Thread Closed 
Week 11:Twitter Chatter During the Super Bowl
11-06-2014, 01:41 AM (This post was last modified: 11-06-2014 01:53 AM by u0875378.)
Post: #1
Wink Week 11:Twitter Chatter During the Super Bowl
The visualization:
http://www.nytimes.com/interactive/2009/...itter.html
   

Description:
This visualization shows the location and frequency of commonly used words on Twitter about a game between Steelers and Cardinals in Super Bowl. In this visualization there is a map in which we can find the most frequently used words on twitter by location, game timeline and words category. By moving the mouse on the map we can see details of top words in a specific place. By moving the icon on the game timeline axis we an see top words during different period in the game, and we can also see the scores of both teams at the corresponding time. By clicking on the frames on the upper left corner we can choose different words categories to specify certain kinds of words we want to see.

Things I liked about this visualization:
The whole visualization is clean and straightforward, which makes it easy to use.
The usage of different hues to represent categorical words makes the visualization distinguishable.
The size of the words based on the frequency highlights import words and makes it easy to tell different frequency of words across different locations.
By clicking any place on the map, it shows a detailed description of the frequently used words near a specific city according to the place you click on.
The game timeline axis enables users to choose a specific time to see the words at that time, and the timeline is divided according to the football game period, which is clear and convenient for football fans to use.
Different words categories for users to choose makes the visualization more effective, which can be utilized by users to choose the specific category of words they want to check on, and get information they need more quickly.

Things I disliked about this visualization:
Words on the map are distinguished by size and colors, but only two colors(red and black) are used in it, which means some different words share the same color. And for some categories of words only black color is used. I think it would be better that more colors are used to represent different words or emotions, which will make it easy to tell different words apart on the map.
The vis just showed frequency of commonly used words, but the proportion of the words is not showed. I think providing a ratio based on the proportion of a specific words in all the twitted words will form a comparison for readers and provide more information.
I also think there should be a overall frequency of words for the whole country, which is lacked in the vis.

Overall I like this visualization very much because I think it is an effective visualization with interesting contents, and is very easy and enjoyable to use, though there are some little flaws which needs more improvement.
Find all posts by this user
11-06-2014, 11:32 AM
Post: #2
RE: Week 11:Twitter Chatter During the Super Bowl
I like this visualization and I actually laughed when I saw the explosion of players' names. This is a very vivid and attractive vis, and the interactive legend at the left allowing the users to choose the topic displaying, it make the graph very efficient.
(11-06-2014 01:41 AM)u0875378 Wrote:  Words on the map are distinguished by size and colors, but only two colors(red and black) are used in it, which means some different words share the same color.
From my understanding, I think the black and red color representing the two teams. By using these two colors, it's easy for the users to tell what team are most of the people talking about. I think it's a good way to represent the data for Superbowl because personally I want to know which team people are talking about more as well. So I don't think it's a problem.

Another thing I think there is still room for improvement is the transmission between different words. In this vis, the designer just let the word appear and disappear suddenly, making the whole vis kind of interrupted by the changes of the keywords. I'm thinking maybe by introducing some kind of animation between different words (eg, slowly disappear, or zoom-in-zoom-out effect like in powerpoint), the vis can becomes more attractive.
Find all posts by this user
11-06-2014, 02:34 PM
Post: #3
RE: Week 11:Twitter Chatter During the Super Bowl
Yes, in "steelers vs cardinals" and "player names" categories black and red color represent two teams and makes good sense. But in other words categories only black color is used, for which I think more colors used would make it better. Thanks for your proposal.

(11-06-2014 11:32 AM)yaoyao Wrote:  I like this visualization and I actually laughed when I saw the explosion of players' names. This is a very vivid and attractive vis, and the interactive legend at the left allowing the users to choose the topic displaying, it make the graph very efficient.
(11-06-2014 01:41 AM)u0875378 Wrote:  Words on the map are distinguished by size and colors, but only two colors(red and black) are used in it, which means some different words share the same color.
From my understanding, I think the black and red color representing the two teams. By using these two colors, it's easy for the users to tell what team are most of the people talking about. I think it's a good way to represent the data for Superbowl because personally I want to know which team people are talking about more as well. So I don't think it's a problem.

Another thing I think there is still room for improvement is the transmission between different words. In this vis, the designer just let the word appear and disappear suddenly, making the whole vis kind of interrupted by the changes of the keywords. I'm thinking maybe by introducing some kind of animation between different words (eg, slowly disappear, or zoom-in-zoom-out effect like in powerpoint), the vis can becomes more attractive.
Find all posts by this user
11-07-2014, 05:18 PM
Post: #4
RE: Week 11:Twitter Chatter During the Super Bowl
I think that all the pros and cons posted by the critic are accurate. I would like to add some points to it.
1. I feel the data represented in this visualization is incomplete, or the basic assumptions regarding the data have not been provided. I would think that majority of the cities in USA will have people tweeting about the Superbowl, while only some of the cities have been represented in the map.
2. Some of the markers have only counts for Steelers or Cardinals mentioned. I think it would be better if the count for the other team also be mentioned, even if it is zero.
Find all posts by this user
11-07-2014, 06:14 PM
Post: #5
RE: Week 11:Twitter Chatter During the Super Bowl
Generally I agree with the poster that this is an effective visualization with interesting content and I like it. The vis displays the most frequent tweets word of each city on the geographic map and use word size to represent its frequency. One good feature about this visualization is that it implement some useful interaction and these interaction works very smoothly. You can use the time bar on the top to change timeline with the game, and display top tweet words of different time. You can also use the left frame to choose different topic categories and display that on the map. So people can check the tweet word frequency they care about.

However, there are two thing I don’t like about this visualization. First, I don’t like the idea of showing frequency using word size, especially when the frequency is large. The word will overlap with each other and it is hard to distinguish them apart. When moving the timeline to the each of the game, some word frequency even go too large that the word cover two or three state, and then how can you find what that word belong to just by watching the visualization. The second thing is that I think the author does not show the full data. Sometime the frequency of two words are close, and I think it should be better if showing them together.
Find all posts by this user
11-08-2014, 12:51 AM
Post: #6
RE: Week 11:Twitter Chatter During the Super Bowl
I agree with you that only some of the cities have been represented, and many of them are small cities that I have not even heard before.

(11-07-2014 05:18 PM)u0869331 Wrote:  I think that all the pros and cons posted by the critic are accurate. I would like to add some points to it.
1. I feel the data represented in this visualization is incomplete, or the basic assumptions regarding the data have not been provided. I would think that majority of the cities in USA will have people tweeting about the Superbowl, while only some of the cities have been represented in the map.
2. Some of the markers have only counts for Steelers or Cardinals mentioned. I think it would be better if the count for the other team also be mentioned, even if it is zero.
Find all posts by this user
11-08-2014, 03:35 PM
Post: #7
RE: Week 11:Twitter Chatter During the Super Bowl
The idea of this vis is good, to show the people's focus and reaction during the the game. This can be used in advertisement or the analysis how the game is developing. But the words in the map have been showed too many and some of them are overlapped which is hard to tell. The words like "go", "wow", "ass" should be filtered. The map should can be zoomed in because the words in northeast are really massive.
Find all posts by this user
11-09-2014, 04:37 PM
Post: #8
RE: Week 11:Twitter Chatter During the Super Bowl
This visualization is pretty cool, i really like it, because it offer you the options about different event at the left top corner, and it add the timeline whcih can display, and in the map, the size of words means that frenquency of this word be used, and by display the timeline, i can know how the frequence of the words being used changed during the game. And what i dislike is that, some times, it looks massive, the words are overlap.
Find all posts by this user
Thread Closed 


Forum Jump:


User(s) browsing this thread: 1 Guest(s)