
Most commonly used words in #OzHA2015 tweets during the 2015 Australian Historical Association Conference. Click on the word cloud to enlarge it. Click on it again to be taken to the data behind this word cloud. I love this facility from Voyant Tools!
During the four days of open sessions at the conference, participants tweeted over forty thousand words excluding hashtags and Twitter handles. This year’s conference had the biggest Twitter stream of any Australian Historical Association conference since 2012 and as my last post showed, more people tweeted the conference than ever before.
A conference Twitter stream is a news service for those who cannot attend the event. It is a crowd note-taking service which participants can refer to in order to jog their memory, find out what happened in sessions they did not attend and to provide added commentary which enriches the conference discourse.
Yet we need to be careful about what a Twitter conference stream can and cannot provide. The fact that there are lots of tweets does not necessarily mean that the event is properly represented in the hashtag. Any digital history papers are more likely to be well tweeted because attendees interested in technology are more likely to be on Twitter. Likewise some papers might miss out on coverage on Twitter even though the room is full, because those attendees are not on Twitter.
How well did this year’s Twitter stream reflect the conference program?
The word cloud above represents the most frequently tweeted words during the four days of sessions at the conference (after excluding stop words such as ‘the’ ‘are’, ‘a’ etc). Predictably words such as history, historians, Australian, Australia and historical were among the most commonly used words in the Conference Twitter stream. If we delve deeper there are some other themes which emerge.
I have taken a closer look at the fifty most tweeted words to identify topics of significant interest to the tweeps during the conference. This is a subjective analysis. I did not use topic modelling software and I ignored over four thousand words which occurred with less frequency in the conference Twitter stream. However, I feel that this limited analysis provides an interesting indicator of at least some of the dominant themes in the Twitter stream. Continue reading