In Italy the most famous is Gatto software. You can find more informations here: http://www.ovi.cnr.it/index.php?page=informazioni-generali.

I used AntConc to create and analyse a corpus of several million words for my MA dissertation. It's free and (I think) easy to use. There are also many useful additional tools available from the same website: http://www.laurenceanthony.net/software.html.

Toolbox is easier to work with.

I absolutely agree with Amy! I once used AntConc 3.2.4. It's convenient and easy to use!

Hi,

I too suggest AntConc. Free download and easy to use. There are also several free useful tutorials you can watch on youtube. Hope this helps?

Check out Sketch Engine (www.sketchengine.co.uk). You can get a 30-day free trial in which to evaluate it. There are already hundreds of corpora available in dozens of languages, or you can create your own.

Dear colleagues,

I answer with the delay because I had the problems with my computer. Thank you for your recommendations.

I knew about Toolbox Mr. Behrooz Barjasteh Delforooz mentioned. It is really useful software, some of my colleagues suggest me the Language Explorer as the substitution of Toolbox.

Mrs. Maria Carmela Benvenuto,  I downloaded your recommended programme, now I am testing it.

AntConc 3.2.4. It is for concordancing firstly? I didn't download, could you share your opinions. According to website, this is for concordancing. I forgot to mention the paculiar feature in my question: can this programme shift files from the database to the Internet or in the format of formed dictionary for publication in paper?

Thank you for the programme Sketch Engine. This soft is not free but I try to evaluate it.

Toolbox does that. FLEx is more advanced and a little confusing, at least for me.

Hi,

I recommend to start AntConc (last version).

Also it is a good tool WordSmith. You should pay for using it. Not expensive.

http://www.lexically.net/wordsmith/

I'm using Sketch Engine. It's not to bad:

www.sketchengine.co.uk

Best of british luck,

I've used AntConc and have had my students use it. It does much more than just concordancing. Since it is free, you should try it. There is another program, Miromaa, which is widely used by minimally-competent community groups for creating dictionaries of their languages.

One of the problems of AntConc is that it does not function well under many language conditions, such as Chinese. 

The one I'm using to build a courpus is a website called 'sketch engine' where you can build a corpus based on web searches for specific word entries.

Dear respect colleagues,

Sorry for my absence. It sounds childish but I test all the programmes I recommended at a certain degree.

Now I have the following problem in Language Explorer - I can not add the set phrase into article within my dictionary, what is the method of adding the phrases within ONE article in the dictionary. If someone knows, do not hesitate to mention it here. Thank you.

Hello Alexander

I would suggest Unitex: http://www-igm.univ-mlv.fr/~unitex

It's pretty much complementary to other tools that were mentioned. It's free (even for commercial applications), it understands Unicode, it's relatively easy to use, and you can work on reasonably huge texts (eg: wikipedia text dumps). It does basic and advanced concordancing (by way of the ubiquitous graphs) and more: you can use the individual executables to process repositories of text files (eg: apply some predefined processing steps to a repository of > 1000 individual files). It also features a mature API to integrate in your own tools (UIMA annotators have been developed thanks to this API, it can be integrated into pretty much any oo-oriented program). So you can go all the way from manual click-and-point corpus processing to fully automated text crunching.

Unitex is an academic tool, which is highly customizable (you're not stuck with just English, you can add your own resources). But it's also a very efficient text processing engine: it's being used by at least 5 French start-ups to provide NLP services. I myself used its ancestor to perform real-time text filtering and information extraction on newswire texts. The latest version is optimized to a very high degree, so it won't run out of memory or complain about the size of your text. Needless to say, it runs on any java/C compatible platform (needs some compilation on linux, though).

The only thing that would make it the perfect tool would  be updating the part of speech tagging philosophy. By default, Unitex performs "lexical tagging", which basically means dictionary lookup, without ambiguity resolution (expect some noise). But it comes with an integrated tagger you can train on your own data, and you can import pre-tagged texts (you must follow the expected format though: {word,lemma.Tag:morphology+additional_features}).

I use Unitex a lot for both teaching and research. I try to make it the default tool for my LTTAC (lexicography, terminography and corpus processing) master students.

Hope this helps.

Dear colleagues, thank you for your pieces of advice. Again, sorry for my delay in answering

AntConc 3.2.4.

AntConc (no programming)

Stanford CreNLP (w/ or w/o programming)

Python NLTK (w/ programming)

antConc

Sketch Engine