By using this site, you agree to our Privacy Policy and our Terms of Use. Close

Forums - General - Second Call: Gamers Wanted as Volunteers for a Research Study

DarkTemplar on 20 July 2013

Hi vgchartz members, would you like to help me in one of the experiments of my PhD? 

During the experiment you will have to judge the relevance of 10 different opinions, it will take around 10 minutes to perform.

In the first call the volunteers managed to judge the relevance of more than 150 opinions but our goal is to have more than 500 judgments. Hence we are now inviting those that did not participate or did not see the first call to help us in the experiment. On a side note, we would appreciate if you could do the poll more than one time (selecting a new game or even a game you selected previously).

Below is the formal Invitation and a mini F.A.Q.:

-------------------------------------------------------------------------------------------------------------------------------------------------

Dear vgchartz users:

You are invited to participate to a web based research study to investigate what make an opinion relevant to an online social network member. In this experiment you must judge the relevance of opinions from reviews written by other member of this site in the past. In order to respect and protect your anonymity this experiment was evaluated and approved by an Ethical Committee in Research at the University of São Paulo. The experiment was also discussed with two moderators of the forums to avoid any problems.

To participate on this research visit the link: http://allanlima.pcs.usp.br/vgpoll/terms/pollTerms.php

Thank you for considering this request.
Sincerely,
Allan Lima - PhD Student at the University of São Paulo.
P.S. If you wrote a public review on the vgchartz but do not want to have none of your opinions judged by other users in this experiment visit the link: http://allanlima.pcs.usp.br/vgpoll/blacklist/userBlacklistForm.php


Around the Network
DarkTemplar on 20 July 2013

Mini F.A.Q.:

What I have to do in this experiment?
You just need to judge the relevance of 10 opinions.
Who is behind this experiment?
We are researchers at the University of São Paulo, we work at the Polytechnic School specifically at the Systems Engineering and Digital Systems Department.

How will you protect my privacy/anonymity?
We will not store any personal information that could be used to identify you (such as your real name), we ask for you username during the experiment so we can know things like the games you have, but we do not intent to store it in its original form (for instance, in your database each username will changed for a random generated fake username like “X102hG023”). We suggest you to read our poll terms in order better understand it.


DarkTemplar on 20 July 2013
RolStoppable said:

I did it for two games and I often had to click on "Not an opinion", because factual statements were highlighted. Seems like there's a random portion of review text taken and then a random sentence highlighted, because sometimes I got the same portion with a different sentence highlighted.

EDIT: Also, quite a few review texts were incoherent.

Thanks, Rol.

I appreciate very much your help.

The opinions showed to you are randomly selected from our database. What is shown to you are the opinions (the highlighted text) what comes before and after it. What probably happened is that the games you choosed did not have many opinions this is because the highlighted part of the text sometimes apears before or after other opinions.

We used an opinion classifier to ensure that most of the text displayed during the experiment are in fact opinions, however since there no way to create a perfect opinion classifier we also offer the opinion to the volunteers replace a sentence he thinks that is an opinions by another one from our database.

Finally, "incoherent" texts are expected the be in our database and it is important to volunteers to judge as well because our I.A. algorhitms can learn what people like you think is "incoherent".

 

 



DarkTemplar on 20 July 2013
RolStoppable said:
DarkTemplar said:

Thanks, Rol.

I appreciate very much your help.

The opinions showed to you are randomly selected from our database. What is shown to you are the opinions (the highlighted text) what comes before and after it. What probably happened is that the games you choosed did not have many opinions this is because the highlighted part of the text sometimes apears before or after other opinions.

We used an opinion classifier to ensure that most of the text displayed during the experiment are in fact opinions, however since there no way to create a perfect opinion classifier we also offer the opinion to the volunteers replace a sentence he thinks that is an opinions by another one from our database.

Finally, "incoherent" texts are expected the be in our database and it is important to volunteers to judge as well because our I.A. algorhitms can learn what people like you think is "incoherent".

One thing I wonder about is how you will judge what it was that made any given opinion relevant to me.

That is essentialy what my PhD is.

We developed many ways to find it out.

For instance, one of the heuristics we use to know if an opinion is relevant or not to you is based on the similarity between you and the author of the opinion.

With this experiment we aim to gather enough data so we can learn (using A.I. algorithms) how important each of the heuristics we created is.



KHlover on 20 July 2013

I'd love to participate, but I'm not 18 yet :/
Will this still be up at 11th August?



Some useful VGChartz tweaks

Member of the "An Unbiased Review" Database team

Around the Network
DarkTemplar on 20 July 2013
KHlover said:
I'd love to participate, but I'm not 18 yet :/
Will this still be up at 11th August?

What a pity!

Well, I do not think the experiment in going to be still running at that time, but if we need to keep it online untill 11th August would you mind if we contact you by a PM to perform our experiment?



KHlover on 21 July 2013
DarkTemplar said:
KHlover said:
I'd love to participate, but I'm not 18 yet :/
Will this still be up at 11th August?

What a pity!

Well, I do not think the experiment in going to be still running at that time, but if we need to keep it online untill 11th August would you mind if we contact you by a PM to perform our experiment?

Not at all :)



Some useful VGChartz tweaks

Member of the "An Unbiased Review" Database team

DarkTemplar on 21 July 2013
KHlover said:
DarkTemplar said:
KHlover said:
I'd love to participate, but I'm not 18 yet :/
Will this still be up at 11th August?

What a pity!

Well, I do not think the experiment in going to be still running at that time, but if we need to keep it online untill 11th August would you mind if we contact you by a PM to perform our experiment?

Not at all :)

Thank you KHlover.

I will add you to list of the volunteers that we intend to contact to perfome a few more judgements if we do not meet the target of at least 500 opinions judged.