clk125
commented
3 years ago It sounds like your group is at a really good point in the process. I think it is really interesting that your results have the potential to show much about the society of the time. My group has not yet started talking about any topic modeling use for our project but I am interested to see how your group utilizes it. My group is also working on our website and the division of labor moving forward. Ill be interested to see how you use topic modeling because I dont believe it would benefit our project much so I dont believe we will use it.
This week we talked about our research questions a little bit more, and also discussed the division of labor and things we wanted to do moving forward now that our markup is done. We definitely want to analyze the policies and values over time, e.g how does the policy type civil rights change in relevance over time? This could yield relevant historical information that we can analyze through the lens of political party (for example, republicans probably talked about it more in the 19th century, but democrats probably talk about it more now). We can do this kind of analysis with many different policies and values, but we generally agreed this meeting that we didn't have much interest in the "tone" markup that we did. We also discussed topic modeling using MALLET and how that could be a relevant data point and tool to visualize trends over time. Our corpus should be large enough to use MALLET. Another thing we discussed was Network Analysis using Cytoscape or some other method. This could be used in data visualization for our project. For the coming week, we are looking into these two technologies as well as continuing to work on our website. Hopefully next week we will have an idea of how exactly we will incorporate such technologies into our project and begin doing so.