Pushing the Score using Twitterdata: Difference between revisions

From XPUB & Lens-Based wiki
No edit summary
 
(21 intermediate revisions by the same user not shown)
Line 1: Line 1:
===Dutch elections 2017===
<h1>Sonification of the Dutch elections 2017</h1>
In the run up to the 2017 general elections in the Netherlands on Wednesday, 15 March 2017 to elect all 150 members of the House of Representatives, a lot of Twitter <ref>Twitter is an online news and social networking service where users post and interact with messages, "tweets," restricted to 140 characters, (http://twitter.com).</ref> traffic is generated.


Since 2002, every cabinet has resigned before completing their full four-year term and five elections were held between 2002 and 2012. The 2012 elections saw the Labour Party (PvdA) and People's Party for Freedom and Democracy (VVD) go head-to-head for the position of prime minister, gathering enough seats in the process to form an absolute majority. The VVD's Mark Rutte formed a coalition government with the PvdA, ousting the Christian Democratic Appeal (CDA) from government, while the Party for Freedom (PVV) went back to full opposition. There are about a dozen political parties taking part in the election due to the fragmented nature of Dutch politics.
<p>In the run up to the 2017 general elections in the Netherlands last Wednesday, the 15 of March 2017, a lot of Twitter<ref>Twitter</ref>  traffic was generated. Literally millions of tweets were send that day over the Internet.</p>


[[File:Data9673248-ac8aea.jpg=250px|*Illustratie Hajo in NRC d.d.27 februari 2017 *]]
<br>
<h1>The online political battle</h1>


===The online political battle===
<blockquote>
<blockquote>
"En meteen is daar de tweet: #waarwasBuma" <ref>En meteen is daar de tweet: #waarwasBuma, De politieke strijd online, door Andreas Kouwenhoven & Hugo Logtenberg, 27 februari 2017, 21:05 (https://www.nrc.nl/nieuws/2017/02/27/en-meteen-is-daar-de-tweet-waarwasbuma-7033073-a1547979).
"En meteen is daar de tweet: #waarwasBuma" <ref>En meteen is daar de tweet: #waarwasBuma, De politieke strijd online, door Andreas Kouwenhoven & Hugo Logtenberg, 27 februari 2017, 21:05 (https://www.nrc.nl/nieuws/2017/02/27/en-meteen-is-daar-de-tweet-waarwasbuma-7033073-a1547979).
Line 12: Line 11:
</blockquote>
</blockquote>


This Special Issue #2 is inspired an article in the NRC about this online political battle.  
<p>This research is inspired on an article in the NRC (27 februari 2017) about the online political battle. Just like in the United States, parties attempt to reach voters through social media and to frame political opponents. The parties use social media more than ever this year’s elections. They try to convey their message to the unprecedented number of undecided voters and they try to frame opponents with catchy slogans or hashtags.</p>
Just like in the United States, parties attempt to reach voters through social media and to frame political opponents. In television debates, such as RTL on Sunday evening, Twitter proves to be extremely popular. Especially with insiders.


The parties use social media more than ever this year’s elections. They try to convey their message to the unprecedented number of undecided voters and they try to frame opponents with catchy slogans or hashtags. During election debates, according to research in 2015 from the University of Amsterdam, the parties especially make use of Twitter - the medium is ideally suited to respond quickly.
<p>Seeing all these tweets passing so quickly it reminds me of a cascade of data, the starting point of a sonification<ref>sonification</ref> to perceptualize this huge amount of data in a <em>score</em>.</p>
<br>


Politicians use Twitter, in the Netherlands a daily log in around 900,000 people, primarily to attract attention, says Niels Spierings, political sociologist at the Radboud University.
<h1>Pushing the score</h1>
<blockquote>
 
"What you want, is that around your party leader a positive buzz is begin created."
<p>This sonification entails three kinds of scores; a) for the whole tweet, b) for the hashtag ikstem (#ikstem), and c) for the hashtag gestemd (#gestemd).</p>
</blockquote>


If one party succeeds in that there is positive twittered about her party leader, or negative about the competitor, this could have great impact according to Spierings.  
<p>a) It captures tweets sed during the day of the elections, between 07.30 and 21.00, when the polling stations were open. In total there are 47613 tweets captured. Some examples of tweets send:</p>


<blockquote>
<blockquote>
"Twitter will affect how the debate is interpreted afterwards, if only because the journalists who describe who is the winner and who is the loser, follow Twitter as one of their resources."
* <p>"Ik wist niet wat ik moest stemmen, dus heb ik uiteindelijk maar een bootje gevouwen van het stembiljet <strong>#ikstem</strong> #tk2017 <strong>#gestemd</strong>" https://t.co/KBqLBkYrpV</p>
</blockquote>
* <p>"Grappig! RT @Mvan_berkel: In Leiden is rekening gehouden met zwevende kiezers. <strong>#ikstem</strong> #TweedeKamerverkiezingen" https://t.co/Un8uJfNZ0v</p>
 
* <p>"Met volle trotst en vrolijkheid voor de eerste keer gaan stemmen vandaag <strong>#ikstem</strong>"</p></blockquote>
 
===Collecting data===
====Twitter API====
Made a Python script to get Twitter data from 26 elected parties.  
 
3200 latest tweets of 2 candidates (highest nominated man and woman) of 14 elected parties.
Elected parties: VVD, PvdA, PVV, SP, CDA, D66, ChristenUnie, GroenLinks, SGP, PvdD, 50Plus, DENK, Niet Stemmers, Piratenpartij based on their currents seats.
Looked at tweets with hashtag tk2017, tk17 and "verkiezingen". Furthermore I have collected all partijslogans of the above mentioned parties.


Used hashtags: #KUZU,#PlanB,#partijkartel,#pgb,#basisbeurs,#zorgfonds,#MakeTheNetherlandsGreatAgain,#MHGA,#GeertTrain,#Geertw,#NederlandWeerVanOns,#15maart2017,#cordonsanitaire,#wordpiraat,#stemvoorverandering,#peilingwijzer,#Oekraïne,oekraine,#factcheck,#trapernietin,#Fritswester,#grenzendicht,#50PLUSdus,#samenstaanwesterker,#JesseMeetup,#iksteundethuiszorg,#nomoreislam,#pakdemacht,#PVVOP1,operatieinterview,stemvoorhetleven,#normaaldoen,#stem15maart,#samenvooruit,#StemBus,#ikstempvda,#babbeltruc,#stemvnl,#lijst13,#wordlid,#stemvoornederland,#nlzorgfonds,#roemer,#nlzorgdebat,#stempvv,#pvv2017,#ikstemcda,#cda,#dienstplicht,#geefnldoor,#d66bustour,#ikstemd66,#stemd66,#hetkanwel,#nietschreeuwenmaarpraten,#grassroots,#beweging,#groenlinksbeltaan,#klaver,#stempvdd,#erisgeenplanb,#mariannethieme,#stem50plus,#50plus,#ikstemdenk,#piratenpartij,#ppnl,#jouwkeuze,#privacy,#stempiraat,#Fvd,#lijst16,#voorhetleven,#eenverdiener,#somszoujemeersgpwillen,#voordeander,#voorhetgezin,#geendrempels
<p>b + c) A hash tag is used to streamline relevant topics by keyword or phrase by grouping them together to make it easier to find and follow tweets from people who are talking about the same thing. In this sonificatoin I used #ikstem and #gestemd. Both are given a different kind of sound.</p>
<br>
<br>
<h1>Used hardware and software</h1>


===Plan of Approach===
*[http://news.bbc.co.uk/ Nieuwspagina BBC]
Need to look at endpoints for accounts and search terms. For accounts endpoint use: https://dev.twitter.com/rest/reference/get/statuses/user_timeline
<p><nowiki>[https://www.python.org/ Python] <nowiki>, specific libraries: <a href="https://www.python.org/dev/peps/pep-0305/">csv</a>, <a href="https://docs.python.org/2/library/os.html">os</a> and <a href="https://pypi.python.org/pypi/python-osc">OSC</a></p>
For search terms use the following endpoint: https://dev.twitter.com/rest/reference/get/search/tweets


Think about what I want to tell - -> music/sound plan --> give meaning to the score --> soundtrack
<p>[https://www.python.org Python] , specific libraries: [https://www.python.org/dev/peps/pep-0305/ csv], [https://docs.python.org/2/library/os.html os] and [https://pypi.python.org/pypi/python-osc OSC]</p>
Make connection with PD
<p>[https://puredata.info Pure Data]</p>
Sound research
<p>[https://pzwiki.wdka.nl/mediadesign/The_Ultimate_RPi_Installation_Guide RaspberryPi] and [https://pzwiki.wdka.nl/mediadesign/Pi_skin_conductivity Pi skin conductivity]</p>
<p>Flyer: [https://issue.xpub.nl/02/" https://issue.xpub.nl/02/]</p>
<p>[https://pzwiki.wdka.nl/mw-mediadesign/images/b/bf/Twittersonification.zip Used files]</p>
<br>


===References===
<h1>References</h1>

Latest revision as of 09:44, 4 April 2017

Sonification of the Dutch elections 2017

In the run up to the 2017 general elections in the Netherlands last Wednesday, the 15 of March 2017, a lot of Twitter[1] traffic was generated. Literally millions of tweets were send that day over the Internet.


The online political battle

"En meteen is daar de tweet: #waarwasBuma" [2] (And right after there was the tweet: #wherewasBuma)

This research is inspired on an article in the NRC (27 februari 2017) about the online political battle. Just like in the United States, parties attempt to reach voters through social media and to frame political opponents. The parties use social media more than ever this year’s elections. They try to convey their message to the unprecedented number of undecided voters and they try to frame opponents with catchy slogans or hashtags.

Seeing all these tweets passing so quickly it reminds me of a cascade of data, the starting point of a sonification[3] to perceptualize this huge amount of data in a score.


Pushing the score

This sonification entails three kinds of scores; a) for the whole tweet, b) for the hashtag ikstem (#ikstem), and c) for the hashtag gestemd (#gestemd).

a) It captures tweets sed during the day of the elections, between 07.30 and 21.00, when the polling stations were open. In total there are 47613 tweets captured. Some examples of tweets send:

  • "Ik wist niet wat ik moest stemmen, dus heb ik uiteindelijk maar een bootje gevouwen van het stembiljet #ikstem #tk2017 #gestemd" https://t.co/KBqLBkYrpV

  • "Grappig! RT @Mvan_berkel: In Leiden is rekening gehouden met zwevende kiezers. #ikstem #TweedeKamerverkiezingen" https://t.co/Un8uJfNZ0v

  • "Met volle trotst en vrolijkheid voor de eerste keer gaan stemmen vandaag #ikstem"

b + c) A hash tag is used to streamline relevant topics by keyword or phrase by grouping them together to make it easier to find and follow tweets from people who are talking about the same thing. In this sonificatoin I used #ikstem and #gestemd. Both are given a different kind of sound.



Used hardware and software

<nowiki>Python <nowiki>, specific libraries: <a href="https://www.python.org/dev/peps/pep-0305/">csv</a>, <a href="https://docs.python.org/2/library/os.html">os</a> and <a href="https://pypi.python.org/pypi/python-osc">OSC</a>

Python , specific libraries: csv, os and OSC

Pure Data

RaspberryPi and Pi skin conductivity

Flyer: " https://issue.xpub.nl/02/

Used files


References

  1. ↑ Twitter
  2. ↑ En meteen is daar de tweet: #waarwasBuma, De politieke strijd online, door Andreas Kouwenhoven & Hugo Logtenberg, 27 februari 2017, 21:05 (https://www.nrc.nl/nieuws/2017/02/27/en-meteen-is-daar-de-tweet-waarwasbuma-7033073-a1547979).
  3. ↑ sonification
Retrieved from "https://pzwiki.wdka.nl/mw-mediadesign/index.php?title=Pushing_the_Score_using_Twitterdata&oldid=105918"