Main Area and Open Discussion > General Software Discussion
Software idea for speed reading/skimming text by color-coding important keywords
David.P:
True, however for a start I'd be perfectly happy with a simple highlighting of the most frequent words of any text, minus a list of common words, like Wordcounter.com already does not too bad, see the result in my orignal post above.
The next step then could be some more sophisticated algorithms like matching with linguistic word lists, and more.
Anyway, the first step should be doable with very simple tools, even if it is simply most frequent words minus words that are shorter than "n" characters.
Most importantly, this way somebody should be able pick this idea up and maybe start hacking together a simple Firefox extension, for a start.
Cheers David.P
Renegade:
I really like the idea a lot. It's really the kind of things that's right up my alley.
There are a few things I would note about it though.
* Keyword frequency is essential
* Thesaurus looking is also essential, especially for well written works
* A grammar engine would also be needed though. The above are not enough
A grammar engine (like a spell checker) would let you work with your frequencies and synonyms to combine them for longer highlighted sections. The effect is that you could then highlight CONCEPTS better, rather than just keywords.
For example:
A)
A grammar engine (like a spell checker) would let you work with your frequencies and synonyms to combine them for longer highlighted sections. The effect is that you could then highlight CONCEPTS better, rather than just keywords.
B)
A grammar engine (like a spell checker) would let you work with your frequencies and synonyms to combine them for longer highlighted sections. The effect is that you could then highlight CONCEPTS better, rather than just keywords.
The second there is easier to read as it groups a unit of thought rather than a unit of language.
The key though is being able to locate a grammar engine that you can use in the software as everything else is relatively easy in comparison. There's no way that anyone could develop a grammar engine for software like this as the scope is simply far to large there.
Renegade:
Here's one example I found: http://www.wintertree-software.com/dev/wgrammar/index.html
They're kind of hard to find, and quite expensive, which I fully expected.
David.P:
All agreed, but this will be the icing. Let's start with the cake.
Cheers David.P
David.P:
Hi forum,
proudly announcing that the first version of the tool discussed above now is ready and working.
Please read on here (sorry only Google Translation for the time being):
Reading 3.0: Tool for professional readers and news junkies - exclusive premiere
Cheers David.P
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version