[Corpora-List] Meta Question regarding sentiment/opinion/reputation was Re: Campagne d’étiquetage en sentiments de tweets et blogs

Alexander Osherenko osherenko at gmx.de
Mon Nov 19 15:51:14 CET 2012


I would differentiate between sentiment, opinion, reputation etc. not using a lexicon but orientate much more towards a general and intuitive psychological disposition. I remember it was a critical question in my research when I was looking for a corpus with strictly emotional texts to perform opinion mining. However, since there are only few of emotional corpora I needed another solution. Hence, I decided to use non-emotional corpora if I can believably map non-emotional details onto emotional labeling. For instance, I chose a corpus with product reviews that are labeled according to technical details such as camera speed etc. and map this labeling onto a customer's satisfaction with a product that is in my opinion reflects emotional labeling.

2012/11/19 Trevor Jenkins <trevor.jenkins at suneidesis.com>


> On 19 Nov 2012, at 12:44, <coralie.reutenauer at orange.com> wrote:
>
> Here is a web site for sentiment tagging of tweets and blogs, where you
> can analyze brand e-reputation :
>
>
> I've been pondering an issue with sentiment analysis and opinion mining
> that the above brings to a head. Are the three things: sentiment, opinion
> and reputation identical? Are we devaluing each or any of those descriptors
> by regarding them as one and same lumping them together like this?
>
> The aggregate analysis of individual statements can be classified and
> conclusions drawn on sentiment and opinion but is it then valid to consider
> that aggregate as reputation. Especially when not every one has e-tools
> available with which to express their personal view. A brand may have an
> entirely different reputation online from offline. To draw conclusions from
> online only surely is an argument from silence.
>
> I'm aware that my comment here may only be true for English rather than
> some dialect of it or for any other languages. For example the three
> concepts have different signifiers in certain signed languages, which match
> the different lexical realisations in English.
>
> Regards, Trevor.
>
> <>< Re: deemed!
>
>
> _______________________________________________
> UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora
> Corpora mailing list
> Corpora at uib.no
> http://mailman.uib.no/listinfo/corpora
>
>

-- Alexander Osherenko Dr. rer. nat, CEO and R&D <http://www.socioware.de/> -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 3686 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20121119/352a36e9/attachment.txt>



More information about the Corpora mailing list