I am not attending Ascilite 2010, but thought it would be useful to get an overview of the twitter stream from the conference. Twapperkeeper contains the archive for the #ascilite10 hashtag. I have used a TreeCloud visualisation, which has the benefit of being able to separate tweets based on word proximity.
What do you think? Is the essence of the twitter stream for Ascilite 2010 captured?
A TreeCloud is a visualization technique that uses a tree structure to display words used in close proximity of each other. I thought the technique would be useful if applied to tweets from an event, given the temporal nature of posts. I used TwatterKeeper to export tweets containing the #ozchi hashtag for the first two days of the OzChi conference. The TreeCloud visualization was able to group together tweets about the keynotes (John Seely Brown, Elizabeth Churchill and Jacob Buur) and other themes such as vote for the student posters, get the OzChi iPhone app, travelling to OzChi in Brisbane and me (@aneesha) tweeting about analysing Ozchi paper abstracts over 5 years. I have included an annotated version of the TreeCloud for OzChi 2010 tweets below.
Software to create TreeClouds is available from http://www.lirmm.fr/~gambette/treecloud/. The code is written in Python. An online tool to create TreeClouds is also available.
I am attending my first OzChi conference - OzChi 2010 in Brisbane, Australia. Having not been to any OzChi conferences before, I was interested in analysing papers across years in order to find common themes. I have used a matrix decomposition algorithm to uncover the main research themes from previous conferences and well as the current OzChi conference. Each year (2006 - 2010) was analysed separately, with only the paper abstracts being included in the analysis. I used jQuery to extract the text for the abstracts from the online proceedings. Each abstract was then added to a term-document matrix with common words such as “design, paper, present, presented, display, displays, displayed, develop, developed” removed. The Non-negative Matrix Factorisation algorithm was used to uncover themes or topics within the accepted papers for each year. The main words found in the top 5 themes for each conference are listed. Mobile and location aware application research seems to be present from across all years analysed.
OzChi 2006
| Theme | Main Words |
| Theme 1 | awareness [16] social [33] technology [50] cue [9] study [35] |
| Theme 2 | usability [29] software [26] model [18] mobile [25] develop [28] evaluate [27] |
| Theme 3 | keyboard [11] key [11] input [13] wearable [7] text [9] |
| Theme 4 | develop [28] technique [19] software [26] tool [21] participatory [7] |
| Theme 5 | probe [20] technology [50] culture [15] method [24] children [9] |
OzChi 2007
| Theme | Main Words |
| Theme 1 | gesture [25] remote [17] physical [19] |
| Theme 2 | mobile [17] tool [21] everyday [11] usage [8] life [7] |
| Theme 3 | visualization [22] support [35] information [18] tool [21] |
| Theme 4 | speech [11] cognitive load [11] multimodal [7] |
| Theme 5 | physical [19] interaction [42] community [21] game [9] social [8] |
OzChi 2008
| Theme | Main Words |
| Theme 1 | community [44] technology [41] social [34] interact [46] design [31] network [14] |
| Theme 2 | game [25] player [8] participation [26] study [37] interact [46] approach [14] |
| Theme 3 | speech [10] cognitive load [6] measure [5] |
| Theme 4 | mobile [30] usage [10] phone [13] feature [11] |
| Theme 5 | interact [46] design [31] body [10] kinesthetic [7] |
OzChi 2009
| Theme | Main Words |
| Theme 1 | technology [47] social [26] design [28] support [31] practice [15] participation [23] |
| Theme 2 | scroll [22] tilt [8] flick [5] task [19] |
| Theme 3 | active [43] physical [13] game [9] user [27] play [7] |
| Theme 4 | location [23] share [10] data [15] map [10] phone [9] |
| Theme 5 | group [23] study [40] older [8] research [39] adult [6] feedback [8] |
OzChi 2010
| Theme | Main Words |
| Theme 1 | experience [61] game [18] interact [51] product [9] person [16] ux [4] |
| Theme 2 | search [21] web [20] inform [38] cognitive [13] style [7] find [23] |
| Theme 3 | cloud [14] tag [13] table [6] visualisation [4] social [32] task [17] |
| Theme 4 | mobile [30] news [6] profession [8] quality [12] study [51] evaluation [25] |
| Theme 5 | technology [47] social [32] people [22] inform [38] older [9] isolated [4] |
| Theme 6 | real-time [8] coordination [5] challenge [11] sound [3] musician [3] explore [26] |
| Theme 7 | inform [38] role [18] interact [51] copresence [3] eparticipation [4] |
I have built an interactive theme or topic explorer tool as part of my research at QUT. The tool is powered by the Non-negative Matrix Factorisation and Latent Dirichlet Allocation algorithms. The tool allows users to interact with algorithm parameters, includes techniques to assist with data interpretation and was used to perform the analysis. Feedback on the analysis is welcome and appreciated.
Making sense of the twitter feed from OzChi 2010 will be published tomorrow. Stay tuned!
| 
Designed By : 
Get this theme at : 