Data visualization helped me graph the frequency, gender, and rank of the name data set. When choosing my graph I automatically had to erase a good 80% of the options because they didn’t fit with the data set. The graphs either required too much information or not enough for the data set. The graph I chose created a circular layout of the baby names. Each name is placed around the edges of the circle, grouped with other names that appeared just as much or at least a similar amount of times. The dots are colored by gender, with the third color (orange) for names that could be either gender. The names are all connected to the center of the graph, I could chose to have cluster dendrogram. This option gave me a little more information about how frequent these names were and where they clustered. I would say one of the issues with my graph is that theres a lot going on. It can be a little overwhelming to look at, at a glance. Many of the labels overlap, the names towards the top ‘rank’ also become less specific because there is not enough space. I experimented with expanding the parameters of the graph but realized it would be far to big to interpret in one image, I also tried flipping the rank from one side to the other but that obviously didn’t change much.
This graph / exercise connects the digital humanities because it shows how data isn’t neutral, it matters what type of ‘visual representation’ we use. If someone looked at this representation they might feel overwhelmed at first but with other, more specific to the data set, graphs we may be able to offset this initial effect. Although this graph shows a lot of information and enough detail for real analysis it goes to show how the actual visualization matters more than we think. Some graphs will make the user do more work while others will seem a lot more obvious. Much like in my blog, the readings by Lin show how much the dictation of images affects the viewers initial reaction. In the AI color generation it could help people understand more about historical photos while running the risk of not being entirely accurate and in the data set what matters is how the graphs represent data; with some being harder to decipher information from than others.
IS vthNice work!
Sorry, I posted this by mistake. I’m not sure how to delete the comment. Anyway, I like how you grouped the names by gender and also included those that can be used for both males and females! Also, I had the same issue as yours, where the labels overlapped with each other. I wonder how to fix that. Nice work!
Very good job on this post Daniel! I really appreciate how you found a graph that fits the data very well, and it seems like you were very thoughtful in creating the layout of this plot. Moreover, I think your addition of gender to the color scheme was a nice touch that I personally would never have thought of doing. Great job!
I think that this visualization is super cool and unique! I especially think the inclusion of the legend makes it very easy to categorize the different names. I never thought about including something to signify unisex names. My model did not allow me to incorporate another label, which is something I could have benefited from.