|
|
|
|
|
Data Visualizations
Using the R-Shief database, we code the data into various visual interpretations, including semantic web, sentiment, and content analysis. |
|
 by: Laila 314 views COMMENT(S) |
|
 by: bobboynton When I was working on the importance of Twitter in the inception of the revolts in Bahrain, Libya, Syria, and Yemen I found that retweets were unusually numerous. (Boynton, 3/9/2011) Sysomos found that only about six percent of all Twitter messages are retweets. In these countries the percentage of messages that were retweets was in the 60% range. That is an extraordinary difference. I argued that the form of the messages worked to constitute a we who could stand against the armed forces of the tyrant. The form of the message invokes a relationship between the person whose message I believe is important enough to be passing along and with the persons who are following me and therefore receive the message. The content may be about facing down the tyrant. The message may be about organizing the protest. The message may be commemorating a fallen hero. Whatever the message it is we who are participating in the communication. It is we who will act. I wanted to see if the use of retweeting was limited to these revolts so I looked at Wisconsin where another revolt was taking place against a governor and majority in the state legislature that were understood to be acting as tyrants. The use of retweet in those messages was 70%. As a shorthand it seemed 'we move' was an appropriate designation. Retweet is a we move. While following the #occupy movement I found the same. Retweets dominated Twitter messages. But there is variation. Some messages are retweeted only a few times and others are retweeted many times. I looked at how that could happen in "Retweeting in big numbers" (Boynton, 11/04/2011) The streams of Twitter messages provided by R-SHIEF lets me examine the extent to which retweets are being used broadly in the #occupy movement. The number of tweets per stream varies greatly. The smallest was 9 for OccupyAiIDS. The Largest was OWS with 5.28 million messages. Figures are not much help with such extreme ranges so to summarize I divided them into six approximately equal groups to show how long the tail is.  At the bottom of the distribution there is high variety. Of the 805 tweets containing the hashtag occupyTheSubway 97.8 are retweets. And occupySanta was a hashtag with 97 messages and only 15.4% are retweets. Since most of the messages are in segments 4 through 6, 12,374,154 compared to 51,403, those are the messages I will examine. The figure shows the percentage of Twitter messages that were retweets for the streams.  The range is from 85% to the outlier 18%. Only two streams fall below 50% of the messages in the stream. The average for the collected streams is 67.2%. The outlier is occupyTheInternet with 6155 posted messages only 18.5% of them are retweets. A we is called for in facing tyrants. One cannot do it alone. The #occupy movement is about changing the culture that has been built to facilitate the 1% dominating political and economic life. They use their wealth to dominate as other tyrants use guns to dominate. Retweeting for #occupy is constituting a we to take on the culture and the tyranny of the 1%. I am reporting to the Hackathon rather late on the last day. I spent most of Saturday and most of Sunday unzipping 120 files and doing a couple of simple computations. I am delighted to have the streams. It is a wonderful collection, and I will use it along with the streams that I have collected -- with acknowledgement to R-SHIEF. And I am looking forward to working with the "raw" files as soon as I get them unzipped. Now I will have to see if I can get this posted. Bob Boynton References Boynton, G. R. ( November 4, 2011) Retweeting in big numbers Boynton, G.R. (March 14, 2011) RT @bobboynton new media and the revolting middle east Sysmos (2010) Replies and Retweets on Twitter © G. R. Boynton, December 11, 2011 1,507 views COMMENT(S) |
|
by: Laila This network visualization was built using gephi:  Experimental work-in-progress of the same data: 955 views COMMENT(S) |
|
by: Laila At USC's School of Cinematic Arts, the Freshmen class are embarking on a remarkable learning experience. Check out Reality Starts Here. Below find visualizations of what the students tweeted during Dr. Henry Jenkins conversational presentation this afternoon. 917 views COMMENT(S) |
|
by: Laila 878 views COMMENT(S) |
|
by: Laila 797 views COMMENT(S) |
|
 by: salty Below are a set of visualizations of the twitter traffic after Breivik's attacks in Norway. This particular dataset includes all of the tweets with #norway, #oslo, #otoya and #blamethemuslims hashtags from the 22nd to the 25th of July.  The tweets have been visualized in Gephi as a bipartite graph (or network) meaning that they show two types of nodes: tweets and Twitter users. A user is connected to all tweets that they've sent and to all tweets that are directed at them (through the @username aspect of a tweet). Each image is an Ego Network of a particular user, showing the tweets that they've sent, any users connected to those tweets, and any tweets that those users have sent. Tweets are color-coded by timing such that the earliest tweets are lighter and the darker tweets are later (relative to the dataset). Further, any tweet that contains the word Christian are colored yellow, tweets that mention terror are colored fuchsia, tweets that mention Muslim or Islam are green and tweets that mention Muslim/Islam AND terror are teal (again, light to dark based on time). The graphs represent frequency of tweets with the listed words. All words searched are actually lemmas, so Islam includes Islamic/Islamist, terror includes terrorism/terrorist/etc.      Visualizations contributed by: https://dhs.stanford.edu/about/ 1,713 views COMMENT(S) |
|
by: Laila Looks like there the mainstream media generated dangerous Islamophobia buzz, not necessarily as loudly as the public sentiment. 1,144 views COMMENT(S) |
|
by: Laila In this interactive timeline of hashtags first commenting on the #oslo attacks, and the new hashtags that were generated from this global conversation on Twitter over the last 3 days. Click on the play button at the bottom to watch the comparison of hashtags tweeted over time. On the right side, you can click on the boxes of the hashtags you want to observe. And as you scroll, up and down that list of hashtags, you will see a word cloud of most frequent words that appeared accordingly. At the top of the graph, there are three views you can choose to view the data. The first is a scatter chart, the second is an interactive bar chart, and the third is a line graph. You can also choose, specifically, which language you might want to graph on both the x and y coordinates -- those are also interactive. Comment if you have any questions. 1,325 views COMMENT(S) |
|
by: Laila   1,273 views COMMENT(S) |
|
by: Laila   754 views COMMENT(S) |
|
by: Laila This graph represents a year of tweets archived and organized in R-Shief's databases. 772 views COMMENT(S) |
|
by: Laila This graph represents a year of tweets archived and organized in R-Shief's databases. 577 views COMMENT(S) |
|
by: Laila Analyzing how meaning is conveyed through Twitter. Looking for words most associated with each other through various conjunctions, "and, but", prepositions, verbs, and articles. Discovered that a big sentiment conveyed from this Twitter data is that "video is history." This is part of larger research initiative by VJ Um Amel. 1,083 views COMMENT(S) |
|
by: Laila Analyzing how meaning is conveyed through Twitter. In the #Syria tweets the word "massacre" sticks out. This is part of larger research initiative by VJ Um Amel. 404 views COMMENT(S) |
|
|
Top contributors
| |
 Laila | 15 |
| salty | 1 |
| MzMona | 1 |
| bobboynton | 1 |
Top Replies |
 #Sa3eedi Geeks Event in Luxor - 25 May 2012 | 0 |
Recent Activity |
| #Sa3eedi Geeks Event in Luxor - 25 May 2012 | 314 |