Dear Anthony, I'm planning on analyzing the structural moves of narratives written by ESL students. I was wondering if it would be possible to have a short video tutorial on how to use AntMover. The word document on your website is very useful but it's still quite advanced for my ICT skill level. I understand that the programme's default settings are geared towards the analysis of academic abstracts. How can settings be retrained to analyze narrative texts based on Lavob & Waletzki's narrative stages? Thanks in advance
Hello dear Mr. Anthony We're building a Persian corpus. Your software - Antconc - is the best corpus manager tool to me. But there's a problem. Antconc doesn't support RTL languages like Persian. I mean it works with some issues: 1- When it extracts wordlist from the texts, it doesn't process ZWNJ properly; which is very important in processing compound nouns or plural nouns. Examples [note: the "-" in the romanized version is the place where ZWNJ is inserted]: آدمها / Aadam-ha / human-plural / humans توقفناپذیر / Tavaghof-napazir / stop-unable / unstoppable But Anctconc treats this instance as two separate tokens, while it shouldn't. It seems it is treating ZWNJ like normal Space. 2- In the concordance section, the text is not right to left. Would you please help us? We can help too. Sincerely yours, Pooya.
Hi. AntConc 3x wasn't fully RTL compatible. Can you try my new AntConc 4 version (just released this week - Dec 2021)? I'll be releasing new RUclips videos for that soon.
@@AntLabJPN Hello. Thanks for your response. I just checked the new version. RTL WORKS fine. [I want to give u a screenshot, but RUclips deletes comments that include links, so I've emailed it to you] Hats off to you, sir. The other problem (ZWNJ) is still there. Is there any way we can help to fix this problem as well? If by any chance you do want to fix the ZWNJ issue, I can rethink/rewrite/recheck everything about it, to provide you with a better explanation of the problem. Mr. Anthony, you have no idea how grateful I am for your efforts. Thanks a million times.
@@pooyaprvsh3948 Hi, that's great. As for the second problem, can you try adding marks to the token definition when you create your corpus My guess is that the token definition is missing the character. If that doesn't work, just append the character directly.
Hello Sir. Thanks for this tutorial. I've downloaded and installed the app on my pc. However, when I select Open Files, I do not see any corpus to choose from. Kindly help me out
Dear Laurence, I tried to do my best but fail to master a function of regex search. I copied the expression for the search you demonstrated: \br[a-z]+?t\b but get No hit as a result of the query. What's wrong with it or what is the secret?
Thankyou so much for this. I'm going to be using this for language learning.
You're very welcome!
Fabulous. Thanks a lot for sharing this, Professor.
thank you Professor
Dear Anthony, I'm planning on analyzing the structural moves of narratives written by ESL students. I was wondering if it would be possible to have a short video tutorial on how to use AntMover. The word document on your website is very useful but it's still quite advanced for my ICT skill level. I understand that the programme's default settings are geared towards the analysis of academic abstracts. How can settings be retrained to analyze narrative texts based on Lavob & Waletzki's narrative stages? Thanks in advance
Sir, please help me. My teacher asked me to divide the brown corpus into its categories. I don't know how to do it?? Any idea!! Thank you sir
You can find the categorized version here:
www.nltk.org/nltk_data/
Thank you a lot . I would like to ask you about textes which textes can I load please?
AntConc 3.5.9 only works with text (.txt) files. But you can use my AntFileConverter tool to convert Word and PDF files into the correct format.
@@AntLabJPN Thank you very much
Hello dear Mr. Anthony
We're building a Persian corpus. Your software - Antconc - is the best corpus manager tool to me.
But there's a problem. Antconc doesn't support RTL languages like Persian. I mean it works with some issues:
1- When it extracts wordlist from the texts, it doesn't process ZWNJ properly; which is very important in processing compound nouns or plural nouns. Examples [note: the "-" in the romanized version is the place where ZWNJ is inserted]:
آدمها / Aadam-ha / human-plural / humans
توقفناپذیر / Tavaghof-napazir / stop-unable / unstoppable
But Anctconc treats this instance as two separate tokens, while it shouldn't. It seems it is treating ZWNJ like normal Space.
2- In the concordance section, the text is not right to left.
Would you please help us? We can help too.
Sincerely yours, Pooya.
Hi. AntConc 3x wasn't fully RTL compatible. Can you try my new AntConc 4 version (just released this week - Dec 2021)? I'll be releasing new RUclips videos for that soon.
@@AntLabJPN
Hello. Thanks for your response.
I just checked the new version. RTL WORKS fine. [I want to give u a screenshot, but RUclips deletes comments that include links, so I've emailed it to you]
Hats off to you, sir.
The other problem (ZWNJ) is still there. Is there any way we can help to fix this problem as well?
If by any chance you do want to fix the ZWNJ issue, I can rethink/rewrite/recheck everything about it, to provide you with a better explanation of the problem.
Mr. Anthony, you have no idea how grateful I am for your efforts. Thanks a million times.
@@pooyaprvsh3948 Hi, that's great. As for the second problem, can you try adding marks to the token definition when you create your corpus My guess is that the token definition is missing the character. If that doesn't work, just append the character directly.
@@AntLabJPN
Thanks for the instructions. The problem is solved.
@@pooyaprvsh3948 That's great!
Hello Sir. Thanks for this tutorial. I've downloaded and installed the app on my pc. However, when I select Open Files, I do not see any corpus to choose from. Kindly help me out
Hi, you have to supply your own corpus. Basically any plain text files will work.
Where do we get the files you opened through the sofware? do you have to download that separately?
either download a ready made corpus from the net like BNC or you start collect your own text corpus
www.laurenceanthony.net/resources/wordlists/
Sir can you guide us regarding MI value and t_score
I suggest you Google the terms. They are very standard in corpus linguistics.
Dear Laurence, I tried to do my best but fail to master a function of regex search. I copied the expression for the search you demonstrated: \br[a-z]+?t\b but get No hit as a result of the query. What's wrong with it or what is the secret?
Did you remember to click the "Regex" option. Perhaps try an easier regex search: [a-z]+
@@AntLabJPN I remebered to do this soon after I had asked you :). Anyway, thank you for your response!
@@AntLabJPN I did it it's working thanks sir