On 6 Sep, 06:53, "x...@sympatico.ca" <x...@sympatico.ca> wrote:
> On Aug 26, 3:47 am, Ray Murphy <ray
...@chariot.net.au> wrote:
> > Hi All,
> > Now that so many old books have been converted to text or PDF files,
> > I've made a small program that might eventually be of some use for
> > establishing who might have published certain astrological words or
> > terms first.
> > It's a simple text analysis program that reads any text file and
> > counts each word and the number of times it was used. The results are
> > printed in both word-count (and %) order plus alphabetical order.
> > It will be easy now to see the following:
> > * When an author first used certain terms
> > * Most frequently used words or terms
> > * What words or terms were not used at all
> > * Vocabulary per 1000 words written
> > * Themes in the text
> > Ray
> Ray, you have deserved yet another medal!
> You really mean this works on pdf files as well as
> text files? With different formats?
RM: Well it's good but not ~that~ good, but I still deserve a medal
or one of those stick-on silver stars for making a start. Users need
to download the freePDF-->TXT program that is available here
http://www.download.com/Free-PDF-Text-Reader/3000-10743_4-10373188.html
"Free PDF Text Reader 1.1.41"; Rated 5 cows with Tucows.
165,000 downloads; Suits Windows operating system.
After installation of the above program, users with the WIN operating
system need only to navigate to any PDF file and then after the
conversion is done, save either all pages or individual pages as a
text file. It's so simple that you don't even have to press a "go"
button.
Once a book or article etc is in text format, my small analysis
program does it's thing and prints out a summary in a few styles.
Before exiting the program the user can select unusual or
interesting words in the batch of text by clicking on them to move
them across to an empty list box and those words are appended to the
automatic analysis.
You've probably noticed the small stack of old books that have been
uploaded by Todd to this Google group, so we've got something to
work with. I also understand that nearly all the very old books of all
types that are in one or more British libraries are freely available
due
to the non existence or expiry of copyright.
This sort of program has all sorts of potential for the more complex
analysis of text, and it should work in a fair number of languages.
I'm also trying to figure out how astrological themes could be
automatically
detected -- you know - reading a a Sagittarian's text and finding a
batch of words that sound all Libran :-)
If anyone wants to nominate a book or article that is available in PDF
or Txt format, I can give a demo quite easily.
Ray
|
