A data visualization allowing users to explore conversational themes and conversation ratios between sexes across decades and genres. You can view the visualization here
Visualization made by -
Our visualization explores movie dialogues spanning 9 decades (1927 - 2010). The purpose of our visualization is to show themes and inter/intra sex ratio in the conversations. We want to bring to light how the ratio of female-female conversations has not really changed over the years. Using overview and then detail, our visualization will highlight these movie characteristics and give our users insight into their favorite movies as well as overall trends.
Some questions you can answer with our viz are:
- How does inter/intra sex ratio in conversations vary over years and themes?
- How do themes and genres intersect?
- Do certain genres have a better gender balance in conversations?
- Do movies that pass the Bechdel test have a higher percentage of female-female conversations? Where does the Bechdel test fail?
- In a specific movie, how do themes intersect with characters? What are the patterns of conversation between characters?
Using overview and then detail, our visualization will highlight these movie characteristics and give our users insight into their favorite movies as well as overall trends.
https://d3-legend.susielu.com, Bootstrap, Adobe Typekit, LIWC, TextBlob, Genderizer, Linear gradient example
Our source of data was the Cornell Movie Dialogs Corpus, released along with the paper Chameleons in Imagined Conversations: A new Approach to Understanding Coordination of Linguistic Style in Dialogs by Cristian Danescu-Niculescu-Mizil and Lillian Lee.