Visualisation Tools for Text Analysis

In this Week, we have learnt about how to analyze a text and count the frequency of words using different tools (http://www.wordle.net) (http://many-eyes.com) and (http://voyant-tools.org) to surf and stumble with the text. In These electronic web sites, we can upload a set of data or information to see which words are most often used and which are used the least in which this information will be analyzed and then displayed in different ways such as maps and charts. I have applied these three tools on the data created previously from Archive Twitter and Almetrics analysis.

Wordle,

Starting with Wordle, which is a tool to perform a basic level of text analysis and produce a word cloud that can be customised with different fonts, layouts, and colors. When I input the data created from Twitter and Almetrics into Wordle, the website asked me to download the JavaScript application in order to produce the word clouds. Wordle is limited in terms of customisation comparing to Voyant tool for example. Word clouds produced by Wordle from both Twitter and Almetrics shown below with different charts.

Word cloud from Twitter Analysis

Word cloud from Twitter Analysis

Word cloud from Almetrics Analysis

Word cloud from Almetrics Analysis

Many Eyes

This website requires sign in to use its features. I have created my account and upload the data created from Twitter and Almetrics analysis, however, the website was very slow and the page took long times to response. This web site has a variety of visualizations can be selected from its collection, and allow some customisation such font and the legend mode. Although this web site is very slow, it allows users to remove some unmeaning words as shown below. In addition, when the size of data uploaded is small, Many Eyes response is quickly and the opposite is true.

Word cloud from Almetrics Analysis

Word cloud from Almetrics Analysis

Word cloud from Twitter Analysis

Word cloud from Twitter Analysis

after removing some unmeaning words

after removing some unmeaning words

Voyant Tools

Voyant is the best tools and it is still under continuing development by its creators. This website performs advanced level of text analysis comparing to the previous methods as it analyses the text in seconds and removes the unwanted words in order to produce a meaning word cloud as shown below. However, Voyant is unable to modify some common features such as font and colours.

Word cloud from Twitter Analysis

Word cloud from Twitter Analysis

Word Cloud from Almetrics Analysis

Word Cloud from Almetrics Analysis

Voyant tools goes beyond the word cloud and provides users with more information about the text including a full text reader, summary, a word trends graph, the frequency and the count of words in the entire corpus, and to exclude individual words and see how they appears in the documents and the context. The screenshots below demonstrate the data created from Twitter and Almetrics analysis. for Example, from the screenshot of data created by Almetrics, as we can see the most frequent words were disaster and health. On the other hand, from the data created by Twitter Analysis, the most used word in the entire corpus was the hashtag cyclist. We can also exclude any word and see more information about it. For example, we highlighted the word Analysis to obtain more information how it appears in document and context.

Screenshot of data from Twitter analysis

Screenshot of data from Twitter analysis

Screenshot of data from Almetrics analysis

Screenshot of data from Almetrics analysis

Summary

Word cloud may be a good method to know the most important keywords of the text and to obtain quantitative analysis. However, it is not a good qualitative method. Another tools such as Voyant can tell more about the text and provide users with more information about the text, but, more development in these tools is required to obtain advanced quantitative and qualitative analysis.

Advertisements
This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s