Thank you sooooo much !! I always find it hard to understand these information,and with my thesis presentation starting tomorrow, you are certainly a life-saver!
You're very welcome! I'm glad you found the video useful. Good luck with your research. Also, check out the new AntConc version 3.5.6, which offers more stats and options!
Thank you for your kind reply! Just found out you also developed a series of free softwares addressing segmentation and tags, which is really a fortune for linguistic students!! Thank you so much for doing so. I'm just wondering what's the difference between the Effect and Keyness in the key word part appearing in the version of AntConc3.5.2. I would be really grateful if you can tell me where to find the algorithm or the formula used to calculate these two statistics. Thank you again!
Hi Laurence, Your videos are a life saver. Thank you so much for sharing. I was wondering if there is a way to make a key word cluster list. looking at multi-word units as well as single words? THank you
Hi. I think there is with the new dev version of AntConc. It allows you to upload target corpus word lists (and reference corpus word lists), so if they consists of MWUs, you should be able to do it.
Hello Laurence Anthony I don't know if you answer questions but may I ask why a few of the infrequent words in the A,B, C corpus had a higher keyness value than the frequent words in the same corpus. For example, words like 'you' and 'I' in the Brown corpus A, B and C had a keyness score of over 300 The other infrequent words quickly drop down to values that did not surprise me but the first few infrequent words had high keyness values. Why please?
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 Laurence, once again thank you for your tutorials. I'm an Italian psychology student and I'm writing a thesis for my master's degree. I would like to use the AntConc program, but I can’t find the material. I would like to ask if you have or know where I can find a Lemma list or Keyword list for Italian language (possibly spoken). Thanks for your help
Hallo Mr Anthony, I am an Italian student and I am trying to learn how to use this software. I have uploaded the reference corpus, but the keywords list indicates articles as words with the highest keyness. What could be wrong? Thank you for your help
HI, thank you for your really clear explaination. however I could not creat a word list as I do not have your interface loading file in the wordlist tool partition. Do you have any solution for that...thanks in advance
What do you mean by "interface loading file"? To create a corpus, you just need to load your corpus files. If you try AntConc 4, you'll find that it now comes with several pre-built corpora.
@@dyousreyaalhamshary8454 Make sure you download the latest version (on April 24, 2020, that's version 4.0.9). Also, you load files via the "Raw Files" tab. Perhaps you need to read the Help page.
Hi, yes it can. You just need to append the hyphen character to the token definition. In AntConc 3, this is in the global settings. In AntConc 4, which I recommend you use, you'll find it in the corpus manager.
Hello, is there a way to make the keyword list omit articles and prepostions and similar words? M keyword list keeps showing them no matter how many times I repeat the process shown in the video. Thank you in advance.
I suspect you are not loading the ref corpus correctly. But, if you are, you can omit words using a wordlist tool stoplist in the word list preferences.
Thank you for the fast reply. The stoplist did help to filter the word list, but then something strange happened. I loaded the filtered word list to the key word tool preferences and when I generared the keyword list, only the words in the stoplist appeared as keywords. I checked the word list I loaded and none of the words of the stoplist appear there. Even when I closed the program and opened it again, loading the filtered stoplist again resulted in the same strange keyword list.
Forget my first reply. I guess I wasn't loading the reference corpus correctly. I had loaded 10 files for analysis then loaded the same 10 files into the reference corpus. Loading a single file for analysis then loading the rest of the files into the reference corpus fixed the keyword list.
Dear Laurence Anthony I was trying to calculate keyness myself, because I am very interested in how it works mathematically. I tried to do it according to ucrel.lancs.ac.uk/llwizard.html , but I don't get the same results. So I am very curious what the formula looks like? Can I find it somewhere I have not looked yet. Or do you mind sharing it?
@@AntLabJPN Thank you for your answer! the example I try to get the same result as AntConc is: 29 appearances in 2287 word corpus vs. 87 appearances in a 54'018 word corpus. So, I hope I understand your formula right. I calculate (if you put it in Wolframalpha it looks better): 2*(29*ln(29/(2287*(116/56305)))+87*ln(87/(54018*(116/56305)))+2258*ln(2258/(2287*(56189/56305)))+53931*ln(53931/(54018*(56189/56305)))) which doesnt get me the result AntConc gets when I use (any) Keyword test. AntConc gives 4-term value 52.17 and 2-term value 51.97 when I add the corpora files as shown in the video. Thus, the confusion.
@@ggames33 Hi. How do you know these are the numbers that AntConc is using for your particular data? Can you send a screenshot of the results page to my email addres?
Hi Bing, yes, you can load in a word list via the Word List Tool preferences menu. Make sure that the format matches the format that AntConc outputs. Good luck!
![BLACK BAG - Official Trailer [HD] - Only in Theaters March 14](http://i.ytimg.com/vi/Du0Xp8WX_7I/mqdefault.jpg)
You manage to explain everything so clearly in this videos...Really, thank you so much for your excellent work. You saved me there!
You're welcome!!
Thank you sooooo much !! I always find it hard to understand these information,and with my thesis presentation starting tomorrow, you are certainly a life-saver!
You're very welcome! I'm glad you found the video useful. Good luck with your research. Also, check out the new AntConc version 3.5.6, which offers more stats and options!
Thank you for your kind reply! Just found out you also developed a series of free softwares addressing segmentation and tags, which is really a fortune for linguistic students!! Thank you so much for doing so.
I'm just wondering what's the difference between the Effect and Keyness in the key word part appearing in the version of AntConc3.5.2. I would be really grateful if you can tell me where to find the algorithm or the formula used to calculate these two statistics. Thank you again!
Finally i found it. Thank you sir☺️
Great!
Hi Laurence,
Your videos are a life saver. Thank you so much for sharing.
I was wondering if there is a way to make a key word cluster list. looking at multi-word units as well as single words?
THank you
Hi. I think there is with the new dev version of AntConc. It allows you to upload target corpus word lists (and reference corpus word lists), so if they consists of MWUs, you should be able to do it.
Wonderful. Thank you very much for your response. And thank you again for your videos.
Hello Laurence Anthony
I don't know if you answer questions but may I ask why a few of the infrequent words in the A,B, C corpus had a higher keyness value than the frequent words in the same corpus.
For example, words like 'you' and 'I' in the Brown corpus A, B and C had a keyness score of over 300
The other infrequent words quickly drop down to values that did not surprise me but the first few infrequent words had high keyness values. Why please?
I don't know if the problem comes from me, but I can't download a reference corpus! Antconc keeps crashing everytime!
Could you help please ?
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 Laurence, once again thank you for your tutorials.
I'm an Italian psychology student and I'm writing a thesis for my master's degree. I would like to use the AntConc program, but I can’t find the material. I would like to ask if you have or know where I can find a Lemma list or Keyword list for Italian language (possibly spoken).
Thanks for your help
Hi Asya, how about typing "italian lemma list" into Google. I think you'll find lots of results that are relevant.
Hallo Mr Anthony,
I am an Italian student and I am trying to learn how to use this software. I have uploaded the reference corpus, but the keywords list indicates articles as words with the highest keyness. What could be wrong?
Thank you for your help
HI, thank you for your really clear explaination. however I could not creat a word list as I do not have your interface loading file in the wordlist tool partition. Do you have any solution for that...thanks in advance
What do you mean by "interface loading file"? To create a corpus, you just need to load your corpus files. If you try AntConc 4, you'll find that it now comes with several pre-built corpora.
@@AntLabJPN thanks very much for your response. when I try to load my file in the prebuilt the application closes.
@@dyousreyaalhamshary8454 Make sure you download the latest version (on April 24, 2020, that's version 4.0.9). Also, you load files via the "Raw Files" tab. Perhaps you need to read the Help page.
@@AntLabJPN Thank you I will.
thanks for this good presentation, can Antconc accept words with - like porte-feuille; cache-colle...
Hi, yes it can. You just need to append the hyphen character to the token definition. In AntConc 3, this is in the global settings. In AntConc 4, which I recommend you use, you'll find it in the corpus manager.
@@AntLabJPN thank you very much, i will search it.
i m using AntConc '4 . I have an other question please, it is about propre nouns without capital letter in arabic, how can i mark them?
@@AntLabJPN dysfonction antconc4
@@nad-im6629 What do you mean by "mark them"?
So helpful 😍
Hello, is there a way to make the keyword list omit articles and prepostions and similar words? M keyword list keeps showing them no matter how many times I repeat the process shown in the video. Thank you in advance.
I suspect you are not loading the ref corpus correctly. But, if you are, you can omit words using a wordlist tool stoplist in the word list preferences.
Thank you for the fast reply. The stoplist did help to filter the word list, but then something strange happened. I loaded the filtered word list to the key word tool preferences and when I generared the keyword list, only the words in the stoplist appeared as keywords. I checked the word list I loaded and none of the words of the stoplist appear there. Even when I closed the program and opened it again, loading the filtered stoplist again resulted in the same strange keyword list.
Forget my first reply. I guess I wasn't loading the reference corpus correctly. I had loaded 10 files for analysis then loaded the same 10 files into the reference corpus. Loading a single file for analysis then loading the rest of the files into the reference corpus fixed the keyword list.
Dear Laurence Anthony
I was trying to calculate keyness myself, because I am very interested in how it works mathematically. I tried to do it according to ucrel.lancs.ac.uk/llwizard.html , but I don't get the same results. So I am very curious what the formula looks like? Can I find it somewhere I have not looked yet. Or do you mind sharing it?
Hi. The Lancs calculator uses a two-term version. You can set AntConc to use this, too, in the settings. The equation is:
# df["LL"] = 2 * (
# df["o11"] * (df["o11"] / df["e11"]).apply(math.log)
# + df["o21"] * (df["o21"] / df["e21"]).apply(math.log)
# + df["o12"] * (df["o12"] / df["e12"]).apply(math.log)
# + df["o22"] * (df["o22"] / df["e22"]).apply(math.log)
# )
Here are some more stats using difference notation:
www.sketchengine.eu/wp-content/uploads/ske-statistics.pdf
@@AntLabJPN Thank you for your answer!
the example I try to get the same result as AntConc is: 29 appearances in 2287 word corpus vs. 87 appearances in a 54'018 word corpus. So, I hope I understand your formula right. I calculate (if you put it in Wolframalpha it looks better):
2*(29*ln(29/(2287*(116/56305)))+87*ln(87/(54018*(116/56305)))+2258*ln(2258/(2287*(56189/56305)))+53931*ln(53931/(54018*(56189/56305))))
which doesnt get me the result AntConc gets when I use (any) Keyword test. AntConc gives 4-term value 52.17 and 2-term value 51.97 when I add the corpora files as shown in the video. Thus, the confusion.
@@ggames33 Hi. How do you know these are the numbers that AntConc is using for your particular data? Can you send a screenshot of the results page to my email addres?
@@AntLabJPN Well I thought I know. But it does not work like that. I sent you some screenshots!
Thanks for such a wonderful thing. Is it possible to load a word file into this cool software?
Hi Bing, yes, you can load in a word list via the Word List Tool preferences menu. Make sure that the format matches the format that AntConc outputs. Good luck!