You're quite right, I didn't spot the second demo! Well, it's a fair cop then...<div><br></div><div>I just tried "I'm fairly unimpressed" and got the result "neutral". It's clearly not a British English sentiment detector then.</div>
<div><br></div><div>Mandy Schiffrin,</div><div>Researcher, RightNow Technologies</div><div><br></div><div><br><br><div class="gmail_quote">On 16 December 2011 14:31, Justin Washtell <span dir="ltr"><<a href="mailto:lec3jrw@leeds.ac.uk">lec3jrw@leeds.ac.uk</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Not quite so. The instructions you quote are for the first demo, and I followed these.<br>
The second demo is designed to deal specifically with "short texts", reportedly relying on parsing rather than statistics.<br>
<div class="im"><br>
Justin Washtell<br>
University of Leeds<br>
________________________________________<br>
</div>From: Amanda Schiffrin [<a href="mailto:a.schiffrin@gmail.com">a.schiffrin@gmail.com</a>]<br>
Sent: 16 December 2011 13:20<br>
To: Justin Washtell<br>
Cc: Alexander Osherenko; <a href="mailto:Corpora@uib.no">Corpora@uib.no</a><br>
Subject: Re: [Corpora-List] EmoText - Software for opinion mining and lexical affect sensing<br>
<div class="im"><br>
To be fair, the description does say: "For reliable result classified texts should contain not less than 200 words."<br>
<br>
Mandy Schiffrin<br>
Researcher, RightNow Technologies<br>
<br>
<br>
<br>
</div><div class="im">On 16 December 2011 14:06, Justin Washtell <<a href="mailto:lec3jrw@leeds.ac.uk">lec3jrw@leeds.ac.uk</a><mailto:<a href="mailto:lec3jrw@leeds.ac.uk">lec3jrw@leeds.ac.uk</a>>> wrote:<br>
Hello Alexander,<br>
<br>
I tried both of your demos out of interest.<br>
<br>
For the first demo I used the default options (the movie reviews and Naive Bayes). I did not understand the output, or how it was supposed to relate to the various parts of the input (if indeed it is?)<br>
<br>
For the second demo I entered the following sentences and received the following classifications:<br>
<br>
This demo is terrible. low_neg<br>
This demo is no good at all. high_pos<br>
This demo is far from brilliant. low_pos<br>
<br>
This demo is excellent. low_pos<br>
This demo is not bad at all. low_pos<br>
Thid demo couldn't be better! low_neg<br>
<br>
Am I doing something wrong?<br>
<br>
I would presently dispute your claimed "undisputable advantages". I am not sure whether your intended customers - who are presumably not language technology experts - will require less or more convincing.<br>
<br>
Justin Washtell<br>
University of Leeds<br>
<br>
________________________________________<br>
</div>From: <a href="mailto:corpora-bounces@uib.no">corpora-bounces@uib.no</a><mailto:<a href="mailto:corpora-bounces@uib.no">corpora-bounces@uib.no</a>> [<a href="mailto:corpora-bounces@uib.no">corpora-bounces@uib.no</a><mailto:<a href="mailto:corpora-bounces@uib.no">corpora-bounces@uib.no</a>>] On Behalf Of Alexander Osherenko [<a href="mailto:osherenko@gmx.de">osherenko@gmx.de</a><mailto:<a href="mailto:osherenko@gmx.de">osherenko@gmx.de</a>>]<br>
<div class="im">Sent: 16 December 2011 08:46<br>
</div>To: <a href="mailto:Corpora@uib.no">Corpora@uib.no</a><mailto:<a href="mailto:Corpora@uib.no">Corpora@uib.no</a>><br>
<div class="im">Subject: [Corpora-List] EmoText - Software for opinion mining and lexical affect sensing<br>
<br>
Dear all!<br>
<br>
Recently I made an announcement of a book about opinion mining and lexical affect sensing. In this contribution I would like to point you to the EmoText demo program that relies on the findings in this book. It was implemented for the European CALLAS project.<br>
<br>
The link is:<br>
</div><a href="http://www.socioware.de/products.html" target="_blank">www.socioware.de/products.html</a><<a href="http://www.socioware.de/products.html" target="_blank">http://www.socioware.de/products.html</a>><<a href="http://www.socioware.de/products.html" target="_blank">http://www.socioware.de/products.html</a>>.<br>
<div class="im"><br>
I apologize for some advertising.<br>
<br>
Kind regards<br>
Alexander Osherenko<br>
<br>
_______________________________________________<br>
UNSUBSCRIBE from this page: <a href="http://mailman.uib.no/options/corpora" target="_blank">http://mailman.uib.no/options/corpora</a><br>
Corpora mailing list<br>
</div><a href="mailto:Corpora@uib.no">Corpora@uib.no</a><mailto:<a href="mailto:Corpora@uib.no">Corpora@uib.no</a>><br>
<a href="http://mailman.uib.no/listinfo/corpora" target="_blank">http://mailman.uib.no/listinfo/corpora</a><br>
<br>
</blockquote></div><br></div>