Artículos
Tuesday, May 1, 2012, 08:21 AM
If you are a regular reader of this blog you probably know my position regarding predicting elections with Twitter. In case you don't know it: you cannot predict elections! [1] [2] [3].Tuesday, May 1, 2012, 08:21 AM
The thing is that time and again I read about the endless possibilities of Twitter regarding politics and, specially, electoral predictions.
Time and again I see the same papers cited; papers that, needless to say, claim that predictions are easy and absurdly accurate.
And time and again I see new papers stubbornly ignoring the (few) negative results showing that predictions are neither that easy nor possible with the current "state of the art" methods.
Therefore, I've reached my limits; I've felt that "someone's wrong on the Internet" moment and I've written a short survey on the topic.
More than a survey the paper is an annotated bibliography covering anything written on the topic of predicting elections with Twitter. Both positive and negative results are commented but more importantly I suggest what I see as key future lines of research.
The survey is available as a preprint at arXiv:
"I Wanted to Predict Elections with Twitter and all I got was this Lousy Paper" -- A Balanced Survey on Election Prediction using Twitter Data
Predicting X from Twitter is a popular fad within the Twitter research subculture. It seems both appealing and relatively easy. Among such kind of studies, electoral prediction is maybe the most attractive, and at this moment there is a growing body of literature on such a topic.
This is not only an interesting research problem but, above all, it is extremely difficult. However, most of the authors seem to be more interested in claiming positive results than in providing sound and reproducible methods.
It is also especially worrisome that many recent papers seem to only acknowledge those studies supporting the idea of Twitter predicting elections, instead of conducting a balanced literature review showing both sides of the matter.
After reading many of such papers I have decided to write such a survey myself. Hence, in this paper, every study relevant to the matter of electoral prediction using social media is commented.
From this review it can be concluded that the predictive power of Twitter regarding elections has been greatly exaggerated, and that hard research problems still lie ahead.
The first part of the title is deliberately provoking but I feel it's moment for the research community involved in this area to think twice before making flamboyant claims on print.
Needless to say, I'll be happy of hearing your comments and you can find me on Twitter: @PFCdgayo.
| enlace relacionado |
( 3 / 5 ) | TopMonday, October 17, 2011, 12:26 PM
God, grant me the serenity to accept the things I cannot change,
Courage to change the things I can,
And wisdom to know the difference.
It's the wisdom part that appeals me the most and, in fact, I've been struggling with something somewhat related since yesterday morning.
I was talking to a few students about the positive aspects of autodidacticism and pursuing "tangential knowledge", in contrast to merely adhering to the prestablished curriculum.
In the midst of that digression I warned them against "yak shaving" and since that moment I've been mulling it over:
- How can we know the difference?
- How can we distinguish between pursuing knowledge that can only be useful in the long time and yak shaving?
- Is there a threshold or, much likely, a continuum between valid (but of little use just now) interests and yak shaving?
Do you have any tip, trick, advise or heuristic to tell them apart? If you have it, please, tweet me.
| enlace relacionado |
( 3.1 / 79 ) | TopHOT!
Monday, September 26, 2011, 03:48 PM
More than a year ago I discussed on this blog why you cannot predict elections from Twitter (here and here).Monday, September 26, 2011, 03:48 PM
In the second of those posts I talked about a paper that, at that moment, was just submitted to a journal (actually a magazine). Today, I can announce it has been published by Communications of the ACM under the title "Don't Turn Social Media Into Another 'Literary Digest' Poll". (You can find author's version here).
During these months, some joint work was conducted with Takis Metaxas and Eni Mustafaraj: "Limits of Electoral Predictions Using Twitter".
Combining contents from the first paper with discussions held with Eni and Takis, and also during ICWSM 2011, I prepared an invited lecture for the people at Yahoo! Research in Barcelona that was held on July 21, 2011.
I think the slides are a good companion for the CACM paper and, hence, I've just uploaded them to Slideshare.
Enjoy and tweet me if you please (@PFCdgayo).
| enlace relacionado |
( 2.9 / 69 ) | TopThursday, July 21, 2011, 03:00 PM
There are lots of nice ideas but I think that when dealing with these issues we should think "bigger" instead on trying to "fix" what we have (although by thinking bigger for sure you are going to fix it). Hopefully, some kind of new Berners-Lee will appear somewhere with something new, fresh and brave.
In the meantime, I'll put together a number of tweets I published on september 2010. It's not a proposal, neither a position, just a dream but dreams are needed for thinking big:
- I've had a dream where I did not tell my friends and family to sign up for the same online services I'm using.
- I've had a dream where I did not chase information down, but information flew to me.
- I've had a dream where my thoughts were not broken in tiny pieces, but neatly organized though not centralized.
- I've had a dream where my information is MY information, not UGC subjected to TOS and EULAs.
- I've had a dream where standards were made to fit people, not people to adhere to standards.
- I've had a dream where there were only a box in which to write, no matter if I was searching or publishing. It simply KNEW.
- I've had a dream where I was able to be truly anonymous when I wanted to, but truly recognizable if I chose so.
- I've had a dream where social networks emerged from daily use, not from exchanging cards.
- I've had a dream where privacy is a natural by-product, not a clumsy patch.
As usual you can find me at Twitter.
| enlace relacionado |
( 3 / 88 ) | TopBreves
Wednesday, June 22, 2011, 12:18 AM
Wednesday, June 22, 2011, 12:18 AM
- Presentation software DOESN'T matter.
- Tell a story: your story.
- The main actor in your presentation is YOU, not your slides.
- Small bites are better (i.e. few text in each slide, enough to transmit a message).
- Corollary: the text in the slide should match your speech, if you provide more text they'll read and won't listen to you.
- Images are not a decoration, they should back up your message.
- Layout is key (or the Rule of thirds rules!)
- Be structured, both during your presentation and preparing it (do not start by opening your presentation software!)
- Don't stuff your audience but leave them a little hungry (e.g. don't tell everything, finish early).
- Do not use your last slide to say "thank you" but to show a remarkable conclusion of your presentation.
- Gadgets can be a presenter's best friend (e.g. a wireless presentation remote control).
- Practice, practice, practice (aka paying in sweat).
A student asked me about software for presentations because, apparently, he thought I was not using PowerPoint (indeed I use it). I told him a number of things (summarized above) and concluded telling him that PowerPoint is great provided you ignore 99% of the features.
Needless to say, I'm not inventing anything new here. Most of the points are extracted from the great books "Presentation Zen" or "Slide:ology" and the blogs from the corresponding authors. I learned the little trick in number 10 in "El Arte de Presentar" (in Spanish) and number 12 is, well, rather obvious but it's usually ignored by most of us.
| enlace relacionado |
( 3 / 141 ) | TopSiguiente