Data visualization is a powerful tool when it comes to surveying vast amounts of content in a timely manner. While text-backed data tools come in several different formats, the two most prominent and perhaps most useful methods of data visualization are word clouds and usage graphs. For this lab I chose to use the Wordle and Tagxedo engines to analyze George W. Bush and Barack Obama’s State of the Union addresses. The engines work by generating word clouds which are groupings of words where the size of each word is based on its level of occurrence in the submitted text.
Before settling on developing my first clouds in Wordle, I took a look at Tagxedo which is another popular generator. Given my computer’s pre-existing software and my lack of chops when it comes to internet browser technology, I did not initially attempt to use it. After a brief consultation of Reddit, I was able to get the proper extensions installed and developed word clouds in both engines. Both Wordle and Tagxedo work in a fairly similar manner – they both build a generic looking cloud in seconds from user inputted text – however Tagxedo seemed to be more nuanced in terms of presentation and text accessibility. While Wordle was capable of chugging out a clean cut and simple word clouds for both Obama and Bush, Tagxedo offers users a much wider array of input methods from RSS feeds to twitter handles (something I attempted after developing word clouds). Another Tagxedo feature I really liked was the fact that it was easy to shape clouds into basic forms and the engine provides templates to do so. While I tried to take it one step further by creating word clouds in the shape of Bush and Obama‘s faces (which did not go as well as I had hoped). The potential of this feature not only provides the user with a more aesthetically pleasing result but a topical one as well. For a website based on a historical figure or setting, using a silhouette or a map of the subject matter and developing a word cloud for its related texts could be a really neat topical addition.
While I had hoped to extract some deeper meaning from the word clouds in terms of themes in the Bush and Obama presidencies, the word cloud generators focused in on the larger buzzwords that are found in almost every state of the union such as America, American, and Americans. That being by taking a look at some of the secondary and tertiary sized words in both clouds I was able to extract certain themes and subjects from each presidency. For Bush a few words that jumped out at me instantly were “terrorism”, “freedom”, “world”, and “security”. For Obama I noticed “jobs”, “businesses”, “economy”, and “energy. Unsurprisingly, Obama’s cloud displayed “Republicans” whereas Bush’s did not display “Democrats” – I attribute this to the divisive nature of Congress during Obama’s presidency. However, because of the nature of word clouds it’s important to remember that their purpose is to provide a survey of relevant issues instead of providing a platform to develop contextual understanding.
In conclusion both Wordle and Tagxedo are effective engines that could augment a website or any other form of digital teaching materials, however they could definitely be improved by allowing for specific filters especially in the case of texts that have a somewhat standardized format like the annual State of the Union. Augmented filter settings could help word clouds be a more accurate gauge of the subject matter of each text. Furthermore, enhanced settings could definitely help make word clouds more user friendly (as I mentioned earlier, I had to look to second-tier size words to find the heart of each set of addresses’ subject matter).