In the past phase of the project we’ve been focussing (amonst others) on the manual annotation process. Finally, it’s are getting along well. It has been quite a struggle appropriately placing the entity stamps, having the computerized annotations in the back of our heads (even though this shouldn’t be influential…)
Computer-wise, it’s now a matter of figuring out do we deal with the relative and absolute dates and times in the text, with reference to the publication date of the article.
Hopefully we can finalize the first article batch next week in order to start an evaluation!
Fascinating to see how worlds collide in where computer scientists and social scientists are working together to see how we can make sense of large datasets to understand activist groups’ tactics and their targets’ responses.
Clearly, not a question for which a ready-made package is available so we’re working hard to assemble helpful tools and to provide some well-coded bits of data to serve as gold standards for the tools. Once again we note how complicated language can be – but we’re making progress, using a sample data set on that gives enough food for (coding) thought…
Historically, whenever we wanted to convey large amounts of potentially complex information to others, we would write it down using a copious amount of text. While this has worked fairly well for many centuries – and in fact still does for many types of information – it is not necessarily the most efficient nor the most enticing method to convey information. But what alternative is there then?
Earlier this month we presented some of our ideas at different conferences. While Lora went to Boston to present a paper at the “Workhop on Detection, Representation, and Exploitation of Events in the Semantic Web” (see the slideshare here), Frank went to Copenhagen to discuss some of the ideas at a workshop at Copenhagen Business School which was followed by a seminar.
At the workshop a range of scholars was present, including Lance Bennett who discussed the work he is doing with the Center for Communication & Civic Engagement in Seattle, Zizi Papacharrissi who works in Chicago on social media and identity and a range of other researchers such as Itziar Castello, Mette Morsing, Tobias Goessling, Anne Vestergaard, Friederike Schultz, Julie Uldam and Dennis Schoeneborn.
After two days in Copenhagen and a trip by train into the countryside, part of the participants continued the conversation in Vejle at the first nSICE workshop ’Social Media and Civic Engagement: Contesting the Mainstream’.nSICE is the Network on Social Innovation and Civic Engagement and the workshop was embedded in the biannual conference of the Danish Association of Media Researchers (SMiD). A range of topics was debated, ranging from CSR, institutional change, Occupy and tactics.
In both workshops our event-based approach attracted several questions – we received a range of helpful comments and ideas for further research, once again assuring us that the ideas are intriguing but that the proof of the pudding is in the eating. Let’s go get some results!
We created a research poster to present our work. You can find it at the following link.
MONA Research Poster 2012
Social scientists studying the activity of activist organizations are faced with an overwhelming amount of possibly biased and out of context source material.
The MONA project aims at developing a tool that gathers source material (e.g. news articles, blog posts), extracts events and their properties (e.g. actors, locations, timestamps) from these materials, analyzes links between events, and visualizes the results. By doing this, MONA attempts to alleviate the aforementioned problems.
Through this blog the MONA project team will keep you up-to-date on the progress of the project.