ALOE: Affect Labeler of Expressions
Status: open source
We developed ALOE to train and test machine learning classifiers for automatically labeling chat messages with different emotion or affect categories. The software takes as input a CSV file containing timestamped chat messages with labels for training and produces a trained Support Vector Machine classifier as described in our CSCW 2013 paper, Statistical Affect Detection in Collaborative Chat.
Once a classifier has been trained, ALOE can use the classifier to label other unlabeled message data. It also features an interactive mode, where chat messages can be entered one at a time for classification.
ALOE is open source software issued under the GNU General Public License.
Hoptrees: visual hierarchy navigation
Status: open source
Hoptrees are a visual navigation interface developed by SCCL in collaboration with researchers from Eigenfactor.org. Hoptrees make it easier for people to keep track of where they are and where they’ve been when navigating hierarchical information. For example, they could help you avoid getting lost in your file system, or on a complex hierarchical website. A paper about hoptrees, Hoptrees: Branching History Navigation for Hierarchies, will be published at INTERACT 2013.
- Read more about our research on hoptrees
- See a hoptree in action
- Check out our open-source prototype on GitHub
Text Prizm: qualitative analysis of large social media data
Status: internal development
To support our research on emotion in collaborative work, we are developing Text Prizm, a web application to help you analyze social media content (such as chat logs or Twitter data). Text Prizm will provide the following features:
- Simultaneous open coding by a team of researchers over large text data sets.
- Minimally painful coding – a lightweight coding interface driven by easy keyboard shortcuts.
- Exposure to all coded data in a portable, open format.
- Collaborative iteration on a structured coding scheme through memo-writing and visualizations.
- Develop custom analyses based on human-in-the-loop active learning to extend coding over the entire data set
- Interactive visualizations that provide access to the underlying text, preserve context, and support qualitative research methods.
If you are interested in hearing from us when Text Prizm is available, Join this mailing list!