AntConc (Windows, Macintosh OS X, and Linux)

Build 3.4.1
Laurence Anthony, Ph.D.
Center for English Language Education in Science and Engineering, School of Science and
Engineering, Waseda University, 3-4-1 Okubo, Shinjuku-ku, Tokyo 169-8555, Japan
January 31, 2014

Introduction
AntConc is a freeware, multiplatform tool for carrying out corpus linguistics research and data-driven learning.
It runs on any computer running Microsoft Windows (tested on Win 98/Me/2000/NT, XP, Vista, Win 7),
Macintosh OS X (tested on 10.4.x, 10.5.x, 10.6.x), and Linux (tested on Ubuntu 10, Linux Mint). It is developed
in Perl using ActiveState's PerlApp compiler to generate executables for the different operating systems.

Getting Started (No installation necessary)

Windows
On Windows systems, simply double click the AntConc icon and this will launch the program.
Macintosh OS X
On Macintosh systems, simply double clicAk the AntConc icon and this will launch the program.
Linux
On Linux systems, change the permissions to allow AntConc to be run as an executable file. Next, double click
the AntConc executable and it will launch.
Caution: Do not place user settings files from older versions of AntConc in the same folder as new versions.
This can cause unexpected problems and may even prevent AntConc from starting. It is recommended that you
delete your earlier settings file and export it again from the AntConc file menu.

Overview of Tools
AntConc contains seven tools that can be accessed either by clicking on their 'tabs' in the tool window, or using
the function keys F1 to F7.
Concordance Tool:
This tool shows search results in a 'KWIC' (KeyWord In Context) format. This
allows you to see how words and phrases are commonly used in a corpus of
texts.

Concordance Plot Tool

This tool shows search results plotted as a 'barcode' format. This allows you to
see the position where search results appear in target texts.

File View Tool

This tool shows the text of individual files. This allows you to investigate in more
detail the results generated in other tools of AntConc.

Clusters/N-Grams
The Clusters Tool shows clusters based on the search condition. In effect it
summarizes the results generated in the Concordance Tool or Concordance Plot
Tool. The N-Grams Tool, on the other hand, scans the entire corpus for 'N' (e.g. 1
word, 2 words, ) length clusters. This allows you to find common expressions in
a corpus.
Collocates:
This tool shows the collocates of a search term. This allows you to investigate
non-sequential patterns in language.

Word List:
This tool counts all the words in the corpus and presents them in an ordered list.
This allows you to quickly find which words are the most frequent in a corpus.

Keyword List:
This tool shows the which words are unusually frequent (or infrequent) in the
corpus in comparison with the words in a reference corpus. This allows you to
identify characteristic words in the corpus, for example, as part of a genre or ESP
study.

Concordance Tool
This tool shows search results in a 'KWIC' (KeyWord In Context)
format. This allows you to see how words and phrases are
commonly used in a corpus of texts.
The following steps produce a set of concordance lines from a
corpus and demonstrate the main features of this tool.
1)
Select one or more files for processing from using the 'Open
File(s)...' or 'Open Dir...' options in the 'File' menu. The list of
selected files is shown in the left frame of the main window.
2)
Enter a search term on which to build concordance lines in the search box.
3)
Choose the number of text characters to be outputted on either side of the search term, using the
increase and decrease buttons on the right of the button bar under the "Search Window Size" title.
(default value is 50 characters)
4)
Click on the 'Start' button to start the concordance lines results generation. The concordance generation
can be halted at any time by clicking on the 'Stop' button.
5)
Use the Kwic Sort options to rearrange the concordance lines at three different levels. 0 is the search
word, 1L, 2L... are words to the left of the target word, 1R, 2R... are words to the right of the target word.
6)
Click on the 'Sort' button to start the sorting process.
7)
Move the cursor over the highlighted search term in one of the concordance lines. The cursor will change
to a small hand icon. Clicking on the highlighted search term, will allow you to view the search term hit
as it appears in the original file via the File View Tool (see below).
8)
Click on the Clone Results button to create a copy of the results so that different sets of results can be
compared.
The total number of concordance lines generated (Concordance Hits) is shown at the top of the tool window.
This number will flash with the word "FINISHED" when processing has been completed, and will flash with the
word "NO HITS", if not hits are generated for a particular search term.
Search terms can be specified as being "words" (default) or "character strings" by activating or deactivating the
"Word" search term option. Also, searches can be either case insensitive (default) or case sensitive by
activating or deactivating the "Case" search term option. Searches can also be made using full regular
expressions by activating the "Regex" option. For details on how to use regular expressions, consult one of the
many texts on the subject, e.g., Mastering Regular Expressions (O'Reilly & Associates Press) or type "regular
expressions" in a web search engine to find many sites on the subject (e.g., http://www.regularexpressions.info/quickstart.html). AntConc supports Perl regular expressions.
By clicking on the "Advanced Search" button, more complex searches become possible. The first advanced
search option allows you to import a set of search terms, either by typing them one per line, or by loading in a
list of search terms from a file. Here, each line will be treated as a separate search term. This feature allows
you to use a large set of search terms without having to re-type them each time. The second advanced search
option allows you to define context words and a context window within which the search term(s) must appear.
For example, to search for "student" where it appears at least three words to the left or right of the word
"university," set the search term as "student," the context word as "university," and set the context window as
'From' 3L 'To' 3R.
A number of menu preferences are available with this tool. (See below).

Concordance Plot Tool

This tool shows concordance search results plotted in a 'barcode'
format, with the length of the text normalized to the width of the bar
and each hit shown as a vertical line within the bar. This allows you to
see the position where search results appear in target texts. The tool
also allows you to see which files include the target search term, and
can also be used to identify where the search term hits cluster
together. An example of the use of the Plot Tool is in determining
where specific content words appear in a technical paper, or where
an actor or story character appears during the course of a play or
novel.
The number of hits and length of each text is shown to the right of the barcode plot, and the plot itself can be
enlarged or reduced in size using the Plot Zoom buttons.
If you move the cursor over the highlighted search term in one of the concordance lines, the cursor will change
to a small hand icon. Clicking on the highlighted search term will allow you to view the search term hit as it
appears in the original file via the File View Tool (see below).
Search terms can be specified as being "words" (default) or "character strings", and searches can be case
insensitive (default), case sensitive, or "Regex" based. Advanced searches are also available. For details see
the Concordance Tool explanation.

File View Tool

This tool shows the raw text of individual files. This allows you to
investigate in more detail the results generated in other tools of
AntConc.
The following steps produce a view of the original file and
demonstrate the main features of this tool.
1)
Select a file to view in the Corpus Files list on the left of the
main window.
2)
If a search term has been specified, the search term hits will be
highlighted throughout the text. Search options are the same
as for the Concordance Tool and Concordance Plot Tool.
3)
Use the "Hit Location" buttons to jump to the appropriate hit in the file.
4)
Change the search term and click on the 'Start' button to view other hits in the file.
5) Click on the highlighted text to generate a set of KWIC lines using the highlighted text as the search term.
6)
Click on the Clone Results button to create a copy of the results so that different sets of results can be
compared.
Search terms can be specified as being "words" (default) or "character strings", and searches can be case
insensitive (default), case sensitive, or "Regex" based. Advanced searches are also available. For details see
the Concordance Tool explanation.
The following shortcut is unique to the File View Tool:
CTRL-Click = Jumps to the nearest hit in the window

Clusters/N-Grams Tool
The Clusters Tool
This allows you to search for a word or pattern and group (cluster) the
results together with the words immediately to the left or right of the
search term. In effect it summarizes the results generated in the
Concordance Tool or Concordance Plot Tool.
The clusters can be ordered by frequency, the start or end of the word,
the range of the cluster (number of files in which the cluster appears),
or the probability of the first word in the cluster preceding the
remaining words. All list orderings can also be inverted by activating
the Invert Order option. Also, you can select the minimum and maximum length (number of words) in each
cluster, and the minimum frequency of clusters displayed. It is also possible to select if the search term always
appears on the left (default) or right of the cluster.
Note: In the current version, if more than one word is specified as the search term, only the first word will
appear on the right if the "Search Term on Right" option is selected.)
The following steps produce a set of cluster results and demonstrate the main features of this tool.
1)
Choose the appropriate ordering options (see above for details).
2)
Press the 'Start' button. At any time, the generation of the clusters list can be halted using the 'Stop'
button.
3)
Click on the cluster to generate a set of KWIC lines using the text as the search term.
4)
Click on the Clone Results button to create a copy of the results so that different sets of results can be
compared.
The N-Grams Tool
This allows you to scan the entire corpus for 'N' word clusters (e.g. 1
word, 2 words, ). This allows you to find common expressions in a
corpus. For example, n-grams of size 2 for the sentence "this is a pen"
are 'this is', 'is a' and 'a pen'.
All ordering options available in the Clusters Tool are also available in
the N-grams tool. You can also select the minimum and maximum size
(number of words) in each n-gram, and the minimum frequency and
range of n-grams displayed.
The following steps produce a set of N-gram results and demonstrate
the main features of this tool.
1)
Click on the "N-Grams" option above the search entry box.
2)
Choose the appropriate ordering options.
3)
Press the 'Start' button. At any time, the generation of the n-grams list can be halted using the 'Stop'
button.
4)
Click on the n-gram to generate a set of KWIC lines using the text as the search term.
5)
Click on the Clone Results button to create a copy of the results so that different sets of results can be
compared.
In both the Clusters Tool and N-Grams Tool, search terms can be specified as being "words" (default) or
"character strings", and searches can be case insensitive (default), case sensitive, or "Regex" based.
Advanced searches are also available for the Clusters Tool. For details see the Concordance Tool explanation. A
number of menu preferences are available with this tool. (See below).

Collocates Tool
This tool allows you to search for collocates of a search term. This
allows you to investigate non-sequential patterns in language.
The collocates can be ordered either by total frequency, frequency on
the left or right of the search term, or the start or end of the word.
They can also be ordered by the value of a statistical measure
between the search term and the collocate. The value measures how
'related' the search term and the collocate are. Current possible
statistical measures are listed below. All list orderings can also be
inverted. Also, you can select the span of words to the left and right
of the search term in which to find collocates, and the minimum frequency of collocates displayed. If only a
one-word span is required, for example, to see which words appear directly on the right of the search term,
check the "Same" box, to keep the minimum and maximum span size the same.
Statistical Measures:
(MI) Mutual Information: Using equations described in M. Stubbs, Collocations and Semantic Profiles, Functions
of Language 2, 1 (1995)
(T-Score) T-Score: Using equations described in M. Stubbs, Collocations and Semantic Profiles, Functions of
Language 2, 1 (1995)
The following steps produce a set of collocate results and demonstrate the main features of this tool.
1)
Choose the appropriate ordering options.
2)
Press the 'Start' button. At any time, the generation of the collocates list can be halted using the 'Stop'
button.
3)
Click on one of the collocates to generate a set of KWIC lines using the text as the search term.
4)
Click on the Clone Results button to create a copy of the results so that different sets of results can be
compared.
Search terms can be specified as being "words" (default) or "character strings", and searches can be case
insensitive (default), case sensitive, or "Regex" based. Advanced searches are also available for the
Collocates Tool. For details see the Concordance Tool explanation. A number of menu preferences are available
with this tool. (See below).

Word List Tool

This tool counts all the words in the corpus and presents them in an
ordered list. This allows you to quickly find which words are the most
frequent in a corpus.
The words can be ordered either by frequency or the start or end of
the word, and the ordering can be inverted. The word list can also be
generated in case-insensitive mode, where words in upper and lower
case are treated the same (default) or case-sensitive, where words in
upper and lower case are treated separately.
The following steps produce a word list and demonstrate the main features of this tool.
1)
Choose the appropriate ordering options.
2)
Press the 'Start' button. At any time, the generation of the word list can be halted using the 'Stop' button.
3)
Click on the word to generate a set of KWIC lines using the text as the search term.
4)
Click on the Clone Results button to create a copy of the results so that different sets of results can be
compared.
A number of menu preferences are available with this tool. (See below).

Keyword List
This tool shows the which words are unusually frequent (or
infrequent) in the corpus in comparison with the words in a reference
corpus. This allows you to identify characteristic words in the corpus,
for example, as part of a genre or ESP study.
The following steps produce a keyword list and demonstrate the main
features of this tool.
1)
Select a set of target files.
2)
Go to the 'Preferences' menu and chose the 'Keyword
Preferences' option.
3)
Choose the keyword generation method (a statistical measure) to calculate the 'keyness' of the target
file words. The default setting of Log Likelihood is recommended. When using either Log Likelihood or
Chi-squared as the statistical measure, the following significance values apply (see:
http://ucrel.lancs.ac.uk/llwizard.html):
95th percentile; 5% level; p < 0.05; critical value = 3.84
99th percentile; 1% level; p < 0.01; critical value = 6.63
99.9th percentile; 0.1% level; p < 0.001; critical value = 10.83
99.99th percentile; 0.01% level; p < 0.0001; critical value = 15.13
4)
Choose a threshold for the number of keywords to be displayed.
5)
Choose whether or not to view 'Negative Keywords' (target file words with an unusually low frequency
compared with the frequency in the reference corpus)
6)
Choose one of the reference corpus options. Select "Use raw file(s)" when you will use raw text (.txt) files
to serve as the reference corpus. Select "Use word list(s)" when you will use one of more word lists that
are generated from a reference corpus. The "Use word list(s)" option allows you to generate keywords
even when the original reference corpus is not available. The format for a word list is:
RANK
FREQUENCY
WORD (separated by any type of white space, including spaces and tabs).
1
12838
the
2
11289
a
3
8583
of
...
Note that blank lines and lines beginning with # will be ignored. Also, AntConc will check that the file(s)
are correctly formatted and report any errors.
7)
Load the reference corpus of text (.txt) files, in the same way that the target files are chosen.
8)
The reference corpus directory will be shown (if appropriate), and the list of reference corpus files will
appear at the bottom of the Keyword Preferences option menu.
9)
Click Apply in the Keyword Preferences menu and return to the main Keywords window.
10) Choose suitable options for displaying the list of generated Keywords (in a similar manner to the options
for generating a Word List).
11) Press the 'Start' button. At any time, the generation of the keyword list can be halted using the 'Stop'
button.
12) Click on the keyword to generate a set of KWIC lines using the text as the search term.
13) Click on the Clone Results button to create a copy of the results so that different sets of results can be
compared.
A number of menu preferences are available with this tool. (See below).

MENU OPTIONS
Menu options are divided into three groups, "File", "Global Settings" and "Tool Preferences". The options
available in each group will be described below.
<FILE>
Options here relate to reading files into AntConc and writing files to the hard disk containing data of various
types. There are also options to export all current settings to a file and import user settings from a file. If a user
settings file becomes corrupted for any reason, simply restart the program or use the "Restore Default
Settings" option to return the program to its original state.
<GLOBAL SETTINGS>
Categories here will have an effect on multiple tools in AntConc:
<Character Encoding> AntConc is fully Unicode compliant, meaning that it can handle data in any language,
including all European languages and Asian languages. The character encoding of the data to be read by
AntConc should be specified here. For example, if you are working with data saved in a Western language,
it will usually be encoded in iso-8859-1. On the other hand, Japanese texts are usually encoded in shiftjis.
By specifying the correct encoding, data from all languages can be processed correctly within AntConc. The
default is Unicode UTF-8, which is an international standard designed to display all characters of the
languages of the world in a single encoding. I recommend you use this encoding if you create any corpus.
<Colors> In the Color Settings category, you can edit the colors used to display results and other
information.
<Files> In the File Settings category, you can choose to display the full path of a file or just the name.
<Fonts> In the Font Settings category, you can edit the font types, sizes, and styles used to display file
names, results, and the search term.
<Tags> In the Tag Settings category, you can choose to display or hide any tags that are contained in the
corpus files. You can also choose to search with tags but hide them in the results display. Embedded tags
(e.g. book_NN1), non-embedded tags (e.g. <noun>book</noun>, and header tags can be shown or hidden
by activating or deactivating the option.
<Token (Word) Definition> In the Token (Word) Definition category, you can choose which characters,
numbers and so on will define a "word". For example, in some cases only letters will be considered words,
but at other times, in might be desirable to include numbers, dashes and so on. AntConc is fully Unicode
compliant, meaning that it can handle data in any language, including all European languages and Asian
languages. For this reason, the default option refers to 'letters' in the broadest sense. Letters, for example,
include all English letters (a to z, A to Z) but also all Japanese 'letter' characters. It is also possible to define
your own "token" definition, or append characters to the standard classes.
For more information on the Unicode standards see:
http://www.cs.tut.fi/~jkorpela/unicode/guide.html
http://www.unicode.org/
http://www.unicode.org/Public/5.0.0/ucd/UCD.html
http://www.unicode.org/Public/UNIDATA/PropList.txt
http://www.unicode.org/charts/
<Wildcard Settings> In the Wildcard Settings category, you can edit the default wildcard characters so that
they do not clash with a search entry. For example, the "or" wildcard default character (a 'pipe' character
| ) can be changed to a backslash (/) here. There are special wildcards to deal with whitespace issues.

<TOOL PREFERENCES>
Each tool (with the exception of Concordance Plot Tool and File View Tool) has a preferences category, where
settings can be fine-tuned. All tool preference categories allow you to show or hide the different frames in
which the results are displayed. For example, you can choose to hide the frame showing file names in the
Concordance Tool display window.
<Concordance> In addition to the above, the following settings can be made:
Using the Treat case in sort option causes capitalized words to appear before lower-case words.
Using the Sort by characters instead of words option, it is possible to arrange the results by
CHARACTERS to the left or right of the first letter of the search term. This makes it possible to search
for spelling differences.
Using the Hide search term in KWIC display option, search term can be hidden in the KWIC lines,
allowing instructors to quiz students on possible words to fit the gap.
Using the Put delimiter around hits in KWIC display option (the default), the chosen delimiter
character is added around the hit in the KWIC display. This makes it easier to see the hit and also
eases later processing of the data in a spreadsheet software program.
Use the Delimiter option to select the delimiter character.
Use the Line break replacement option to select a character to replace line breaks with.
<Clusters/N-Grams> In addition to the above, the following settings can be made:
Using the Treat all data as lowercase option (default) causes all words to be transformed to lowercase words. This is useful to get accurate counts of words in certain cases.
Using the Treat case in sort option causes capitalized words to appear before lower-case words.
<Collocates Preferences> In addition to the above, the following settings can be made:
Use the "Selected Collocate Measure" to choose the statistical measure for measuring collocate
strength. Currently, two statistical measures can be used: Mutual Information (MI) and T-Score. See
the tool explanation above for references to the statistics.
Using the Treat all data as lowercase option causes all words to be transformed to lower-case
words. This is useful to get accurate counts of words in certain cases.
Using the Treat case in sort option causes capitalized words to appear before lower-case words.
<Word List Preferences> In addition to the above the following settings can be made:
Using the Treat all data as lowercase option causes all words to be transformed to lower-case
words. This is useful to get accurate counts of words in certain cases.
Using the Treat case in sort option causes capitalized words to appear before lower-case words.
Use the lemma list options to select a lemma list. A 'lemma list' can be loaded from a file, which can
then be used to generate a lemma list instead of a word list. When the lemma list function is used,
the 'lemma word form(s)' column will show the words in the corpus associated with each lemma.
A lemma list can be created by specifying the 'lemma entry' follow by '->' followed by one or more
'words' that should be assigned to the lemma separated by one of more non-tokens. See the
example below:
be->is, are
play->play, plays, playing, played
Note that in the example above, commas and spaces are assumed to be NOT defined as tokens. For
this reason, if the lemma list available on the AntConc webpage is used, a 'dash' needs to be added
to the token (word) definition for the lemma list to be processed correctly as the hyphenated words
are used to the right of the lemma definition.

Using the "Word List Range" option, a wordlist can be generated using all words (Use all words),
or a specific set of words (Use specific words below), or ignoring a certain set of words (Use a
stoplist below). The range of words to be used (or ignored) can be entered directly, or can be
stored in files which are then read by AntConc by pressing the 'Open' button. A combination of
words in a file and words directly entered can also be used.

<Keyword List Preferences> In addition to the above the following settings can be made:
Using the Treat all data as lowercase option causes all words to be transformed to lower-case
words. This is useful to get accurate counts of words in certain cases.
Using the Treat case in sort option causes capitalized words to appear before lower-case words.
Use the "Keyness Generation Method" to choose the statistical measure for measuring keyword
strength. Currently, two statistical measures can be used: Chi-Squared and Log-Likelihood (the
default). The default option for the 'keyness' measure is recommended.
Use the "Threshold Value" option to choose a cut-off for the keyness values generated. The default
option (All values) for the threshold value is recommended.
Use the "Show Negative Keywords" option to view words that are unusually INFREQUENT in the
target corpus compared with the reference corpus.
Use the "Use raw file(s)" option to use raw reference corpus file(s) as the reference corpus.
Use the "Use word list(s)" option to use word list(s) that correspond to a reference corpus. The word
list(s) should be formatted as described in the tool explanation.
Click the "Add Directory" or "Add Files" buttons to select the reference corpus files.
Click the "Swap with Target Files" button to swap the main and reference corpora. Note that this
will only make sense when raw corpus files are being used.
<HELP>
The help menu provides access to this detailed readme file. It also shows some general information about
AntConc, including the current version number and date of release.

SHORTCUTS
Here is a list of Shortcuts that apply to all tools using window panes for results.
CTRL-C = Copies the currently selected text
CTRL-A = Selects all text in the window pane
ALT-A = Selects all text in all window panes showing
Double click = Selects the current word
Triple click = Selects the current line in the window pane
SHIFT-click = Selects continuous lines across all window panes showing
CTRL-click = Selects discontinuous lines across all window panes showing
DELETE = This deletes any selected lines that span across all window panes
INSERT = This keeps any selected lines that span across all window panes, and deletes all others
For any 'spinbox' widgets (e.g. the search term entry box) the 'UP' and 'DOWN' arrow keys on the keyboard can
be used to activate the up and down buttons.

NOTES
Saving Results
Results can be either saved to the clipboard, saved to a text file (..txt) or saved to a new window using
keyboard commands, the appropriate option in the 'File Menu', or by clicking on the "Save Window" button in
each tool, respectively. Also, it is possible to launch multiple clones of AntConc by double clicking on the .exe
file.
Comments/Suggestions/Bug Fixes
All new editions and bug fixes are listed in the revision history below. However, if you find a bug in the program,
or has any suggestions for improving the program, please let me know and I will try to address the issues in a
future version. Indeed, the revisions that have been made are largely due to the comments of users around the
world, for which I am very grateful.
This software is available as 'freeware' (see Legal Matter below), but it is important for my funding to hear
about any successes that people have with the software. Therefore, if you find the software useful, please send
me an e-mail briefly describing how it is being used.

CITING/REFERENCING ANTCONC
Use the following method to cite and reference AntConc according to the APA style guide:
Anthony, L. (YEAR OF RELEASE). AntConc (Version VERSION NUMBER) [Computer Software]. Tokyo, Japan:
Waseda University. Available from http://www.antlab.sci.waseda.ac.jp/
For example if you download AntConc 3.3.2, which was released in 2012, you would cite/reference it as
follows:
Anthony, L. (2012). AntConc (Version 3.3.2) [Computer Software]. Tokyo, Japan: Waseda University. Available
from http://www.antlab.sci.waseda.ac.jp/
Note that the APA instructions are not entirely clear about citing software, and it is debatable whether or not
the "Available from ..." statement is needed. See here for more details:
http://owl.english.purdue.edu/owl/resource/560/10/

ACKNOWLEDGEMENTS
I would like to say thank you to the users of AntConc who have taken the trouble to e-mail me with feedback
on the software and suggestions for improvements and/or changes.
The development of AntConc has been supported by a Japan Society for Promotion of Science (JSPS) Grant-inAid for Scientific Research (C): No. 23501115, a Japan Society for Promotion of Science (JSPS) Grant-in-Aid for
Young Scientists (B): No. 18700658, a Japan Society for Promotion of Science (JSPS) Grant-in-Aid for Young
Scientists (B): No. 16700573, and a WASEDA University Grant for Special Research Projects: No. 2004B-861.

LEGAL MATTER
AntConc can be used freely for individual use for non-profit research purposes, and freely distributed on the
condition that this read me file is attached in an unaltered state. If the software is planned to be used in a
group environment, you are required to inform me how the software is to be used, and I will then determine if
you can have permission to use it. The software comes on an 'as is' basis, and the author will accept no liability
for any damage that may result from using the software.

KNOWN ISSUES
The current version of the AntConc does not run well on Macintosh OS X 10.9 (Mavericks). When run on
Mavericks, unpredictable refresh problems will appear. This is a bug in the programming language that I use. I
am waiting for ActiveState to release a bug fix the programming language and will immediately update the
software when it becomes available.

REVISION HISTORY
3.4.1
This is a minor upgrade addressing several issues noted with the release of 3.4.0.
Bug fixes:
1. On Macintosh and Linux systems, I have heard reports that the advanced search window sometimes
does not appear correctly when the "Advanced Search" button is clicked. To address this, I have set the
window size explicitly (instead of calculating the size from the internal widgets) and forced a refresh of
the screen. This addresses the problem on the systems that I have checked (Macintosh 10.6.8, 10.7.2
and Linux Mint 10). I hope it fixes the problem on other systems. This is difficult to test so please let me
know if you continue to find a problem.
2. If no search term is inputted into the search box, an error that "No search term has been entered" is
reported even when terms have been entered using a file based search. This is now fixed.
3. In 3.4.0, when the program was first started, the collocates display would incorrectly show the span to
be 5L to 5R, when it was in fact 1L to 1R. Although changing the span setting would rectify the problem,
on first load, the results would appear to be incorrect. In this new version, the display continues to
show the span to be 5L to 5R, but the results now match the setting.
4. In all versions of AntConc, the results produced when using context horizons (e.g. in the advanced
search of the Concordance tool) are dependent on the window span (e.g. set in the main windows of
the Concordance Tool). If the span of the horizons goes beyond the window span, spurious results can
be generated. In 3.4.0, due to the nature of results being stored in a set of columns in database, if the
left horizon span went beyond the number of columns estimated by the window span, no hits would
be reported. In some ways, this was a good thing as it showed that something is wrong with the span
settings. However, the results no longer matched those in previous versions of AntConc that simple
ignored the mismatch and produced results for the largest horizon span possible. I have now edited the
storage routine so that 3.4.1 also ignores the mismatch and creates results for the largest column size
possible. I am still unsure if this is best approach to take. I am considering if a proper error or warning
should be reported when a mismatch between the context horizon span and the window span arises.
5. Although not strictly a bug, due to changes in the way the Clusters/N-Grams tool treated line breaks
introduced in 3.4.0, results stopped matching those from earlier versions of AntConc (e.g. 3.3.5). In this
version, I have introduced an option in the Clusters/N-Grams preferences to either replace line breaks
with a space or other character (as happens with the Concordance Tool - the default), or leave line
breaks as is. If line breaks are left as is, the results will match those from previous versions of the
software.
6. Although not strictly a bug, on startup with a user settings file, AntConc would not remember the size
or position of the main window and instead show it at the default size and position. This has been fixed.
7. When AntConc is launched on a Windows system with a magnified view (e.g. 125%), some of the
settings options were hidden from view. To compensate for this feature of Windows, I have redesigned
the settings windows so that all options are viewable when first launched in magnified view at 125%.
Viewing at 150% is likely to still hide certain options. I have also allowed the settings windows to be
resized so that even at 150%, it should still be possible to resize the settings window to see the options.
8. The way the program reads in files has been changed to avoid a strange bug that occurred when
reading in some Unicode files that resulted in the last character on a line being misread on very rare

occasions. The way file names are displayed in the program has also been changed to avoid a
potentially similar problem (although I have heard of no reports on this to date).
9. The "Range" heading of cloned Clusters/N-Grams tool results did not show correctly. This has been
fixed.
10. The "Stat" heading of the cloned Collocates tool results was shown as "Prob". This now appears
correctly.
3.4.0
This is a semi-major upgrade fixing a few small bugs and adding several important features to improve the
output display of the Concordancer and Plot tools.
New Features:
1. The Concordancer tool output now correctly centers the search terms regardless of language. This is
especially useful for Asian languages such as Japanese, Chinese, and Korean that have two-byte
character widths. Note that this is achieved by using two columns in the 'KWIC' pane. If you click CTRLA in the pane, only one of these hidden columns will be selected. Use ALT-A to select all columns.
2. Line breaks in the Concordance Tool output can now be replaced by a user defined character (default
being a single space). Again, this is especially useful for Asian languages because traditional
replacement by a single space might inappropriately split words that cross line-break boundaries.
3. The Collocates Tool no longer includes the center word in the collocation table. This results in more
meaningful collocate values.
4. Not exactly a feature, but Concordance Tool searches are now only centered on completion rather
than after finding the first line. This is generally a more pleasant experience.
5. Whitespace options in the wildcards menu have now been extended.
6. A database now runs in the background and is used to store KWIC and Plot concordance results. This
alleviates out of memory problems with earlier versions and also allows much faster sorting of the
KWIC. The new database model also improves the performance of general KWIC and Plot searches.
7. In the Advanced Search, if you add multiple context words, AntConc will look for *any* of these in
surrounding context of the search term rather than *all* of them (as in earlier versions). This change
has been implemented as a result of many requests for this function. I hope it does not inconvenience
too many people.
8. In the Collocates tool, if you click on a word, if will search for that word in the Concordance Tool but
restrict results to only those where the original search term is in the neighborhood of the collocate as
set by the Collocates tool. Again, this change has been implemented as a result of many requests for
this function.
9. Colors for the highlight and ordered words in the KWIC display have been changed to be a little easier
to see.
10. The font size for the main results is now a little larger.
11. Mismatches in results across tools (e.g. different frequency counts for the Concordance, Plot, File View
and Clusters/N-grams tools) due to the different ways in which tools handle line breaks and spacing
have been fixed. In particular, by changing the default processing to replace line breaks with spaces,
the frequencies reported in the Clusters/N-grams tool will now match the Concordance tool hits.
12. All searches are now performed to find the maximum number of hits for a particular search. This
means that a search for "aa" in a string "aaa" will now produce two hits instead of one (as in earlier
versions of AntConc). For most 'word' based searches, very few differences will be noticed. However,
results can be quite strikingly different for string-based searches using wildcards. The new way of doing
searches is intuitively more sensible than the previous search approach. However, many other
software programs adopt the previous approach so you should expect some differences when
comparing results.
13. Generic line break characters representing line breaks on different systems (e.g. \r\n for Windows and
\n for Linux/OS X) are now used for all file processing.

Bug fixes:
1. The Collocates Tool sometimes use to report values of "-1" (when the only collocates of the center
word was itself) and "-2" (when the collocate value was not able to be calculated due to an out-of-sync
word list). The first case ("-1") no longer exists (see feature 3 above). In the second case ("-2"), the
value has now been replaced with "0" to avoid confusion.
2. In the Word List Tool, Alt-A now correctly selects all the panes including the Lemma list.
3. When no corpus file names were selected in the left panel, sometimes the Close file menu option
would delete the first file name in the list instead of reporting that no file was selected. The has been
fixed.
4. After changing the size of the Search Entry font option in the global settings, the new font size would
not be reflected in the sample display. This is now fixed.
5. If the "invert sort" option was activated in any tool with this feature, an "Array XXX" error was
produced in the results window. I apologize for this very inconvenient bug.
6. Several other minor bugs including one that sometimes led to tags not being hidden properly after
applying tag settings have been fixed.
3.3.5
This is a minor upgrade fixing a few small bugs. One very minor new features have been added.
New Features:
1. The "Order by Stat" sort option is now set as the default option in the Collocates Tool.
Bug fixes:
1. Fixes a problem that caused the window span controller values to become out of sync if the "from"
controller changed the "to" controller or vice versa.
2. Fixes a problem that caused the window 'same' checkbox controller to not work properly.
3. Fixes a problem that caused the scrollbars to not show correctly in the cloned view of the
Concordancer Tool window.
4. Fixes a problem that caused the background color of the global-settings->character encoding options
to appear white on a Mac OS X system. This made the options impossible to read because the OS X
native foreground color is also set to white, which is different from that on Windows and Linux
computers.
3.3.4
This is a minor upgrade fixing a single bug. No new features have been added.
Bug fixes:
1. Sort by word end in the Word List tool produced a list in reverse order. This is now fixed.
3.3.3
This is a minor upgrade fixing a couple of bugs. No new features have been added.
Bug fixes:
1. A tab delimiter character incorrectly appeared in the KWIC concordance display. This is now fixed.
2. If multi-character delimiters were inputted, the highlighting became misaligned. This is also fixed.
3. A bug causing the software to keeping reopening after clicking the close button (top right of screen on
Windows) was fixed. The problem only appeared after importing user settings files or reloading the
default settings.
3.3.2
This is a semi-major upgrade that addresses an important bug in version 3.3.1. I've also added a few small but
useful features.
New Features:

4. Performance has been greatly improved (> 10 times) for all but the Concordance and Plot tools. I hope
to improve the performance of these tools from now.
5. It is now possible to append user-defined characters to the standard Unicode character classes.
6. Users can now choose to ignore or not ignore white space in the search box. The default is set to ignore
white space
7. Users can now choose which delimiter character to use in the Concordance tool.
8. The "treat all data as lowercase" option is now set as default on all tools except the Concordance and
Plot tool. I hope this is a much more sensible default setting.
9. Files created on different platforms (Windows, OS X, Linux) can have different end-of-line characters.
These are now treated correctly regardless of which platform AntConc is launched on.
Bug fixes:
1. Some user token definitions caused the output to display strange results that appeared to ignore any
word boundaries. This problem is now fixed.
3.3.1
This is a minor upgrade that addresses a few small bugs version 3.3.0.
Bug fixes:
1. All windows now resize correctly to show widgets even when the screen display setting on Microsoft
Windows is set to 125%. In 3.3.0, an enlarged display caused some setting areas to be cut.
2. Windows are now centered correctly regardless of the screen display setting on Microsoft Windows. Also,
there is no flicker of windows moving from the top left of the screen to the center as they are positioned.
3. The checkbox for Level 1 sorting now includes a greyed out check mark.
4. Underline marks that appeared under list items (e.g. corpus files in the main window) have now been
removed.
5. A few bold headers have now been corrected to normal typeface.
6. Word and case checkbuttons in the main window are now correctly disabled if the Regex option is marked
as active in a user settings file.
3.3.0
This is a major upgrade introducing many new features and addressing bugs that appeared in previous 3.2.x
versions.
New Features:
1. Now AntConc runs completely natively on all platforms. This is especially important for Macintosh OS X
users as it means no X11 installation is required. Also, the interface is considerably improved over
previous 3.2.x versions most spectacularly on Macintosh OS X but also on Windows and Linux systems.
2. The range information for a word has now been added to the Clusters Tool. This allows lexical
bundles of a certain minimum range to be generated.
3. The full path of a corpus file is now shown by hovering the cursor over the file name in the main
window.
4. Loading of lemma lists is now greatly simplified.
5. Loading of reference corpus lists are now greatly simplified.
6. When loading reference corpus lists, checks are now made to see if the format is correct.
7. The interface has been simplified a little by removing little-used buttons, in particular, the Reset
button and Exit button on each tool tab. Also the ambiguous Save Window button has now been
renamed Close Results to more accurately reflect its action.
8. Hit and frequency counts now appear above the results window in all tools.
9. The Clusters tool has now been renamed Clusters/N-Grams to more accurately reflect its functions.
10. The actions associated with spinbox arrows have now been reversed. This is perhaps a more natural
way for them to work.
11. The term Language Encoding used in 3.2.x versions has now been corrected to Character Encoding.

12. Various default settings have been changed to reflect modern corpus linguists preferences. In
particular, the default character encoding has been changed from Latin 1 to UTF-8.
13. This user file has now been updated to be a little easier to read and understand.
14. After opening or saving files, AntConc will now remember the folder location.
Bug fixes:
1. Selection highlighting of corpus files now accurately reflects which files can be viewed in the File View
tool.
2. AntConc now correctly remembers the positions of window panes after global and tool preferences
have been accessed.
3. AntConc now correctly remembers which was the last preference window accessed by the user.
4. Numerous interface oddities associated with 3.2.x versions of AntConc should now be fixed with the
introduction of a native interface for all platforms (see New Features 1 above).
5. File names with a different character encoding the file contents will now appear correctly.
3.2.4
This is a minor upgrade fixing two problems (one only applicable to OS X). No new features have been added.
Bug fixes:
1. Modified the OS X version so that lemma list files will load correctly regardless of the OS version. The
bug was more general and possibly caused errors when using any single file open file dialog box, e.g.,
importing settings files, search list files, and word list files. I only heard one or two reports of problems
on OS X, so perhaps the problem was introduced with upgrades to recent versions of OS X. It is difficult
to tell.
2. Fixed a bug that caused the total number of lemma entries to not be displayed properly.
3.2.3
This is a major upgrade introducing several new features and addressing bugs that appeared in version 3.2.2.1.
New Features:
1. Massively simplified the engine used for sorting and displaying results. This will allow for improved
performance later.
2. Introduced a Concordance preference setting allowing tab spaces to be added around the search hit in
the KWIC concordance view. Introducing this means that the toggle shortcut key x to show/hide the
KWIC search term has had to be removed. This function can now be accessed via the (fixed) menu
option. I also hope to reintroduce this toggle shortcut key later.
3. Introduced the option to use word list(s) of reference corpus files instead of directly processing the raw
files. This allows fast generation of keywords and also allows users to generate keyword lists even
when the reference corpus is not available (but a word list is). See the Keyword List section for further
details.
4. Introduced feature to output the counts of types and tokens for each tool when saving results as a text
file.
5. Changed some of the default settings in response to user feedback:
6. Concordance Tool sort level s are set as 1R, 2R, 3R.
7. The Collocates Tool Stats measure is now displayed and calculated.
Bug fixes:
1. The splash screen that appeared when a user settings file was added has been removed. Now, the
words User Settings appears next to the title of the software at the top of main window, whenever a
user settings file is used.
2. The layout of the main window has been improved so that the main work area is maximized for all sizes
of main window.
3. Unicode Marks can now be added/removed in the token definition settings

4. Occasionally, the last bar of the progress bar remained blank at the end of processing. This has now
been fixed.
5. Under tag settings, the Hide Header Tags option will now work for headers than span across multiple
lines.
6. The main window and all pop-up windows now appear centered in the display.
7. When viewing the global settings preferences, it was possible to accidentally activate the main window
by clicking on it. This has been fixed.
8. The list of files in the left pane was often too narrow to view easily. This has been fixed by introducing a
two-pane approach, with the left pane showing the files, and the right pane showing the results and
controls. The width of both panes can be adjusted with the default size being the minimum size.
9. It was possible to adjust the main window to be smaller than the default size of 800px by 600px. This
caused the positioning of some widgets to become misaligned. Now, 800px by 600px has been set as
the minimum size of the main window.
10. Various problems experienced when cutting, copying, and pasting text to/from the Search boxes in
each tool have been addressed by removing the Spinbox widget type and replacing this with a
standard Entry widget. The history of previous searches can be now accessed using the Up
(previous) and Down (next) arrow keys.
11. After moving the cursor over the corpus files list while using the File View Tool, it should change into a
Pointing Finger cursor. However, this cursor shape was returned incorrectly back to a standard Edit
cursor when one of the files was selected. This has now been fixed.
12. Changed foreground to white when AntConc reports that the KWIC analysis has finished or has no
results. This should be easier to read.
13. Fixed a bug that caused the Concordance preference option Hide search term in KWIC display to only
show the hit. (The opposite of the stated function!). This is now fixed.
14. Fixed some typos in this Read Me document.
15. Fixed a major bug that caused a warning to always appear when using the T-Score statistical measure
with the Collocates Tool. Note that the warning could be ignored as the results would always be correct.
3.2.2.1
This is a minor upgrade addressing bugs that appeared in version 3.2.2.
Bug fixes:
1. The legacy character encodings for Japanese, Chinese, Taiwanese, and Korean were inadvertently left
out of the compiled executable for version 3.2.2 meaning that files encoded in these languages would
not open correctly in AntConc. These have now been added.
2. In the Mac OS X version, the compiled version did not include several important components which are
essential for the X11 graphical toolkit to run correctly. These have been added.

3.2.2
This is a minor upgrade addressing several minor bugs that appeared in version 3.2.1. It is also the first version
of AntConc to be complied with Perl 5.10.
Bug fixes:
1. Fixed a bug that caused back references to not work correctly when the 'Regex' search option was
selected. This now works correctly. However, note that the first back reference should be 2 and not 1
(due to an implicit back reference 1 that holds the entire search result.
2. Fixed error message when the sort by Stat option was selected.
3.2.1

This is a minor upgrade addressing several bugs that appeared in version 3.2.0, as well as introducing a few
new features requested by users.
New Features:
1. Better display of long lists of fonts and font sizes in the global options/font menu. Now the lists appear
as an easy to navigate list with attached scrollbar.
2. When results windows are saved, the cloned windows now display summary results information.
3. Word range lists can now be used as lemma range lists.
4. New feature allowing tagged data to be searched while remaining hidden. See the tag settings
preferences. Pressing CONTROL and the START button (or the ENTER button if the search entry box has
the focus) temporarily disables the new feature, allowing the user to switch easily between a 'nontagged' or 'tagged' display.
5. New options in the tag settings preferences that allow embedded and non-embedded tags to be shown,
ignored, or hidden. This enables data of the form of the BROWN and BNC corpora to be processed
easily.
6. Improved the updating of the progress bar display, which may also improve the speed of processing in
some cases.
7. Improved the images used for icons within the program.
Bug fixes:
1. Fixed bug that caused user defined token definitions containing special regular expression characters
from not working properly
2. Fixed bug that caused "Treat all data as lowercase" option to ignore the wordlist range and lemma lists
3. Fixed bug that caused the lemma list "Load" button to not ignore the currently opened file if a new file
dialog was opened and then "Cancel" was pressed
4. Fixed bug that caused file searches to not work if the search entry box was blank.
3.2.0
This is a major upgrade with a completely redesigned interface, several new features, and several bug fixes.
The new interface follows the basic design used in previous versions, although users should find it 'cleaner' and
more intuitive. In particular, all global and tool menu settings have been combined into two groups, where all
the related settings can be accessed and adjusted within the same window. This should dramatically improve
the usability. All tools now have access to the search engine (including the Word List Tool and Keyword List
Tool) and there is also a new advanced search window that can be used to perform list (file) searches, and
searches within a particular context. Due to the nature of the changes, this version will not be compatible with
the settings files for previous versions. Another huge change is that this version will run on Macintosh OS X
systems.
New Features:
1. Completely redesigned interface
2. Added search and advanced search features to all tools (including the Word List Tool and Keyword
List Tool).
3. Created new list (file) search available in all tools.
4. Created new context search option in all tools except the Word List Tool and Keyword List Tool where
it has no meaning.
5. Busy cursors are used to indicate when very long sorting operations are being carried out (e.g. when
sorting large N-gram list results).
6. Case options affecting whether or not data is converted to lower case are now more intuitive. For
example, the 'Case' option in the main window now only affects the operation of the search itself and
has no impact on the data under observation. Data can be treated as lowercase (for example in Word

7.
8.
9.
10.

list tool) by choosing the 'Treat all data as lowercase' under the appropriate category in the 'Tool
Preferences' menu.
The number of corpus files (and reference corpus files) being analyzed is now displayed.
Correct some mistakes in this readme file
My name has been removed from the top of the main window! However, please remember that my
name is Laurence with a U if you are ever citing me in your research papers!!
Now works with Macintosh OSX

Bug fixes:
1. The program no longer crashes when the 'All Values' option is chosen as the threshold value in
Keywords.
2. Negative keywords are now highlighted correctly when the 'Show Negative Keywords' option is
selected.
3. The KWIC lines are now aligned correctly even when the hit appears near the very start or end of a file.
4. Collocates frequency values are now correctly calculated even when the span extends further left than
the start of the file
5. The action of the 'one word only' wildcard is now more intuitive.
6. Some operations (e.g. creating a word list) now do not crash after restoring the default settings and
then performing an operation.
Bug fixes (since beta1 version):
1. The program now (correctly) only shows files that generate hits in the Concordance Plot tool.
2. The sort function in the Keywords Tool now works correctly. In previous versions, even when the
'Frequency' option was selected, the sort would be based on Keyness. Also, in some cases inverted
sorting did not work.
Bug fixes (since beta2 version):
1. The program now (correctly) hides the various Concordance Tool panes depending on the chosen
Display Options. In the earlier beta versions, the options were ignored.
2. The default file type to use when opening directories now works correctly. In previous beta versions,
after hitting the apply button, the default file type reverted back to the .txt type.
3. Fixed a bug that prevented the n-grams option in the Clusters Tool from working when the search
term entry box was empty.
Bug fixes (since beta3 version):
1. Fixed bug that caused the program to not be able to open files with non-English names correctly if the
full-pathname option was selected. There are potentially many problems with non-English filenames,
so I recommend that users use English filenames for their corpus files, and also save them under a
pathname which only contains English characters.
2. Fixed bug that caused the 'OR' wildcard to not work correctly if a character other that '|' was user
defined.
New Features (since beta4 version):
1. Made some small changes so that the program could be more easily ported to Macintosh and Linux
platforms.
3.1.303
This is a very minor upgrade with just one change:
Bug fixes:

1. Corrected problem that caused the No. of Hits to not be indicated correctly in the Concordance Plot
Tool display when more than one corpus file was being used.
3.1.302
This is a very minor upgrade with the following changes:
Bug fixes:
1. Corrected problem which caused the program to not launch when the path of the default temporary
folder on the system contained non-English characters.
2. (Linux only): Corrected problem that caused the Open Dir menu option to not work correctly.
3. (Linux only): Corrected problem that caused font selections to not work correctly.
New Features: Improved speed and memory handling when calculating collocates. Over 10 times faster than in
previous versions (including version 3.1.3).
3.1.3
This is a minor upgrade containing an important bug fix that prevented files with non-ASCII filenames being
used. There are also some major performance improvements. For example, n-grams will now be processed
over 10 times faster on small corpora and many more times faster on larger corpora. A list of all important
changes is below:
New Features:
The history feature for search term entries has been changed. I have heard two reports of the 3.1.2 version not
starting on computers. Hopefully, this change will allow the program to start correctly on all machines.
The performance of tools such as Collocates, Clusters and N-grams, has been significantly improved. (Over 10
times faster on small corpora and many more times faster on larger corpora.)
The Open Dir option now open files in all sub-directories too.
The program will automatically look for a user defined settings file named "antconc_settings.ant" in the
directory where the program is saved. If this file is found, this settings file will be used instead of the default
settings. Also a splash window will be displayed when a user settings file is found. If no file is found, the default
settings will be used. In this case, no splash file will be shown. This feature allows users to save their setting
preferences and use them again without having to load the preferences each time.
Bug Fixes:
4. Files with non-ASCII file names were incorrectly processed preventing them being used. Introduced in
version 3.1.2. Fixed.
5. In the N-grams tool, non-word units (e.g. spaces) at the beginnings of lines were treated as words. This
caused some n-grams of n-1 size to also appear in the results list. Fixed
6. When "ALL" was selected as the file type option in Open Dir, sub-directories were also displayed even
though these could not be opened or processed. Fixed
7. In some tools, non-ASCII filenames were sometimes displayed incorrectly, even after the correct
encoding was chosen. Fixed.
8. If 'Exit' was chosen from the File menu options, the program exited without a warning. Fixed.
9. The "Add" button for the 'Add Word' option was accidentally deleted from the Word List Preferences
menu. This meant that words could only be added by hitting the return button. Fixed.
Changes:
10. The default file type for Open Dir has been changed to .txt
11. The 'Directory' displays in both the main window and keywords list preferences window have been
deleted. As directories and sub-directories can now be used, this feature has become redundant and
perhaps confusing.

3.1.2
This is a semi-major upgrade containing a new Lemma List tool, and numerous bug fixes and interface
improvements. A list of all major changes is below:
1. Binding to plot canvas lines to allow jumping to hit in file (same as in concordance view).
2. Binding to file list allowing the user to simply click on the file name when using the File View tool to
view the file
3. History feature for all entries (use up and down arrows on keyboard)
4. Much faster concordancer processing (especially sorting... up to 10 times faster)
5. Ability to clear content of tools (Clear Tool, Clear All Tools, Clear All Tools and Files)
6. Ability to save list of all loaded files in settings file
7. Redesigned "one or more words" and "any one word" wildcards to act more sensibly. Now, the
wildcards incorporate any non-tokens between possible words. Therefore, use "is#dog" and
"is@@dog" etc. to search for the hit "is a dog" in the sentence "This is a dog."
8. New Lemma tool allowing all lemmas of a word list word to be displayed. Note: A lemma list is required.
9. Added bindings to Wordlist Range list to allow words to be deleted (<Delete> button) or kept (<Insert>
button) as with other tools.
10. Swap button to switch the main and reference corpora when doing keyword analyses. (Accessible from
the Keyword preferences Menu).
11. Fixed serious bug that caused the program to freeze or crash when an incorrectly formed regular
expression was used as a search term.
12. Fixed serious bug that caused some word, keyword and font encodings to not be loaded correctly from
the user settings files
13. Fixed bug that caused the 'Reset' buttons to not work correctly.
14. Fixed bug that caused 'user defined token definitions' to not be loaded correctly.

3.1.1
This is a minor upgrade although it contains a new T-Score stats feature, and has a new editing feature,
allowing you to select and then delete, or keep certain results lines. A list of all major changes is below:
1. Added T-Score to stats measure in the Collocates Tool.
2. Changes the Collocates Tool menu to allow one a several statistical measure to be chosen.
3. Adding a feature so that any selected results lines can be either deleted or kept (after deleting all
others).
4. The tag settings options in the File Settings menu have now been moved to a separate menu.
5. Labels in the software (in particular those related to language encodings) have been slightly altered to
make them clearer.
6. Improved the warnings given when the Collocates Tool is not in sync with the Word List Tool, during
the calculation of collocate significance.
7. Changed the internal workings to allow LINUX and WINDOWS ports of AntConc to be easily created
from the same source code. This should allow future versions to be released at the same time for both
operating systems.
8. Fixed a bug that caused the Transition probabilities calculation to not initialize correctly. This meant
that although the first calculation was correct, all later calculations would be give false measurements.
3.1
AntConc 3.1 is a major, major upgrade. I was very tempted to name this AntConc 4.0, but in the end chose to
keep it in the 3s. To list the improvements and changes will take up many pages, so only the major differences
are listed here in no particular order. Of course, with so many changes, there will inevitably be new bugs that
have crept into the program. However, I hope you will find that this release is an improvement over previous
releases.

1.
2.
3.
4.
5.
6.

7.
8.

9.

10.
11.
12.

13.
14.
15.
16.
17.

Implementation of new Collocates Tool.

Total reworking of the programming underneath all tools for performance enhancements.
Adding of "sort" features to many tools, enabling quick and efficient re-ordering of results information.
Sensible treatment of word case in all tools. Now searches are truly case sensitive or insensitive, and
results can be displayed according the
case setting.
All settings can now be imported or exported to a file. This allows users to easily upload their
preferred settings, avoiding the need to constantly make changes to the defaults. This is a HUGE
improvement (I think!).
The "Open Dir" option has been restored. File types that will be read into AntConc via the "Open Dir"
option can now be specified in the "File Settings" menu.
AntConc is now FULLY Unicode compliant. It should work with any language in the world. I am very
interested to hear how AntConc works in different languages. Please let me know. Token definitions
can be made using all Unicode character classes, or a user defined token definition can be made.
However, I would advise against using the user defined token definition, as it is so easy to overlook
possible characters that might need to be processed.
As AntConc is now fully Unicode compliant, all possible encodings of characters are listed under the
Encodings menu option. For English and all other Western Language, the default option, iso-8859-1
should work fine. Note that many Windows systems actually save files in the cp-1252 encoding which
resembles iso-8859-1 but is a little different, just to confuse people! Users with Japanese texts will
probably find shifts to be the best option.
Some users have said they could not open the font selection windows. This problem has been fixed.
Many new ordering options have been added, for example, from word ends.
The overall design of the interface has been improved, allowing for new options to be easily added
without cluttering the display. Also most menu options now use simple checkboxes instead of the
confusing "yes", "no" radio boxes used in previous versions.
Some of the tool names have been changed. In particular "lexical bundles" are now called "N-Grams",
which is a more familiar term.
The Word Stems feature has been removed. This only worked with English texts, and the results were
of questionable value. If a user would like to see the feature in a future version, please let me know.
The size of the program has increased by around 1MB. This is so that all the fonts and encodings for
true internationalization of AntConc can be included.
This read me file has been largely rewritten. I hope there are fewer spelling mistakes this time!
Many, many small bugs have been fixed, mainly related to the way the interface responded to user
actions.

3.0.1
A few bugs in 3.0 have been corrected. Also, the View Files tool has been redesigned to work much, much
faster. To do this, the program now does no processing at the end of lines. Therefore, ambiguous line endings
will stay ambiguous!
1. Improved performance of View Files tool
2. Corrected bug which caused the first file in Keywords to appear even after the list of files is deleted
from the system.
3. Added feature to allow files to be added to the system from different folders. Now, if a file is opened
into the system it will be added to the current list without deleting the previous list files. The same
applied to files added in the Keywords List preference menu.
3.0
There have been so many changes that it is almost impossible to list them. Here are some of the more obvious
differences between AntConc3.0 and previous versions.

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.

Wildcard implementation
Word stem implementation
Save current window feature added to all tools (except 'Concordance Search Term Plot')
Three levels of sort implemented
Complete redesign and implementation of the 'View Files' tool, making it generate results much faster.
(But it is still too slow).
Hyperlinks added to the results of all tools.
Rearrangement of the menu options
Redesign of the file list window, allowing one or more files to be closed.
Redesign of the main results windows, placing information in different window 'panes'. Each pane can
be resized, or hidden.
Redesign of the data selection methods. The selection methods now comply with most other software
products.
Complete redesign of many sub-routines enabling quicker processing.
Many, many bugs found and corrected. Please tell me if you find a bug because I WILL correct it.

2.6.0
1. Corrected a bug (introduced in version 2.5.3) which caused cluster lists to be not be displayed
alphabetically.
2. Corrected a bug (introduced in version 2.5.3) that prevented the 2nd sort color to be properly updated
from the settings menu. Thanks to a user for pointing this out to me.
3. Corrected a bug (introduced in version 2.5.3) that prevented files on a Japanese system to be open
correctly. This bug is related to the new implementation of the Perl programming language. Please
point out any bugs due to the new version of Perl, as they are very difficult for one person to discover.
4. Added an option to either show or ignore tags embedded in files (which is useful when processing
HTML or XML files).
5. Added a 'lexical bundles' option to the Word Clusters tool that generates "word n-grams".
6. Deleted the "Open Directory" option, as it prevented files other than .txt from being uploaded.
7. Added menu setting options to allow the font for searches and results to be changed.
8. Added menu setting options to show Japanese fonts (in SHIFT-JIS encoding) to be displayed properly.
9. Reduced the overall size of the program by approx. 1 MB.
10. A number of small bugs were corrected.
2.5.4
1. Corrected a bug (introduced in version 2.5.3) which caused word lists and keyword lists to be not be
displayed alphabetically.
2.5.3
1. Adjusted the positioning and size of application widgets so that the application will be displayed
properly on an 800 x 600 or higher resolution monitor.
2. Corrected several spelling errors in this read me file.
3. Adjusted the parameter settings for the Word Clusters tool.
4. Re-labeled several buttons so that they display properly on low resolution monitors.
2.5.2
1. Enabled a new 'Clusters' tool, for generating word clusters centered on a target search term.
2. Re-ordered Open File and Open Dir options in the File Menu. I think file navigation combined with
'select all files' <CTRL A> is easier than directory navigation,
3. Fixed a couple of small, insignificant bugs.
4. Fixed some spelling errors in this read me file.

2.5.1
1. Enabled a new feature, whereby clicking on the search term hit in any concordance line, allows the
user to view the hit in the original file via the View Files tool.
2. Fixed a small bug in the View Files tool, which caused the searches to ignore the 'Case' setting.
3. Improved the performance when generating View Files by caching already processed files
2.5.0
A fairly major upgrade since 2.4.1
Here is a list of changes that have been made
1. Bug fix. When viewing files, and locating the next or previous hit, if the target file was changed and the
hit number did not exist in the new file, the program would crash. This problem has been fixed.
2. Extension: In the view file window, hits would only appear if they occurred on a single line in the
original file. This would result in different numbers of hits depending on if the search was made in the
concordance window or the view files window. The view file processing has been completely revised
enabling view file searching to correspond exactly with that used in the concordance window.
Unfortunately, this has resulted in a small loss in performance when generating the highlighting in the
view file. Also, clicking in the View File window now allows the user to immediately jump to the
nearest hit.
3. The ability to show or not show full path names to files has been added as a system preference
4. The ability to show or not show file names in the concordance window has been added as a system
preference.
5. The ability to set a wordlist 'range' has been added as a system preference.
6. Highlighting in the View Files tool has been changed to make the hits easier to see.
7. Pop-up windows that showed how many concordance hits were generated, and that reported when
no hits were found have been removed. Instead, the status of the concordance hit processing is now
shown in the top right of the main window.
8. Many small bugs relating to how the various tool displays are updated after preference changes are
made have been corrected.
9. Processing that blocks user events (such as mouse clicks etc.) have been reduced.
10. The internal workings of the program have been re-written so that problems and future additions can
be easily handled.
11. The general layout of the README file has be re-designed.
2.4.1
New since AntConc2.4.1 is the ability to choose whether or not to view 'Negative Keywords'. These are words
in the target file that have an unusually 'low' frequency. In previous versions of AntConc, Negative Keywords
were not distinguished from Keywords. However, now they are treated separately, and if the user chooses to
display them, they appear after the Keywords, with a highlight color.
2.4.0
A major upgrade since 2.3.0
1. First, progress indicators were added to 'pages' of AntConc.
2. Second, a new file view feature was added to view target files in their original state.
3. Third, a keyword generation feature has been added using log-likelihood and chi-squared methods.
4. Finally several bugs were found, in particular, bugs centered on the wordlist generation feature. This
feature of the software should work much quicker now. Also, the user can interrupt the processing of
files in any 'page' of the software.
2.3.0

A major upgrade since 2.2.3

1. First, the ability to view concordance search results as a barcode plot graph and a feature to produce
wordlist according to different criteria were added.
2. Numerous bugs centered on the way the software entered a 'Busy' mode were corrected.
3. The main core of the software was also updated resulting in a quicker, 'cleaner' processing of the data.
4. Performance improvements should be noticed as a result.
2.2.3
1. Updated file and directory selection dialog boxes to run smoothly in a Windows environment.
2. Also, changed the default colors for sort highlighting, and search window frame size.
3. A number of small bugs were also corrected
2.2.2
1. Corrected critical fault with compiler than caused program to expire when evaluation version of
ActiveState Perl Development Kit expired. Sorry folks!! I didn't realize this would happen!!
2.2.1
1. Corrected bug which prevented new concordance lines being generated if the search term was left the
same and then new files were selected. Port to Linux also completed.
2.2
1. Designed new subroutines for selecting directories and files to solve rendering of dialog windows
problems. This also enables an easier port to Linux.
2.1
1.
2.
3.
4.

Added a second level of sort.

Added ability to restrict searches to full-words only, case sensitive.
Added ability to search using full Perl implemented regular expressions.
Added ability to save results either to a file or the clipboard

2.0
1. Added new sort feature, for rearranging concordance lines.
2. Tidied up the interface. Made the system more robust for novice users. (Now bad input will not cause
the system to crash so easily).
1.1
1) Added binding to allow return key to launch concordance search.
2) Also, recompiled software so that no console is required.
1.0
First version
Copyright Laurence Anthony