Literature authors with similar styles
Matt B
As part of a university project I'm working on a piece of software
which is able to attribute authorship of a document, when presented
with two candidate authors.
It uses various statistical techniques in order to do this, and
requires a large quantity of "training" text - ie works which are
*known* to have been written by each author.
I'm currently gathering together various works from different authors
(from the Project Guttenberg web site) to act as training texts and
test texts for experiments on the software. The works available are
mostly pre-1923 due to copyright.
In order to provide a challenge to the software in determining the
correct author of a certain text, I need to test it with authors whose
writing styles are similar. (Being able to distinguish between works
by Chaucer and Dickens is all well and good, but it wouldn't be that
impressive really!)
I'm not an expert on literarature, so can anyone suggest any pairs of
stylistically similar authors that I can challenge the software with?
Obviously, they must also have literature works available from the
Guttenberg archives.
Cheers for any advice!
Matt
Purl Gurl
(snipped)
> In order to provide a challenge to the software in determining the
> correct author of a certain text, I need to test it with authors whose
> writing styles are similar. (Being able to distinguish between works
> by Chaucer and Dickens is all well and good, but it wouldn't be that
> impressive really!)
http://www.google.com/search?hl=en&lr=&q=%22authors+with+similar+styles%22
Purl Gurl
Purl Gurl
> Matt B wrote:
(snipped)
> > In order to provide a challenge to the software in determining the
> > correct author of a certain text, I need to test it with authors whose
> > writing styles are similar. (Being able to distinguish between works
> > by Chaucer and Dickens is all well and good, but it wouldn't be that
> > impressive really!)
Horace Smith and Percy Shelley
Purl Gurl
Matt B
wrote:
Many thanks. I'd Googled earlier, but it didn't turn much up, which is
why I posted here.
Would you say Dickens and Carroll were stylistically similar?
Matt
John Dean
That's a pity. I'd love to see a comparison between Margaret Drabble and
AS Byatt.
If you're doing poetry then there's always Wordsworth / Coleridge and
Shelley / Byron. Novels? How about Dickens / Thackeray or Dickens /
Trollope?
And while Shakespeare / Marlowe has been done by many others, it's maybe
one to cut your teeth on.
--
John Dean
Oxford
Purl Gurl
(snipped)
Agatha Christie is well known for her mimics of many other authors.
Mark Twain and Will Rogers
John Keats, Percy Shelley, Lord Byron and Mary Wollstonecraft
Purl Gurl
Purl Gurl
> On Wed, 25 Jan 2006 17:18:24 GMT, "Purl Gurl" <purl...@purlgurl.net>
> wrote:
>
> >Purl Gurl wrote:
> >
> >> Matt B wrote:
(snipped)
> >> > In order to provide a challenge to the software in determining the
> >> > correct author of a certain text, I need to test it with authors whose
> >> > writing styles are similar. (Being able to distinguish between works
> >> > by Chaucer and Dickens is all well and good, but it wouldn't be that
> >> > impressive really!)
> >Horace Smith and Percy Shelley
> Would you say Dickens and Carroll were stylistically similar?
No, different writing styles; reality versus fantasy.
However, both authors are known for being rebellious and questioning.
Dickens and Carroll do often write of disparity between an elitist view
of life and a realistic "salt of the earth" view of life.
Their chosen topics are very similar but not their writing styles.
You "might" find similar styles between Dickens' _A Christmas Carol_
and Carroll's _Hunting of the Snark_
Purl Gurl
Purl Gurl
> Matt B wrote:
> > Purl Gurl wrote:
> > >Purl Gurl wrote:
> > >> Matt B wrote:
(snipped)
> > >> > In order to provide a challenge to the software in determining the
> > >> > correct author of a certain text, I need to test it with authors whose
> > >> > writing styles are similar. (Being able to distinguish between works
> > >> > by Chaucer and Dickens is all well and good, but it wouldn't be that
> > >> > impressive really!)
> > Would you say Dickens and Carroll were stylistically similar?
> No, different writing styles; reality versus fantasy.
If asked to label their writing styles, my personal perspective is
Dickens is "heavy and burdening" and Carroll is "light and lively."
Reading those two, Dickens gives me a feeling of shouldering
a burden and Carroll gives me a feeling of being light on my feet.
Dickens trudges along and Carroll dances along, although both
cast an ironic light, if not a hypocritical light upon authority figures.
Purl Gurl
Alan Jones
"Matt B" <matt...@hotmail.co.uk> wrote in message
news:pfdft19oh89ggq4hm...@4ax.com...
Not very. Both are rather strikingly idiosyncratic, as indeed are most
authors whose works have become classics. But if you want 19th century
novelists perhaps Trollope and Thackeray would be a more useful pair. As a
"challenge" you could also try samples from early and late work by the same
author - or do you assume that certain characteristics are unchanging?
There is a long history of such stylistic tests, though of course not
computerised until very recently, and their results have been most
convincing to readers whose prior assumptions happen to be supported by a
particular test. One classic instance is the attempt to distinguish between
the styles of Shakespeare and his collaborators in what are assumed to be
joint compositions, and then to identify those collaborators by examining
the work of various contemporaries. Another is the question of how many of
the scriptural Letters attributed to St Paul are by the same person.
Alan Jones
Mike Lyle
I can see a non-specialist running into the sands with this one, through
no fault of his own. You'll need to bone up a lot on language before you
can ask the right questions: simple vocabulary-counts will only get you
so far (I'm not saying they're useless: some interesting work's been
done that way -- but don't ask me for references), as language is based
on combinations of words, not really the words themselves. Is there any
chance of collaborating with at least one sympathetic person from the
English Department?
--
Mike.
Matt B
Thanks, John, for all of those suggestions.
I'll be including prose, plays and poetry in the tests; the software
will be able to handle all of those formats. Marlowe vs Shakespeare
should be a good test of its ability.
I'm also going to try Shakespeare vs Fletcher on Henry VIII as a
genuine attribution test, just to see what results the software comes
up with. I know the authorities on the matter attribute most of it to
Fletcher, so it will be interesting to see what results I get.
Matt
Matt B
wrote:
Nicely put! You do seem to know your literature.
I have to say, I'm not familiar with Carroll's works (IT student, as
opposed to English lit), but I'm actually a big fan of Dickens. His
stuff can be miserable, but in a way that's what I like about it.
There was a BBC adaptation of Bleak House screened at the end of last
year, which was quite simply *the best thing* I have ever seen on TV.
http://www.amazon.co.uk/exec/obidos/ASIN/B000CCE25K/203-2178738-9867949
I would highly recommend it to anyone. A true masterpiece.
Matt
Tony Cooper
wrote:
>I'm not an expert on literarature, so can anyone suggest any pairs of
>stylistically similar authors that I can challenge the software with?
Purl Gurl and Chess One.
>Obviously, they must also have literature works available from the
>Guttenberg archives.
Oh. Never mind.
--
Tony Cooper
Orlando, FL
R H Draney
>
>Matt B wrote:
>>
>> In order to provide a challenge to the software in determining the
>> correct author of a certain text, I need to test it with authors whose
>> writing styles are similar. (Being able to distinguish between works
>> by Chaucer and Dickens is all well and good, but it wouldn't be that
>> impressive really!)
>>
>> I'm not an expert on literarature, so can anyone suggest any pairs of
>> stylistically similar authors that I can challenge the software with?
>> Obviously, they must also have literature works available from the
>> Guttenberg archives.
>
>Shelley / Byron. Novels? How about Dickens / Thackeray or Dickens /
>Trollope?
>And while Shakespeare / Marlowe has been done by many others, it's maybe
>one to cut your teeth on.
It's up to you to scan the archives, but try Benchley/Thurber or
Poe/Bierce...(gotta get some Yanks in there)....r
Matt B
<mike_l...@REMOVETHISyahoo.co.uk> wrote:
>
>I can see a non-specialist running into the sands with this one, through
>no fault of his own. You'll need to bone up a lot on language before you
>can ask the right questions: simple vocabulary-counts will only get you
>so far (I'm not saying they're useless: some interesting work's been
>done that way -- but don't ask me for references), as language is based
>on combinations of words, not really the words themselves. Is there any
>chance of collaborating with at least one sympathetic person from the
>English Department?
Mike -
Funnily enough very little knowledge of language is going to be
necessary. The analysis of the texts is based on stylometry
(statistical analysis of literary style).
Although traditional content analysis has been widely used in many
previous (non computer-based) authorship attribution studies, most
experiments carried out in recent years have used techniques borrowed
from statistics, including Naive Bayes, K-Nearest Neighbour, CUSUM
etc. Some have proven more fruitful than others.
I'll definitely need some help from the statistics department, but I
should be able to get by with my limited knowledge of linguistic
styles.
Matt
Sara Lorimer
> There was a BBC adaptation of Bleak House screened at the end of last
> year, which was quite simply *the best thing* I have ever seen on TV.
>
> http://www.amazon.co.uk/exec/obidos/ASIN/B000CCE25K/203-2178738-9867949
>
> I would highly recommend it to anyone. A true masterpiece.
I watched the first few minutes of it. I couldn't stand the way it was
filmed and edited. I don't want to notice camera angles.
--
SML
Matt B
Sara, Sara, Sara... what can I say?
You missed a treat! Ignore the camera angles! Concentrate on the
drama.
Honestly, it's worth it. I even wrote a fan letter to Anna Maxwell
Martin (who played Esther) and I never do stuff like that.
Buy it on DVD and then when you've watched it, you can come back here
and say "thanks Matt - I owe you one".
Matt
Matt B
wrote:
>
>"Matt B" <matt...@hotmail.co.uk> wrote in message
>news:pfdft19oh89ggq4hm...@4ax.com...
>>
>> Would you say Dickens and Carroll were stylistically similar?
>
>Not very. Both are rather strikingly idiosyncratic, as indeed are most
>authors whose works have become classics. But if you want 19th century
>novelists perhaps Trollope and Thackeray would be a more useful pair. As a
>"challenge" you could also try samples from early and late work by the same
>author - or do you assume that certain characteristics are unchanging?
No, and that's a good idea. The more assorted the range of tests, the
more robust the tool will seem.
I have until May before I have to demonstrate it, so plenty of time to
mix it around and try different approaches.
Matt
Pat Durkin
> On Wed, 25 Jan 2006 12:20:29 -0800, que.sara....@gmail.com
> (Sara Lorimer) wrote:
>
>>Matt B <matt...@hotmail.co.uk> wrote:
>>
>>> There was a BBC adaptation of Bleak House screened at the end of
>>> last
>>> year, which was quite simply *the best thing* I have ever seen on
>>> TV.
>>>
>>> http://www.amazon.co.uk/exec/obidos/ASIN/B000CCE25K/203-2178738-9867949
>>>
>>> I would highly recommend it to anyone. A true masterpiece.
>>
>>I watched the first few minutes of it. I couldn't stand the way it was
>>filmed and edited. I don't want to notice camera angles.
>
> Sara, Sara, Sara... what can I say?
>
> You missed a treat! Ignore the camera angles! Concentrate on the
> drama.
>
> Honestly, it's worth it. I even wrote a fan letter to Anna Maxwell
> Martin (who played Esther) and I never do stuff like that.
>
think it is an excellent production. No matter how realistic the
filming attempts to be, however, I found the interior scenes' darkness a
real bother.
And I could stand for more close-ups.
I have been thinking of reading it for a long time, but the length of
the work and the Victorian style of writing quite put me off. And now I
need extra-strong lighting just to do crossword puzzles.
jerry_f...@yahoo.com
> "Matt B" <matt...@hotmail.co.uk> wrote in message
> news:pfdft19oh89ggq4hm...@4ax.com...
> > On Wed, 25 Jan 2006 17:18:24 GMT, "Purl Gurl" <purl...@purlgurl.net>
> > wrote:
> >
> >>Purl Gurl wrote:
> >>
> >>> Matt B wrote:
> >>
> >>(snipped)
> >>
> >>> > In order to provide a challenge to the software in determining the
> >>> > correct author of a certain text, I need to test it with authors whose
> >>> > writing styles are similar. (Being able to distinguish between works
> >>> > by Chaucer and Dickens is all well and good, but it wouldn't be that
> >>> > impressive really!)
Maybe Addison and Steele?
> > Would you say Dickens and Carroll were stylistically similar?
>
> Not very. Both are rather strikingly idiosyncratic, as indeed are most
> authors whose works have become classics. But if you want 19th century
> novelists perhaps Trollope and Thackeray would be a more useful pair. As a
> "challenge" you could also try samples from early and late work by the same
> author - or do you assume that certain characteristics are unchanging?
...
Another challenge is single works in which an author uses different
styles. _Ulysses_ is at Gutenberg. Then there's _Tristram Shandy_.
Can your software tell the authorship of the main text from Yorick's
sermon (probably not) or Shandy's father's stuff (maybe)? Should it?
I don't think I've read any epistolary novels from the pre-copyright
era, but they might be good places to look for deliberate attempts to
change one's style.
--
Jerry Friedman
the Omrud
What he said. Except for the stalking of the heroine. Shake me up,
Judy.
--
David
=====
replace usenet with the
Purl Gurl
> As part of a university project I'm working on a piece of software
> which is able to attribute authorship of a document, when presented
> with two candidate authors.
> It uses various statistical techniques in order to do this, and
> requires a large quantity of "training" text - ie works which are
> *known* to have been written by each author.
First order of business for a text parser of this nature is to
scan for character names.
Puck
Willy Loman
Lopakhin
McKendrick
Celimene
Bianca
Oedipus
Laura Wingfield
Thomas Stockmann
Those are character names, a few of thousands of famous character
names which all readers here should instantly recognize, if literate.
A reader should be able to provide the title of the work and the author
without hesitation.
From a programming point of view, character names are very unique.
Parsing for character names would be a logical initial parsing point.
However, with my having over a decade of programming experience
in Natural Language Emulation, I do know this task of yours is
almost impossible, to any degree of acceptable success.
You will need a very powerful machine and a very fast machine
supported by thousands of gigabytes of data for comparison.
To be successful, you have no choice but to parse complete
sentences and a good number of sample sentences. You
cannot parse for individual words.
I have doubts you can create a database which reflects unique
writing styles of equally unique authors.
Developing an effective database will take years. For my Roberta,
I am now in my tenth year of programming and data development
and her general conversation skills are barely equal to a fourth or
fifth grade student. Her ability to cite hard data such as science
and history, her ability to perform math, those types of abilities
infinitely surpass the human mind, but she cannot attain a level
of common conversation above an eight to ten year old child.
You face very serious programming challenges.
Purl Gurl
R H Draney
>
>Puck
>Willy Loman
>Lopakhin
>McKendrick
>Celimene
>Bianca
>Oedipus
>Laura Wingfield
>Thomas Stockmann
>
>Those are character names, a few of thousands of famous character
>names which all readers here should instantly recognize, if literate.
>A reader should be able to provide the title of the work and the author
>without hesitation.
I got four without thinking about it, and one other that looked vaguely
familiar...now, that's names that ought to be familiar in the English-speaking
parts of the world...how many fictional characters of recent (say the last 150
years) vintage would be recognizable *anywhere* in the world?...a few examples:
Batman
Mickey Mouse
Zorro
Tarzan
Sherlock Holmes
Maybe Popeye too, but I'm not holding my breath....r
John Dean
Ditto Two Noble Kinsman.
And it would be nice if you could try Bard with the Earl of Oxford,
mainly because I don't like to pass up an opportunity to encourage
others to spit in the eye of the idiots on
humanities.lit.authors.shakespeare who think Shakespeare was too base
born to have amounted to anything.
--
John Dean
Oxford
Purl Gurl
> Purl Gurl filted:
> >Puck
> >Willy Loman
> >Lopakhin
> >McKendrick
> >Celimene
> >Bianca
> >Oedipus
> >Laura Wingfield
> >Thomas Stockmann
> >Those are character names, a few of thousands of famous character
> >names which all readers here should instantly recognize, if literate.
> >A reader should be able to provide the title of the work and the author
> >without hesitation.
> I got four without thinking about it, and one other that looked vaguely familiar...
Most likely, Puck, Willy Lohman, Oedipus and Laura Wingfield are
the most often recognized character names amongst the literate.
> now, that's names that ought to be familiar in the English-speaking
> parts of the world...how many fictional characters of recent (say the last 150
> years) vintage would be recognizable *anywhere* in the world?...a few examples:
> Batman
> Mickey Mouse
> Zorro
> Tarzan
> Sherlock Holmes
I would guess Mickey Mouse to be the best known name
around the world. Ignoring recent Batman movies, maybe
Tarzan or Sherlock Holmes would be second best known
of character names.
Personally, Tarzan entered my life long before Sherlock Holmes.
Purl Gurl
Stephen Calder
You mean Lewis Carroll? He's miles from Dickens and not a good one to
compare with anyone. Carroll's language is bizarre and unique.
How about Jane Austen and Emily Bronte?
Anthony Trollope and Henry James?
--
Stephen
Lennox Head, Australia
bert
> As part of a university project I'm working on a piece of software
> . . . can anyone suggest any pairs of stylistically similar authors
> that I can challenge the software with?
No-one yet seems to have suggested John Buchan and Arthur Conan Doyle.
Their language, and many of their plot elements, are remarkably
similar. H. Rider Haggard uses a similar language style, but rather
different plot elements.
John Dean
I hope you plan to report back here, or link us to a site where you
announce results.
--
John Dean
Oxford
Mike Lyle
[...]
> Ditto Two Noble Kinsman.
> And it would be nice if you could try Bard with the Earl of Oxford,
> mainly because I don't like to pass up an opportunity to encourage
> others to spit in the eye of the idiots on
> humanities.lit.authors.shakespeare who think Shakespeare was too base
> born to have amounted to anything.
Hear, hear! I reckon we can do it without benefit of compution, though.
I tried reading some Oxford aloud to a friend who'd just been all but
convinced by one of the Oxfordian books, and he agreed at once that the
theory leaked like a colander. It was a very palpable pro-am tournament.
--
Mike.
Matt B
wrote:
>
>What he said. Except for the stalking of the heroine. Shake me up,
>Judy.
Aw come on - it was just a simple fan letter. Hardly a marriage
proposal!
Although, that said, anyone know if she's hitched?
Matt
Matt B
Sure will, my friend.
If I get time (and can work out how) I'll stick it up on a web page,
so that everyone can have a play. That's assuming I can get it working
of course, which is far from guaranteed.
Matt
the Omrud
> On Wed, 25 Jan 2006 23:03:05 GMT, the Omrud <usenet...@gmail.com>
Interestingly, IMDB says:
Though has worked rather extensively on stage, Anna has graced the
big and small screen. Guest appearances in "Midsomer Murders" (1997)
and opposite Christopher Eccleston and Billie Piper in "Doctor Who"
(1963)
She certainly started young.
However, FWIW, I continue to be impressed by the calm career of Denis
Lawson. OK, he won't reach the parts his nephew has reached, but
he's never disappointed and he was utterly perfect as Mr Jarndyce.
And he was in Star Wars Eps 4, 5 and 6.
Matt B
wrote:
>First order of business for a text parser of this nature is to
>scan for character names.
>
>Puck
>Willy Loman
>Lopakhin
>McKendrick
>Celimene
>Bianca
>Oedipus
>Laura Wingfield
>Thomas Stockmann
>
>Those are character names, a few of thousands of famous character
>names which all readers here should instantly recognize, if literate.
Oh dear. I got 2. I think that makes me deeply illiterate :(
>A reader should be able to provide the title of the work and the author
>without hesitation.
>
>From a programming point of view, character names are very unique.
>Parsing for character names would be a logical initial parsing point.
Well, you are correct in one way and wrong in another. The starting
point is indeed to remove all proper nouns. However, these form no
part of the analysis.
There are several techniques available for authorship attribution, but
the technique I will be using finds similarities between documents,
based on the frequencies of words they contain.
An author will generally have a favourite set of words that he / she
consciously or unconsciously uses often. Finding similar frequencies
of these words across two documents indicates homogeneity. Although
it's a bit more complex than that, that's the basic premise.
Including proper nouns such as character names in the process would be
of no benefit. In fact it would be detrimental because it would lead
to the software choosing a load of character names as the author's
favourite words. Since a set of character names is generally unique to
one text, any other text that didn't contain those names would be seen
as unlikely to have been written by the same person.
>
>However, with my having over a decade of programming experience
>in Natural Language Emulation, I do know this task of yours is
>almost impossible, to any degree of acceptable success.
No - not at all. There have been loads of computer-based attribution
studies done over the last few years, some of which have been very
successful. If you're interested, I can give you some references.
>
>You will need a very powerful machine and a very fast machine
>supported by thousands of gigabytes of data for comparison.
>To be successful, you have no choice but to parse complete
>sentences and a good number of sample sentences. You
>cannot parse for individual words.
Again, this is just not true. Parsing for individual words *is* the
way the software will work. A bog-standard PC will be able to handle
it easily.
>
>I have doubts you can create a database which reflects unique
>writing styles of equally unique authors.
>
Well, fortunately there's no database work involved.
>Developing an effective database will take years. For my Roberta,
>I am now in my tenth year of programming and data development
>and her general conversation skills are barely equal to a fourth or
>fifth grade student. Her ability to cite hard data such as science
>and history, her ability to perform math, those types of abilities
>infinitely surpass the human mind, but she cannot attain a level
>of common conversation above an eight to ten year old child.
>
What / who is Roberta? Are you talking about a program?
Matt
Laura F. Spira
>
> However, FWIW, I continue to be impressed by the calm career of Denis
> Lawson. OK, he won't reach the parts his nephew has reached, but
> he's never disappointed and he was utterly perfect as Mr Jarndyce.
> And he was in Star Wars Eps 4, 5 and 6.
>
I gave up on "Bleak House" very early on - it's a favourite book of mine
and I rarely enjoy adaptations of my favourites. I also found all the
faces familiar from other settings very disconcerting. But one reason
that I bothered at all was Denis Lawson - he is a fine actor.
--
Laura
(emulate St. George for email)
Matt B
wrote:
>Matt B <matt...@hotmail.co.uk> spake thusly:
>
>> On Wed, 25 Jan 2006 23:03:05 GMT, the Omrud <usenet...@gmail.com>
>> wrote:
>>
>> >
>> >What he said. Except for the stalking of the heroine. Shake me up,
>> >Judy.
>>
>> Aw come on - it was just a simple fan letter. Hardly a marriage
>> proposal!
>>
>> Although, that said, anyone know if she's hitched?
>
>Interestingly, IMDB says:
>
>Though has worked rather extensively on stage, Anna has graced the
>big and small screen. Guest appearances in "Midsomer Murders" (1997)
>and opposite Christopher Eccleston and Billie Piper in "Doctor Who"
>(1963)
>
>She certainly started young.
Yikes! Then she's too old for me.
>
>However, FWIW, I continue to be impressed by the calm career of Denis
>Lawson. OK, he won't reach the parts his nephew has reached, but
>he's never disappointed and he was utterly perfect as Mr Jarndyce.
>And he was in Star Wars Eps 4, 5 and 6.
Yes, I was also very impressed. I thought maybe he'd win the best
actor of 2005 but Christopher Eccleston cleaned up once again.
Matt
Matt B
<jerry_f...@yahoo.com> wrote:
>Another challenge is single works in which an author uses different
>styles. _Ulysses_ is at Gutenberg. Then there's _Tristram Shandy_.
>Can your software tell the authorship of the main text from Yorick's
>sermon (probably not) or Shandy's father's stuff (maybe)? Should it?
>I don't think I've read any epistolary novels from the pre-copyright
>era, but they might be good places to look for deliberate attempts to
>change one's style.
Hmm... not too sure about that to be honest. Stylistic changes within
a document would probably make matters difficult - especially if the
changes brought about a drastic change in vocabulary from the author's
usual style.
Of course, for each author, I'll need a large amount of training text
- consisting of several thousand words of other work - so obviously I
won't be able to use it to attribute authorship of a document if there
are no previous examples of such work.
Matt
Matt
K. Edgcombe
Matt <matt...@hotmail.co.uk> wrote:
>>Matt B wrote:
>>> As part of a university project I'm working on a piece of software
>>> which is able to attribute authorship of a document . . .
>>> . . . can anyone suggest any pairs of stylistically similar authors
>>> that I can challenge the software with?
>>
Any two writers of the same period will be liable to show certain
resemblances: for instance Chesterton and Shaw, who are really very different,
might show up as closer than they should be. An interesting one would be
Chesterton and Dickens, who to my ear ought to be closer than - say - Trollope
and Dickens; but in most cases I suppose differences of vocabulary will
dominate when comparing across centuries. (Oh, and Chesterton and C.S.Lewis,
but I don't know how much of the latter is freely available).
Are you analysing just word
frequencies, or phrases? I think most non-specialist readers will recognise
styles from whole sentences, not from vocabulary, and in the case of Chesterton
it's the sentence formation and punctuation that gives him away.
Hardy and George Eliot would be worth a try; they are easily
distinguishable in some ways but there are whole passages it would be possible
to confuse. I can't think of anyone else to pair with Hardy.
Katy
the Omrud
Purl Gurl
> Purl Gurl wrote:
> >From a programming point of view, character names are very unique.
> >Parsing for character names would be a logical initial parsing point.
> Well, you are correct in one way and wrong in another. The starting
> point is indeed to remove all proper nouns. However, these form no
> part of the analysis.
Parsing is not removing. Parsing is to seek per specific parameters.
Typical parsing is regular expression (regex) matching, not removal.
A programmer knows what is parsing. Clearly you do not know.
You are not a programmer.
Parsing for character name is the most logical initial method to use.
A matching of the character name "Puck" instantly prompts your
program to return a work title and author. Secondary parsing would
be to verify "Puck" is not an inclusion in a different work.
> There are several techniques available for authorship attribution, but
> the technique I will be using finds similarities between documents,
> based on the frequencies of words they contain.
> An author will generally have a favourite set of words that he / she
> consciously or unconsciously uses often.
Many authors use the same expressions. All authors use the same
words with great frequency; the, a, his, their, your....
> Including proper nouns such as character names in the process would be
> of no benefit. In fact it would be detrimental because it would lead
> to the software choosing a load of character names as the author's
> favourite words.
You are clearly not a programmer. Character names are the most powerful
and best key terms to use for initial parsing. You are dead wrong.
> >However, with my having over a decade of programming experience
> >in Natural Language Emulation, I do know this task of yours is
> >almost impossible, to any degree of acceptable success.
> No - not at all. There have been loads of computer-based attribution
> studies done over the last few years, some of which have been very
> successful. If you're interested, I can give you some references.
Language parsing is still in a very primitive stage of development.
There are no successful language parsers, to date. This is a new
frontier to be explored and this frontier is being explored by the
likes of Bell Labs, IBM, Microsoft and many others. None have
developed a successful language parser. There is a reason for this.
This reason is we cannot write software which can equal the
intellectual capacity of the human mind. This is impossible.
> >You will need a very powerful machine and a very fast machine
> >supported by thousands of gigabytes of data for comparison.
> >To be successful, you have no choice but to parse complete
> >sentences and a good number of sample sentences. You
> >cannot parse for individual words.
> Again, this is just not true. Parsing for individual words *is* the
> way the software will work. A bog-standard PC will be able to handle
> it easily.
You are most incorrect. Parsing for individual words is the worst
approach, the most illogical approach, a serious mistake.
For any given average page of text, picked at random, you will
find common words in all samples; a, the, to, at, this, that....
Another example is "Ozymandias" will be found in a number
of works besides Shelley's work. You cannot parse for select
individual words. This simply cannot be done, with success,
for your given programming objective. This is a classic case
example of parsing for a character name which requires much
need for secondary, even tertiary parsing for verification.
Nonetheless, a hit on "Ozymandias" will narrow down parsing
to perhaps four to six authors, rather than literally any author
which in turn significantly narrows database usage and very
significantly limits database entries to a very narrow range.
You cannot identify a literary work by parsing individual words.
You can only parse for a specific literary work by matching a
series of words strung together in a highly recognizable pattern.
> >I have doubts you can create a database which reflects unique
> >writing styles of equally unique authors.
> Well, fortunately there's no database work involved.
Again, you are indicating you are not a programmer. You _must_ have
a database for parsing comparison, a huge database. You simply
cannot parse for select phrases without a basis for comparison.
> >Developing an effective database will take years. For my Roberta,
> >I am now in my tenth year of programming and data development
> >and her general conversation skills are barely equal to a fourth or
> >fifth grade student. Her ability to cite hard data such as science
> >and history, her ability to perform math, those types of abilities
> >infinitely surpass the human mind, but she cannot attain a level
> >of common conversation above an eight to ten year old child.
> What / who is Roberta? Are you talking about a program?
Roberta and Robby are Natural Language Emulation software which
I have written and continue to write. Roberta receives the most attention.
Her multiple databases, combined, now far exceed fifty-gigabytes which
affords her an ability to converse with humans at roughly the level of a
ten year old child. Her main programming is over six-thousand lines
of Perl code. She is supported by MSDOS Basic, ANSI C, C plus plus
and Visual Basic. Roberta's programming is so complex, so large, she
requires two Pentium 4 class machines each with two gigabytes of
memory with an operating speed of three gigaHertz.
Even with all that technology, Roberta remains a very rudimentary speaker
prone to many errors; language is extremely complex.
Roberta parses textual input, processes that input for contextual
understanding, then formulates a correct, or close to correct coherent
response for a visitor.
You are not a programmer, not even close to being a programmer.
These statements you make are completely contradictory to all
programming languages techniques from FORTRAN to MS Net.
Again, you have made it very clear you are not a programmer.
Incidentally, I have been writing software for Roberta which parses for
well known phrases. She will be able to recognize a work title and
author based on text parsing, which is precisely what you "think"
you are to do. I have been writing this type of software for years.
Despite three years of development, this software has still not
reached a level of success warranting bringing this feature online.
You will find commentary on this here,
http://www.purlgurl.net/~callgirl/roberta/help.html
Matt, I am a good ten years ahead of you on this topic. You are
not even close to being in my league. Be careful about indicating
a professional is wrong, you might bite off more than you can chew.
My very moniker should be clue enough, Perl Girl - Purl Gurl
http://www.purlgurl.net/~callgirl/roberta/roberta.cgi
http://www.purlgurl.net/~callgirl/android.html
Purl Gurl
Adam Grinter
>
> I'm not an expert on literarature, so can anyone suggest any pairs of
> stylistically similar authors that I can challenge the software with?
> Obviously, they must also have literature works available from the
> Guttenberg archives.
The Bronte sisters.
Linz
> However, FWIW, I continue to be impressed by the calm career of Denis
> Lawson. OK, he won't reach the parts his nephew has reached, but
> he's never disappointed and he was utterly perfect as Mr Jarndyce.
> And he was in Star Wars Eps 4, 5 and 6.
Kit Curran, King of the Airways!
Matt B
wrote:
<snip>
>
>Parsing is not removing. Parsing is to seek per specific parameters.
>Typical parsing is regular expression (regex) matching, not removal.
>
>A programmer knows what is parsing. Clearly you do not know.
>You are not a programmer.
Yeah, I know what parsing is. The document is parsed for proper nouns,
so that they can be removed, and thus form no part of the analysis.
>
>Parsing for character name is the most logical initial method to use.
>A matching of the character name "Puck" instantly prompts your
>program to return a work title and author. Secondary parsing would
>be to verify "Puck" is not an inclusion in a different work.
>
No no no, Purl Gurl! I think you misunderstand the point of the
software. It's not intended to identify the title and author of a work
from pre-existing knowledge.
Rather, the software (if I can get away from USENET long enough to
make it) attributes authorship of a document to one of two given
authors, based on features derived from each author's collection of
training text.
If it was searching a database for an author / title based on stored
knowledge, then yes, identifying proper nouns would be a logical
approach.
>
>> There are several techniques available for authorship attribution, but
>> the technique I will be using finds similarities between documents,
>> based on the frequencies of words they contain.
>
>> An author will generally have a favourite set of words that he / she
>> consciously or unconsciously uses often.
>
>Many authors use the same expressions. All authors use the same
>words with great frequency; the, a, his, their, your....
But not necessarily at the same frequency. And what about less
commonly used words? If we gather all of these aspects together, we
can generate a pretty good literary "fingerprint" of the author.
>
>> Including proper nouns such as character names in the process would be
>> of no benefit. In fact it would be detrimental because it would lead
>> to the software choosing a load of character names as the author's
>> favourite words.
>
>You are clearly not a programmer. Character names are the most powerful
>and best key terms to use for initial parsing. You are dead wrong.
>
Depends on the purpose, Purl Gurl. See above.
>
>> >However, with my having over a decade of programming experience
>> >in Natural Language Emulation, I do know this task of yours is
>> >almost impossible, to any degree of acceptable success.
>
>> No - not at all. There have been loads of computer-based attribution
>> studies done over the last few years, some of which have been very
>> successful. If you're interested, I can give you some references.
>
>Language parsing is still in a very primitive stage of development.
>There are no successful language parsers, to date. This is a new
>frontier to be explored and this frontier is being explored by the
>likes of Bell Labs, IBM, Microsoft and many others. None have
>developed a successful language parser. There is a reason for this.
>This reason is we cannot write software which can equal the
>intellectual capacity of the human mind. This is impossible.
AI completeness is far off in the distance - true - and pehaps it will
always remain as such. But we can still create computers to beat
humans at chess can't we?
Like I say, there have been loads of successful computer-based
attribution studies done over the last few years.
<snip>
>
>> >I have doubts you can create a database which reflects unique
>> >writing styles of equally unique authors.
>
>> Well, fortunately there's no database work involved.
>
>Again, you are indicating you are not a programmer. You _must_ have
>a database for parsing comparison, a huge database. You simply
>cannot parse for select phrases without a basis for comparison.
I won't be using a database. I'll be comparing a given work with two
collections of training text. Each of these will be stored in flat
files and read into the software.
This is the *fifth* time you've said this in this post! Message
received, noted and understood!
>
>Incidentally, I have been writing software for Roberta which parses for
>well known phrases. She will be able to recognize a work title and
>author based on text parsing, which is precisely what you "think"
>you are to do. I have been writing this type of software for years.
>Despite three years of development, this software has still not
>reached a level of success warranting bringing this feature online.
>
>You will find commentary on this here,
>
>http://www.purlgurl.net/~callgirl/roberta/help.html
>
>Matt, I am a good ten years ahead of you on this topic. You are
>not even close to being in my league. Be careful about indicating
>a professional is wrong, you might bite off more than you can chew.
>My very moniker should be clue enough, Perl Girl - Purl Gurl
Oh yeah. Well that proves it then.
I've no doubt you do have considerably more programming experience
than I, but that's not so say you know what you're talking about on
this issue. Because you don't, quite frankly.
>
>http://www.purlgurl.net/~callgirl/roberta/roberta.cgi
>
>http://www.purlgurl.net/~callgirl/android.html
>
I've just had a go with Roberta. Actually, it's pretty impressive. You
must have put a lot of time into that.
She was pretty poor at holding a conversation, I must say, and I doubt
that even the most naive person could ever mistake her for human, but
she recognised basic questions such as age etc. She also seemed to
have a good knowledge of facts.
Also had a go with your automatic poem generator and automatic
headline generator. Funny stuff. I can see how they work, but good job
all the same.
Have you tried Alice? Now this girl is amazing:
Matt
Matt B
Thanks, Katy.
I'll be analysing individual words, as well as doubles and triples.
Anything past that and I'm doubtful that the phrases would occur often
enough to make a successful comparison.
It's a purely statistical approach though. There's no analysis of the
semantic content, in the way a person would do.
Matt
jerry_f...@yahoo.com
> On Fri, 27 Jan 2006 01:49:23 GMT, "Purl Gurl" <purl...@purlgurl.net>
> wrote:
>
> <snip>
> No no no, Purl Gurl! I think you misunderstand the point of the
> software.
...
Maybe it's time someone let you know that Purl Gurl's signal-to-noise
ratio in aue has been extremely low. In particular, I'd say that if
you persist in trying to explain something to her that she's
misunderstood, your chance of receiving personal or racial or sexual
insults is higher than your chance of getting her to
understand--despite her undoubted intelligence and her competence as a
programmer.
Of course, this is the sort of thing I like to be proven wrong about.
--
Jerry Friedman
Purl Gurl
> Matt B wrote:
> > Purl Gurl wrote:
(snipped)
> > No no no, Purl Gurl! I think you misunderstand the point of the
> > software.
> Maybe it's time someone let you know that Purl Gurl's signal-to-noise
> ratio in aue has been extremely low. In particular, I'd say that if
> you persist in trying to explain something to her that she's
> misunderstood, your chance of receiving personal or racial or sexual
> insults is higher than your chance of getting her to
> understand--despite her undoubted intelligence and her competence as a
> programmer.
Previously this group enjoyed a sudden appearance of a number of
cultural anthropologists displaying expert knowledge about my
Choctaw culture.
Now this group enjoys a sudden appearance of a number of professional
programmers displaying expert knowledge about my programming culture.
> Of course, this is the sort of thing I like to be proven wrong about.
Stereotypical masculine behavior; always the fault of another.
Purl Gurl
Vinny Burgoo
>Cheers for any advice!
The different styles used by Doyle in his Sherlock Holmes and Brigadier
Gerard stories might be useful for tuning the sensitivity of your prog.
BG is lighter and more relaxed (and more humorous) than SH - and is
supposedly written by a Frenchman, above the market!
--
V
Matt B
<jerry_f...@yahoo.com> wrote:
>
>Maybe it's time someone let you know that Purl Gurl's signal-to-noise
>ratio in aue has been extremely low. In particular, I'd say that if
>you persist in trying to explain something to her that she's
>misunderstood, your chance of receiving personal or racial or sexual
>insults is higher than your chance of getting her to
>understand--despite her undoubted intelligence and her competence as a
>programmer.
>
>Of course, this is the sort of thing I like to be proven wrong about.
From reading some of her latest contributions, I see your point.
However, at the risk of becoming unpopular, I do find her fairly
entertaining.
She's an eccentric woman who has some odd (to say the least)
viewpoints and I can see how that could cause controversy around here,
but at the same time her posts make me laugh - even though I'm sure
they're not intended to.
Gives the group a bit of an edge, I feel.
Of course, I've only been reading the group for 3 days. There is time
for the novelty to wear off...
Matt
jerry_f...@yahoo.com
I just wanted you not to waste your time trying to convince her that
she had misunderstood something. But if you find her posts
entertaining, then you don't need my warning.
I missed the fact that she'd already provided you with insults. At
least, if I were a programmer, I'd be insulted if someone told me I
wasn't. You were very patient.
--
Jerry Friedman
Tony Cooper
wrote:
>On 27 Jan 2006 09:58:04 -0800, "jerry_f...@yahoo.com"
><jerry_f...@yahoo.com> wrote:
>
>>
>>Maybe it's time someone let you know that Purl Gurl's signal-to-noise
>>ratio in aue has been extremely low. In particular, I'd say that if
>>you persist in trying to explain something to her that she's
>>misunderstood, your chance of receiving personal or racial or sexual
>>insults is higher than your chance of getting her to
>>understand--despite her undoubted intelligence and her competence as a
>>programmer.
>>
>>Of course, this is the sort of thing I like to be proven wrong about.
>
>From reading some of her latest contributions, I see your point.
>However, at the risk of becoming unpopular, I do find her fairly
>entertaining.
Some people find dwarf tossing to be entertaining. Some people
actually find watching bowling on television to be entertaining. I'm
not exactly sure where I would place PGs posts in that sequence, but I
think it would be somewhere above the bowling thing. I've never seen
a dwarf tossed, though.
--
Tony Cooper
Orlando, FL
R H Draney
>
>Some people find dwarf tossing to be entertaining. Some people
>actually find watching bowling on television to be entertaining. I'm
>not exactly sure where I would place PGs posts in that sequence, but I
>think it would be somewhere above the bowling thing. I've never seen
>a dwarf tossed, though.
Rent the "Lord of the Rings" videos...I'm sure it happens at least once in
there....r
Skitt
Go to http://tinyurl.com/9ejhx and select 19.
--
Skitt (in Hayward, California)
http://www.geocities.com/opus731/
the Omrud
> Some people find dwarf tossing to be entertaining.
It is somehow comforting, in these days of uncertainty, to be able to
identify the author of a post without needing to scroll down to the
sig.
R J Valentine
...
} Some people find dwarf tossing to be entertaining. Some people
} actually find watching bowling on television to be entertaining. I'm
} not exactly sure where I would place PGs posts in that sequence, but I
} think it would be somewhere above the bowling thing. I've never seen
} a dwarf tossed, though.
How about golf on televistion? How about the Jerry Springer Show? How
about public-television pledge drives? How about Jerry Lewis's marathon
fundraisers? Do they televise frozen-turkey bowling in the two-liter
aisle of supermarkets?
--
rjv
Robert Lieblich
Someome must watch all of those or they wouldn't be televised. Hell,
people watch informercials. There's no accounting for taste.
Or is the question where on a continuum of these various shows to put
dwarf-tossing? Myself, I'd put it to the side, like the rare earths.
--
Bob Lieblich
Bring on the Masters
Matt B
<tony_co...@earthlink.net> wrote:
>
>Some people find dwarf tossing to be entertaining. Some people
>actually find watching bowling on television to be entertaining. I'm
>not exactly sure where I would place PGs posts in that sequence, but I
>think it would be somewhere above the bowling thing. I've never seen
>a dwarf tossed, though.
I'd say, from least entertaining to most entertaining:
Watching bowling on television
Dwarf tossing
Purl Gurl's posts
Topless darts (sadly, no longer on British TV)
Matt
Tony Cooper
I don't think dwarf tossing is a televised event. My cable programming
carries Fox, ESPN, ESPNII, and MTV. I've never seen a dwarf tossing
event listed. I did see a re-run of "Taxi" where "Louie" was knocked
down, but not knocked for any distance.
I would imagine that the appeal of dwarf tossing is that it's a live
event and held in a venue where there is a great deal of beer
guzzling. I would expect to see tattooed women with mostly-exposed
breasts in the crowd, and this would be of more interest to me than
the aerial antics of dwarves. I appreciate art in all forms and feel
that ink-on-skin is an underappreciated genre.
I am tempted to follow your lead and hyphenate "dwarf tossing", but
I'm a bit unsure. I don't think I'd hyphenate "ball tossing" or
"stick tossing", and the only thing different about those phrases is
the nature of that which is tossed.
I've never really grasped the rules about hyphens. I can spot the
obvious need when the result is ambiguity of meaning in the sentence,
but I've never noted an obvious need when ambiguity is not a risk.
I am also prompted to wonder if there is an American style and an
Other-Pondian style of hyphenation. If not, it's one of the few areas
of agreement in use of punctuation.
Drifting a bit...Is dwarf tossing a sport? Or, is it like golf and
auto racing where there are people who insist it is and people who
insist it isn't?
Paul Wolff
<tony_co...@earthlink.net> writes
>Drifting a bit...Is dwarf tossing a sport?
Tiddlywinks, you mean?
Some players have a wooden squidger, but this is mostly for
show.
The scoring:
Winks in a pot are worth three tiddlies.
Those on the mat and not squopped score one tiddly.
Squopped winks score no tiddlies.
Some named compound shots:
The Bristol shot (named after the university whose teams first
employed it on a regular basis) is played with a vertical
squidger and moves a wink and the one below it together.
The Boondock leaves the player's own wink hardly moved, but
sends an opponent's from underneath to a great distance.
The John Lennon is a Boondock where the player's own wink also
achieves a squop on a third wink.
A Gromp has the same result as a Bristol but is played with a
squop-type action.
Indented extracts all from:
http://www.bbc.co.uk/dna/h2g2/classic/A178634
>Or, is it like golf and auto racing where there are people who insist
>it is and people who insist it isn't?
Someone may care to devise a rule for discriminating between sports and
games[1], but the participants will claim whatever suits them at the
time. If it is true that the Cambridge University Tiddlywinks Club was
established in order to provide a niche sport for its founders to
challenge and play Oxford at, and thereby be entitled to be awarded
'Blues' (subject to the Committee's view), then of course it becomes a
sport.
[1] Here goes: improved performance in a sport is best acquired by
practice; improved performance in a game is best acquired by study.
That makes tiddly-tossing a sport.
--
Paul
In bocca al Lupo!
Salvatore Volatile
> I am tempted to follow your lead and hyphenate "dwarf tossing", but
> I'm a bit unsure. I don't think I'd hyphenate "ball tossing" or
> "stick tossing", and the only thing different about those phrases is
> the nature of that which is tossed.
>
> I've never really grasped the rules about hyphens. I can spot the
> obvious need when the result is ambiguity of meaning in the sentence,
> but I've never noted an obvious need when ambiguity is not a risk.
>
> I am also prompted to wonder if there is an American style and an
> Other-Pondian style of hyphenation.
Certainly there are differences. In general, Brits (and their Slavish
Imitators, e.g. Rob Bannister, Steve Hayes, etc., etc.) use hyphenation
more often than we 'Mericans do.
I also would write "dwarf tossing". WHRHR.
--
Salvatore Volatile
ref at freeshell dot org
Father Ignatius
news:drg7fi$182$1...@news.wss.yale.edu...
> Tony Cooper wrote:
>> I've never really grasped the rules about hyphens. I can spot the
You are not alone. Fowler remarks, "The chaos among writers [...] regarding
the use of hyphens is discreditable to English education. [...] it
sufficiently proves by its existence that neither the importance of proper
hyphening nor the way to set about it is commonly known [...]
After the death of Harold Ross, founding editor of _The New Yorker_, there
was concern that his successor, William Shawn, could cut the mustard. "Yet
for all his eccentricities Shawn proved (after a cautious start) to be a
heaven-born editor. [...] he once kept a writer up until 2.30 A.M. in a
dispute over a hyphen."
>> obvious need when the result is ambiguity of meaning in the sentence,
>> but I've never noted an obvious need when ambiguity is not a risk.
Examples of this include extra-marital sex vs. extra marital sex, and
superfluous hair-remover is a hair remover that no-one wants.
>> I am also prompted to wonder if there is an American style and an
>> Other-Pondian style of hyphenation.
Well, let us distinguish style from degree. There can only be one style: as
language develops, separate words become linked by usage. Typesetting
reflects this process first by joining the words with
a hyphen and ultimately by eliding them completely:
collar bone -> collar-bone -> collarbone.
Usage changes constantly, and American and British usages vary. Editing in
this area therefore calls for informed judgment; judgment is usually
informed by lists. When lists fail you, be informed by agreed reference
books (e.g. AHD4): if a particular combination appears, follow the
hyphenation or elision (or lack
thereof) that is given; if the combination does not appear, use separate,
unhyphenated words. If more than one variant is given, tend to prefer the
first.
[Remark: this particular list eschews infelicitous elisions (see below). It
also goes without saying that such lists express opinion, and are therefore
the source of much contention.]
A-frame
afterbirth
aftercare
after-image
afterworld
air hostess
air port
air raid
airlift
airline
airmail
airship
airsick
airspace
airstream
airtight
air-to-air
airworthy
aitchbone (not H-bone)
alfresco
A-line
A list
all clear
all fours
amuse-gueule
antiretroviral
any more
ape-shit
apropos
arm binder
armchair
armrest
au fait
au pair
autopilot
a while
baby-sit
back road
backpack
backstreet
bad-ass
bake sale
baldhead
barstool
bath-house
bathrobe
bathroom
bath water
beanbags
bear hug
bed frame
bed stand
bed sheets
bedsprings
bedtime
bee-line
bellboys
belly button
belly laugh
belt buckle
bench press (n)
bench-press (v)
billy goat
blind date (n)
bloodsucker
blood work
blowjob
boat-shed
bodice ripper
body blow
body-shaper
board games
boiler room
bolt-hole
bona fide
boob tube
book-keeper
bookmaker
box cutter
boyfriend
bra burner
bra-burning
bra-less
bra-strap
brand new
breadbasket
bride-price
broderie anglaise
bulldog
bullseye
bumbag (Anglicism)
busboy
bust line
buttonhole
buzzword
by the by
Bye-bye
by-election
cakewalk
callback
cameraman
campfire
career-oriented
car park
case study
catch phrase
cat-house
cat-o'-nine-tails
catsuit
cell phone
chain link
chat-room
checkbook
checkout (noun)
chest hair
chocolate-brown
clapboard
classmate
classroom
clean-shaven
cleanup (n)
clear-cut
closing time
clothesline
cobblestone
Coca-Cola
co-ed
cold cut
coffee shop
collarbone
con man
comb-over
corkscrew
cornhusk
corpsman
co-star
cotton candy
co-operate
country and western
co-worker
crackdown
crewcut
criss-cross
crop picking
crop sprayer
cross-dress
cross-eyed
cross-legged
cross-question
crosswalk
crow-bar
crow's feet
cue ball
cut-off (noun)
cut-out (noun)
cut-price (adj.)
cut-throat
daredevil
darkroom
de luxe
deadline
deadlock
deadpan
deathbed
death trap
death wish
deed poll
demi-bra
derail
desktop
Detective Sergeant
dining car
dining room
dining table
dishrag
dishtowel
dog collar
dog meat
doorbell
doorframe
doorknob
doorman
door-to-door
double check
double-entendre
double take
down side (opposite of ``up side'')
down stroke
drip-proof
drop-dead
dry cleaner
dumb-ass
dumb-bell
dust rag (but prefer duster)
duty-free
ear lobe
earing
eggnog
eggshell
email (now preferred but e-mail still acceptable)
en suite
expat[riate]
eye shadow
eyehole
eyeliner
eyeteeth
face down
fair-minded
fanny pack
fast track (n.)
fast-track (adj. vt.)
fat cat
feedback
fence-line
field goal
figure drawing
fingernail
fire escape
firefight
first aid
fishing line
fishnet
flat-bottomed
flatmate (British for room-mate)
floor show
footrest
fortune telling
four-poster
free and easy
French fry (n)
French-fry (v)
French kiss
full nelson
gang-bang
garlic bread
garter belt
glass-blower
god-daughter
goodbye
good-looking
goodnight (salutation and adjective)
goodwill
goose bumps
goose pimples
grand-daughter
G-spot
G-string
gutshot
guy-hug
hairgrip
hairline
hair-loss
hairpin
hairstyle
half a dozen
half-and-half
half-breed
half brother
half-dozen
half-erect
half-hearted
half-hour
half-light
half-mile
half-naked
half nelson
halfway
halter-top
hand basin
handcrafted
hand-held
handbag
handhold
handprint
handout
handrail
hangover (but hung over)
hanky-panky
hardball
hard-on
headboard
headpiece
head-rail
headrest
heads-up
heartbeat
heavyset
heeltap
hidebound
high school
high-pitched
hip flask
hitchhiking
hobbyhorse
homeboy
home-made
hometown
honey pot
horseshoe
horse-shoeing
hot-dog (noun)
hotdog (verb)
hot-house
hot-pants
hourglass
houseboy
house cleaning
hung over (but hangover)
ice age
ice axe
ice bag
ice bucket
ice blue
ice cream
ice cube
ice pack
ice pick
ice rink
iceberg
icebox
ice-cold
ice pack
inbox
inter-racial
jackhammer
jackrabbit
jaw line
jet set
jet stream
jet-black
jet-setter
job seekers
joy ride
keyholder
keyhole
kneehole
knee socks
knifepoint
lamp-post
lance corporal
left-handed
lifesaver
Life Savers
light bulb
like-minded
lip-gloss
lime juice
living room
locker drawer
long-leg
longneck [Bud]
lookout
lovemaking
lunchtime
mailroom
make-up
marine plywood
mast-head
mastermind
medevac
mid-day
middle finger
middleman
midnight
mis-shapen
mis-spell
mis-stated
mis-trial
money-grubbing
money-spinner
money's-worth
mortarboard
motherfucker
mother-of-pearl
motorbike
nametag
neckband
neoconservative
newfound
nightcap
nightclothes
nightshirt
nightstand
no man's land
no one
non-coms
non-native
notebook
notecase
notepad
notepaper
noteworthy
offhand
off kilter
off ramp
offline
old-fashioned
once-over
online
on-ramp
outer space
over-age (when of age)
over-analyze
overcook
over-corset
over-rate
over-react
over-ride
over-ripe
overshirt
overuse
paddleboard
pageboy
pant-suits
panty-girdle
part-time
partygoers
pay dirt
peace offering
peeping Tom
penthouse
phonecard
pièce de résistance
pina colada
ping-pong
pinstripe
pissant
play-acting
playlist
plum pudding (n.)
plum-pudding (adj.)
pocketknife
Police Constable
Police Special
ponytail
postman
postmark
postmaster
postmodernist
pot-head
power boat
power cut
power dive
power line
power plant
power station
pre-Byzantine
pre-dated (opp. of ante-dated cf predated ``preyed upon'')
pre-heat
pre-planned
preschool
printout
pro bono
rabbit brush
rabbit-hole
rag-head
R-and-R
re-admit
re-couple
re-create
redeye
redhead[ed]
re-elect
re-enter
re-examine
re-gather
re-learn
re-reading
re-spell
re-word
right angle
ringbolt
ring tone
road hog
roadhouse
roadside
role-play
role player
role-playing
roller coaster
room-mate
rotgut
rough-hewn
saddlebag
sadomasochistic
safeword
sailor suit
sangfroid
savoir-faire
schoolgirl
schoolwork
screen saver
scriptwriter
sea cucumber
sea spray
second floor (n.) second-floor (adj.)
secondhand (adj. and adv.) (US) vs. second-hand (Brit.)
second-rate
self-control
self-defense
self-pity
self-satisfied
semicolon
servomechanism
set piece
setup
sex appeal
sex change
sex kitten
sex maniac
sex object
sex symbol
sex-linked
shareholder
shirtsleeve
shirt-tails
shirtwaist
shirtwaister
shish kebab
shit-eating
shit-hole
shockproof
shoe buckle
shoe leather
shoehorn
shoelace
shop assistant
shop girl
shopfront
short-hand
short-sleeved
shoulder bag
shoulder blade
shoulder pad
shoulder strap
showboat
showerhead
side-by-side
side effect
sidekick (assistant)
side kick (karate)
sidewall
signpost
silver foil
singalong
sketchpad
skiing
skintight
skin tones
ski-wear
slapdash
slap-happy
sleazeball
sledgehammer
sleepwalk
slingback
slot machine
small-scale
small talk
small-time
smart-ass
snakeweed
Snow White
snow-white
soapsuds
Solicitor-General
sonic boom
soundman
spaceport
spearhead
spread-eagled
split second
spoon-feed
stable hand
staff sergeant
stagehand
stainless steel
status quo
stepfather (and similar relationships all one word)
stir-fry
stopgap
stoplight (red brake light)
stop light (red traffic light)
storybook
straight away
strait-laced
stranglehold
streetlights
strong-arm
sunblock
suspender belt
sweet talk (n)
sweet-talk (v)
swimsuits
swivel chair
table tennis
tabletop
Tae Kwon Do
tailgate
take-off
tank top
team building
teenyboppers
telltale
tenfold
ten-gallon hat
terra cotta
theme park
tiptoe
toenails
tollbooth
tomcat
tommyrot
tongue-lashing
toolmaker
tortoiseshell
towel-head
trail bike
transpacific
transsexual
trashcan
troop carrier
truck stop
T-shirt
two-pot
under appreciated
under way
un-erotic
under-rate
ugly-looking
under-educated
up-hill
up side (opposite of ``down side'')
upside down (upside-down when adj.)
upside-down cake
upstanding
uptight
vice president (US) vs. vice-president (Brit)
viceregal
vice versa
victor ludorum
videophone
videotape
viewport
viva voce
voice-over
volte-face
waistband
walking wounded
walking stick
washroom
watch chain
watch keeper
water buffalo
watercolor
webcam
web designer
webmaster
website
weekend
weightlifting
weight training
well being
well-defined|filled [noun]
well-doer
well-to-do
well trained
West Chester County (Delaware and Pennsylvania)
Westchester County (New York)
wet dream
wetsuit or wet suit
wheelchair
whiplash
whistle-blower
whitewash
wide-eyed
wife-swapping
will-o'-the-wisp
willy-nilly
windburn
windowsill
wing tips
wise guys
wish-washy
women's wear
workload
worktable
wristwatch
X-rated
X-ray.
> Certainly there are differences. In general, Brits (and their Slavish
> Imitators, e.g. Rob Bannister, Steve Hayes, etc., etc.) use hyphenation
> more often than we 'Mericans do.
COED11 points out, "...the evidence of modern English indicates a tendency
towards avoiding hyphenation in general, showing a preference for
_airstream_ rather than _air-stream_ and for _air raid_ rather than
_air-raid_. There is an additional tendency for the form to be one word in
US English and two words in British English, e.g. _coffee pot_ tends to be
the more common form in British English, while _coffeepot_ tends to be more
common in US English."
As a general rule, American English hyphenates and elides sooner and more
readily than British English. However, note some exceptions amongst the
differences listed below:
American,British
backup,back-up
bonbon,bon-bon
buildup,build-up
caseload,case load
cheekbone,cheek bone
coffeepot,coffee pot
crisscrossed,criss-crossed
double-entendre,double entendre
firefight,fire-fight
fox hunting,foxhunting
goose flesh,gooseflesh
half brother,half-brother
hipbone,hip bone
homemade,home-made
hometown,home town
ladylove,lady-love
lambchop,lamb chop
nonbreaking,non-breaking
nonnegotiable, non-negotiable
overanxious,over-anxious
overexercise,over-exercise
percent,per cent
pinup,pin-up
pocketknife,pocket knife
purebred,pure-bred
purse,handbag
reelect,re-elect
reenter,re-enter
roommate,room-mate
roundtable,round table
secondhand,second-hand
shirttail,shirt-tail
singsong,sing-song
sling-back,slingback
soul mate,soulmate
spellcheck,spell check
spongecake,sponge cake
stylesheet,style sheet
townhouse,town house
V sign,V-sign
vice president,vice-president
workday,working day
workweek,work week
As another general rule, though, American English is more ready to accept
infelicitous elisions than British English, which would tend to hyphenate in
such examples as:
apeshit
badass
bathhouse
beeline
boilerroom
bolthole
bookkeeper
braless
butthole
cathouse
chessset
coed
cooperate
costar
coworker
grabass
cropsprayer
cutthroat
doubleentry
dripproof
dumbass
dumbbell
fasttrack
goddaughter
granddaughter
halferect
homemade
hotdog (noun)
hothouse
highheeled
interracial
lamppost
lefthanded
loophole
masthead
midday
misshapen
misspell
misspent
misstated
mistrial
nonnative
nonnegotiable
noone
overage (when of age)
overrated
overreact
override
overripe
pantsuits
pothead
pothole
preByzantine
preplanned
rabbithole
readmit
recreate
reelect
reenter
reestablish
reexamine
reexplore
regather
relearn
rereading
respell
reword
roommate
shithole
shirttail
shorthand
skiwear
takeoff
undereducated
underrate
unerotic
uphill
weephole
--
Nat
"Not things, but opinions about things, trouble folk."
--Epictetus, _Manual_
Tony Cooper
<FatherI...@ANTISPAMananzi.co.za> wrote:
>"Salvatore Volatile" <m...@privacy.net> wrote in message
>news:drg7fi$182$1...@news.wss.yale.edu...
>> Tony Cooper wrote:
>
>>> I've never really grasped the rules about hyphens. I can spot the
>
>You are not alone. Fowler remarks, "The chaos among writers [...] regarding
>the use of hyphens is discreditable to English education. [...] it
>sufficiently proves by its existence that neither the importance of proper
>hyphening nor the way to set about it is commonly known [...]
(Interesting screen-fulls snipped)
I wish Charles was still around. Once again my point that the most
trivial of throw-away comments can result in a usage point has been
validated.
Father Ignatius
>
> As part of a university project I'm working on a piece of software
> with two candidate authors.
>
> It uses various statistical techniques in order to do this, and
> requires a large quantity of "training" text - ie works which are
> *known* to have been written by each author.
>
> I'm currently gathering together various works from different authors
> (from the Project Guttenberg web site) to act as training texts and
> test texts for experiments on the software. The works available are
> mostly pre-1923 due to copyright.
>
> In order to provide a challenge to the software in determining the
> correct author of a certain text, I need to test it with authors whose
> writing styles are similar. (Being able to distinguish between works
> by Chaucer and Dickens is all well and good, but it wouldn't be that
> impressive really!)
>
> I'm not an expert on literarature, so can anyone suggest any pairs of
> stylistically similar authors that I can challenge the software with?
> Guttenberg archives.
Hey, Matt: I just serendipitously stumbled on:
http://www.gutenberg.org/files/12478/12478-h/12478-h.htm#SHAKESPEARES_FINAL_PERIOD
which seems to suggest that you might compare Shakespeare with Shakespeare.
Maybe you software could thence infer an ordering to Shakespeare's plays?
--
Nat
"Second to the right, and straight on {till|'til} morning."
Peter Moylan
> On Thu, 26 Jan 2006 00:31:17 GMT, "Purl Gurl" <purl...@purlgurl.net>
> wrote:
[...]
> What / who is Roberta? Are you talking about a program?
Matt, it's only fair to warn you that the person to whom you're
responding has a reputation for talking crap. Most of the other people
in this thread are trustworthy.
--
Peter Moylan http://www.pmoylan.org
Please note the changed e-mail and web addresses. The domain
eepjm.newcastle.edu.au no longer exists.
My e-mail addresses at newcastle.edu.au will probably remain "live"
for a while, but then they will disappear without warning.
The optusnet address still has about 5 months of life left.
Peter Moylan
> Someome must watch all of those or they wouldn't be televised. Hell,
> people watch informercials. There's no accounting for taste.
Once, during a bout of insomnia, I discovered that at 4 a.m. I had a choice
between informercials and someone called Benny Hinn. (Whose audience, I
imagine, consists entirely of people who confuse him with Benny Hill.)
That gave me some understanding of why anyone would watch infomercials.
stev...@gmail.com
Links to all linds of porn on the net
http://www.freeadulthotels.com
Porn, links to porn sites
http://www.pornhardcoreporn.com/
http://midget.freakview.com/1260755420
http://www.14inch.com/1260755420
http://www.14inchdick.com/1260755420
http://www.18vision.com/1260755420
http://www.amateurtour.com/1260755420
http://www.americantranssexuals.com/1260755420
http://www.analbliss.com/1260755420
http://www.asianteenies.com/1260755420
http://www.assmasters.com/1260755420
http://www.bathbang.com/1260755420
http://www.bigpolestinyholes.com/1260755420
http://www.bigtitbabe.com/1260755420
http://www.bizarrefisting.com/1260755420
http://www.bjhotties.com/1260755420
http://www.blackonwhite.com/1260755420
http://www.blowqueen.com/1260755420
http://www.boys4girls.com/1260755420
http://www.camgang.com/1260755420
http://www.camshowsluts.com/1260755420
http://www.carrotfucking.com/1260755420
http://www.casarosso.com/1260755420
http://www.chubbylover.com/1260755420
http://www.cumbarn.com/1260755420
http://www.cumfreckles.com/1260755420
http://www.cummania.com/1260755420
http://www.drippyslits.com/1260755420
http://www.ebonyflics.com/1260755420
http://www.ebonyteenies.com/1260755420
http://www.eruptingpussy.com/1260755420
http://www.evilescorts.com/1260755420
http://www.exsluts.com/1260755420
http://www.farmgirls.com/1260755420
http://www.farmho.com/1260755420
http://www.femfacials.com/1260755420
http://www.freakpeep.com/1260755420
http://www.freakview.com/1260755420
http://www.freakyinsertions.com/1260755420
http://www.friskyfarms.com/1260755420
http://www.fuckfavors.com/1260755420
http://www.gothfuck.com/1260755420
http://www.grannyfantasy.com/1260755420
http://www.gspotfacials.com/1260755420
http://www.gushingfacial.com/1260755420
http://www.hardflics.com/1260755420
http://www.hornyebony.com/1260755420
http://www.hornysecretarys.com/1260755420
http://www.hotel69.com/1260755420
http://www.indianhussies.com/1260755420
http://www.interracialorgasms.com/1260755420
http://www.jackingteens.com/1260755420
http://www.latinoteenies.com/1260755420
http://www.lesbianpuss.com/1260755420
http://www.lesbofisters.com/1260755420
http://www.lesboprison.com/1260755420
http://www.localhoneys.com/1260755420
http://www.lonelyamateur.com/1260755420
http://www.love2pee.com/1260755420
http://www.milftease.com/1260755420
http://www.moretitties.com/1260755420
http://www.musicvideohos.com/1260755420
http://www.oldfucker.com/1260755420
http://www.orgydream.com/1260755420
http://www.peeclub.com/1260755420
http://www.peeteenagers.com/1260755420
http://www.pinkflicks.com/1260755420
http://www.prisonlesbians.com/1260755420
http://www.pussycramps.com/1260755420
http://www.rawmilfs.com/1260755420
http://www.realbjs.com/1260755420
http://www.sexsmoke.com/1260755420
http://www.sexxxylatinas.com/1260755420
http://www.sexxxymomma.com/1260755420
http://www.sexygushers.com/1260755420
http://www.shavingteens.com/1260755420
http://www.sleazyflicks.com/1260755420
http://www.soapingteens.com/1260755420
http://www.spunkfarm.com/1260755420
http://www.spunkforce.com/1260755420
http://www.squirtingfacials.com/1260755420
http://www.stripteenies.com/1260755420
http://www.superlesbians.com/1260755420
http://www.swingingteen.com/1260755420
http://www.teengushers.com/1260755420
http://www.titbanger.com/1260755420
http://www.titfiesta.com/1260755420
http://www.titsquirts.com/1260755420
http://www.tittyplus.com/1260755420
http://www.tokyobitches.com/1260755420
http://www.totalaccessporn.com/1260755420
http://www.trannyhookup.com/1260755420
http://www.trannyview.com/1260755420
http://www.trickortranny.com/1260755420
http://www.virginshemales.com/1260755420
http://www.vulgarteens.com/1260755420
http://www.wildflics.com/1260755420
http://www.xxxsecretions.com/1260755420
Hardcore porn
http://www.freeadulthotels.com
Porn, links to porn sites
http://www.pornhardcoreporn.com/
Steve Hayes
wrote:
>Hardcore porn
If you don't quote what you're replying to, how do we know what you're taling
about?
--
Steve Hayes from Tshwane, South Africa
http://www.geocities.com/Athens/7734/stevesig.htm
E-mail - see web page, or parse: shayes at dunelm full stop org full stop uk
the Omrud
> Robert Lieblich wrote:
>
> > Someome must watch all of those or they wouldn't be televised. Hell,
> > people watch informercials. There's no accounting for taste.
>
> Once, during a bout of insomnia, I discovered that at 4 a.m. I had a choice
> between informercials and someone called Benny Hinn. (Whose audience, I
> imagine, consists entirely of people who confuse him with Benny Hill.)
> That gave me some understanding of why anyone would watch infomercials.
He was here. If it weren't for AUE I would never have heard of him.
I only know that he was here (in Norwich) because the delightful
Kathy Sykes made a series of programmes about alternative medicines
and therapies, so she filmed his healing session at Norwich City.
Purl Gurl
> Matt B wrote:
> > Purl Gurl wrote:
> > What / who is Roberta? Are you talking about a program?
> Matt, it's only fair to warn you that the person to whom you're
> responding has a reputation for talking crap. Most of the other people
> in this thread are trustworthy.
You are "talking crap." Such a fool you make of yourself!
Purl Gurl
JF
<purl...@purlgurl.net> writes
I can see visible steam coming out of your ears.
Robert Lieblich
But you *do* have a reputation for talking crap, Ms. P.G. Unless
you've been ignoring many of the posts discussing you, you are well
aware of that. Nor are you trustworthy. Much of what you post is
sheer nonsense. So Peter speaks truth, not crap. You, on the other
hand ...
Know thyself.
--
The Barrister
Purl Gurl
Purl Gurl
Peter Duncanson
Interpreting "know" in the biblical sense: <chuckle>
--
Peter Duncanson
UK (posting from a.u.e)
Al in Dallas
You continually make the adolescent error of being cocksure of
yourself whether you know what you're talking about or not.
Hence, you possess your well earned reputation for talking crap.
Who the hell would choose perl to implement an AI, anyway?
Oh, someone who overestimates her own intelligence and
judgment--of course!
--
Al in St. Lou
Robert Lieblich
Fact or opinion?