I have a couple of crackers here. In case you did not know, a word cloud is... well, it is a cloud of words. I know, it sounds a bit silly, but it can actually be rather nice. more importantly though, word clouds are usually organised in such a way that the size of an individual word reflects the frequency with which that word is used. For example, the following word cloud uses the text of the American Declaration of Independence:
The word 'people' is the largest, as it occurs 10 times in the text. 'Free' (bottom middle) is a lot smaller, as it only appears 4 times in the text.
There are two kinds of word could which you can create. The first - and the most straightforward - is a cloud of words in which the words themselves are randomly put together in order to create a pattern. One of the best resources for creating these is Wordle.
Wordle is a really simply little website, which offers a neat range of options from the incredibly simple to the rather sophisticated. At its simplest, all you need to do is paste in a bunch of text and click on 'Go'! You can paste a web address as well, if you would prefer. The result will be spewed out from a Java page, in which you have the chance to modify the cloud by changing the font, the colours and the layout to suit your purposes.
If you are feeling brave, there is an 'Advanced' option, in which you can use some really quite simply programming markers to pre-determine the size or colour of particular words. This is a useful function if you want to create a word cloud which emphasises certain key words or phrases - say, for example, you wanted to create a title page for a document called 'Happy Thoughts', then you can ensure that the words 'Happy' and 'Thoughts' stand out.
If Wordle seems just a little too boring for you, then Tagxedo offers something a little bit different. As with Wordle, Tagxedo is Java based and you can paste your text / URL (or Twitter account) directly into it. What Tagxedo offers which is different, is the ability to shape your word cloud into all sorts of interesting patterns. You can choose these patters from a pre-determined list, or you can upload your own. For example, I uploaded an official Twitter logo in order to create a word cloud of my own Twitter account:
Again, this is a rather nifty option for presentations. You will see that I have used the word cloud above in a Prezi presentation (more on that later), and certainly with Wordle this has the advantage that you can upload the image as a PDF vector for near-infinite zoomability. It is certainly another interesting way in which to present words in an attractive format.
Perhaps a little more seriously, word clouds are actually an extremely good way of visualising word frequency data. Alberto Pérez García-Plaza, et al. have explored the effectiveness of 'tag clouds' in engaging people with information, while a good study by Yingca Wui, et al. has demonstrated how effective word clouds can be for both analysing and comparing word frequency data. Certainly I have seen one student use word clouds to good effect in their linguistics dissertation - presenting word frequency information with great clarity (although it is always good to back up cloud images with tables so that that visual impressions can be verified by the actual data).
Warning: When using word clouds for this more serious kind of use, do remember not to get too fancy. When you are using word clouds to present data, the important thing is that that data is clear.
There are two kinds of word could which you can create. The first - and the most straightforward - is a cloud of words in which the words themselves are randomly put together in order to create a pattern. One of the best resources for creating these is Wordle.
Wordle
Wordle is a really simply little website, which offers a neat range of options from the incredibly simple to the rather sophisticated. At its simplest, all you need to do is paste in a bunch of text and click on 'Go'! You can paste a web address as well, if you would prefer. The result will be spewed out from a Java page, in which you have the chance to modify the cloud by changing the font, the colours and the layout to suit your purposes.
If you are feeling brave, there is an 'Advanced' option, in which you can use some really quite simply programming markers to pre-determine the size or colour of particular words. This is a useful function if you want to create a word cloud which emphasises certain key words or phrases - say, for example, you wanted to create a title page for a document called 'Happy Thoughts', then you can ensure that the words 'Happy' and 'Thoughts' stand out.
Tagxedo
If Wordle seems just a little too boring for you, then Tagxedo offers something a little bit different. As with Wordle, Tagxedo is Java based and you can paste your text / URL (or Twitter account) directly into it. What Tagxedo offers which is different, is the ability to shape your word cloud into all sorts of interesting patterns. You can choose these patters from a pre-determined list, or you can upload your own. For example, I uploaded an official Twitter logo in order to create a word cloud of my own Twitter account:
Why might you use it?
Again, this is a rather nifty option for presentations. You will see that I have used the word cloud above in a Prezi presentation (more on that later), and certainly with Wordle this has the advantage that you can upload the image as a PDF vector for near-infinite zoomability. It is certainly another interesting way in which to present words in an attractive format.
Perhaps a little more seriously, word clouds are actually an extremely good way of visualising word frequency data. Alberto Pérez García-Plaza, et al. have explored the effectiveness of 'tag clouds' in engaging people with information, while a good study by Yingca Wui, et al. has demonstrated how effective word clouds can be for both analysing and comparing word frequency data. Certainly I have seen one student use word clouds to good effect in their linguistics dissertation - presenting word frequency information with great clarity (although it is always good to back up cloud images with tables so that that visual impressions can be verified by the actual data).
Warning: When using word clouds for this more serious kind of use, do remember not to get too fancy. When you are using word clouds to present data, the important thing is that that data is clear.
Bibliography:
Alberto Pérez García-Plaza, Arkaitz Zubiaga, Víctor Fresno, and Raquel Martínez, ‘Reorganizing Clouds: A Study on Tag Clustering and Evaluation’, Expert Systems With Applications, 39 (2012), 9483–9493
Wu, Yingcai, Thomas Provan, Furu Wei, Shixia Liu, and Kwan-Liu Ma, ‘Semantic-Preserving Word Clouds by Seam Carving’, Computer Graphics Forum, 30 (2011), 741–750
Comments
Post a Comment