[Corpora-List] corpus of rated words

[Oliver Mason]

A while ago I tried doing this automatically and ran into severe
problems during the training phase, mainly for two (related) reasons:

- words are not adequate units of meaning, so often you cannot decide
a rating for a single word.  Just as it makes no sense to ask for the
meaning of 'surgery', you can not come up with a general list of such
ratings, apart from a few obvious ones ('catastrophe', 'accident'
etc).  I found that it's easier to decide on the negative ones, but
most words I looked at I was not able to categorise.

- some words change their rating depending on context: for example,
'decline' sounds negative, but 'decline in road casualties' [Invented]
is good, 'decline in infant survival rates' [I] is bad, and 'decline
in profits' [I] depends on your personal view of the capitalist
system.

So I would conclude from my (unpublished) pilot study that it's a
futile exercise, but I'd be happy to be convinced of the opposite

Oliver

On 01/06/07, Claire Jessel <claire.jessel at uma.at> wrote:
> hello
> does anyone know of a word corpus in which all words would be rated as
> positive/neutral/negative (for instance, 'good' would have a positive
> rating, 'bad' a negative rating)?
> thanks
> claire
>
> --
>
> uma - separating the signal from the noise
>
> claire jessel . software engineer . claire.jessel at uma.at
> uma information technology GmbH . amerlingstrasse 1 . a-1060 vienna
> http://www.uma.at . phone +43-1-526 29 67-712 . fax +43-1-526 29 67-200
>
> --
> This message contains information which may be confidential and
> privileged. Unless you are the addressee (or authorised to receive
> for the addressee), you may not use, copy or disclose to anyone the
> message or any information contained in the message. If you have
> received the message in error, please advise the sender by reply
> e-mail @uma.at, and delete the message. Thank you very much.
>
>
>