go*_V*, I will just get the resutls with go, not the extended result: go, goes, went, going, gone. If there is a setting can help me get  the extended, it would be helpful.

Assuming go, goes, went, going, gone are in the data, you should get all those hits with this search term. Are you sure you have tagged your data properly?

>When I am creating the searching list, I just want to keep the word "go". If I save it as “go*_V*", the searching results will give me: go, govern. And if I met "govern" later I may add "govern*_V*" in my searching list. Oops, after searching the my list, I may get the concordance lines of govern twice. So, how can I avoid such kind of problem when I am using antconc?

This contradicts the first point you made, where you say that you only get results for "go". I also don't understand your use of "oops". Assuming you are doing the searching correctly, you shouldn't get any duplicate results. Perhaps you are seeing the exact same results from different parts of the original corpus. This suggests that maybe your corpus contains duplicate files.

>And there is another question comes to me: Is there a function that can help me exclude the words that has collected in my searching list? What I am doing right now is extracting some new items. In each round I will exclude the existed items that I have identified with a specific function: showing politeness. When I am trying to extract some new verbs with the function of politeness, I want to exclude these verbs collected in my searching list. And I will just tag other verbs not included in my list. Then I will get some new items.

I would recommend that you don't try to use the KWIC interface to do your data processing. Use it to generate results and then copy those results to some other tool (e.g. Excel). Then generate more results and update your Excel file. This process will be much better for tracking what you are doing.

I hope that helps!

Laurence.

###############################################################
Laurence ANTHONY, Ph.D.
Professor of Applied Linguistics
Faculty of Science and Engineering
Waseda University
3-4-1 Okubo, Shinjuku-ku, Tokyo 169-8555, Japan
E-mail: antho...@gmail.com
WWW: http://www.laurenceanthony.net/
###############################################################

--
You received this message because you are subscribed to the Google Groups "AntConc-Discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to antconc+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/antconc/6bdb6ebd-dd3b-4c6c-9424-5c106018ce84n%40googlegroups.com.

[Li He]

Hi Laurence,

>When I am using Antconc to do the searching, I input go*_V*, I will just get the resutls with go, not the extended result: go, goes, went, going, gone. If there is a setting can help me get  the extended, it would be helpful.

Assuming go, goes, went, going, gone are in the data, you should get all those hits with this search term. Are you sure you have tagged your data properly?

> Yes, I check the data, go, wen, gone are in my data. When I search  "go*_V*", and I get three results: governing, gone and go. As for the coding I tagged the data with Tree Tagger, which was automatically finished by LancsBox. 

I tried with LancsBox, it works. In LacnxBox, I gave the headword + POS function, and the system gernerate the results of: went, gone, go. While in Antconc, I searched "go*_V*", I think the system just treat it as the word stat with "go" + suffixes, that's why I just get: governing, gone, and go. That why I though I may ingored some setting in the Antconc, so that I didn't get the different forms of word "go". Is there a similar function in Antconc, that can help me realize inputting a verb, and I can get different forms of this verb (like go we have mentioned).

>When I am creating the searching list, I just want to keep the word "go". If I save it as “go*_V*", the searching results will give me: go, govern. And if I met "govern" later I may add "govern*_V*" in my searching list. Oops, after searching the my list, I may get the concordance lines of govern twice. So, how can I avoid such kind of problem when I am using antconc?

This contradicts the first point you made, where you say that you only get results for "go". I also don't understand your use of "oops". Assuming you are doing the searching correctly, you shouldn't get any duplicate results. Perhaps you are seeing the exact same results from different parts of the original corpus. This suggests that maybe your corpus contains duplicate files.

> Yes, sorry for the contradiction. The results are: governing, gone and go. And in the results from Antconc, I found the concordance lines with governing appeared twice. That's why I have the second question. 

>And there is another question comes to me: Is there a function that can help me exclude the words that has collected in my searching list? What I am doing right now is extracting some new items. In each round I will exclude the existed items that I have identified with a specific function: showing politeness. When I am trying to extract some new verbs with the function of politeness, I want to exclude these verbs collected in my searching list. And I will just tag other verbs not included in my list. Then I will get some new items.

I would recommend that you don't try to use the KWIC interface to do your data processing. Use it to generate results and then copy those results to some other tool (e.g. Excel). Then generate more results and update your Excel file. This process will be much better for tracking what you are doing.

> Yes, this is what I am doing right now. But it took my a lot of time to manually remove the duplicated items. That's why I think whether there is a way can help me save the time.

Thanks.

Helios