marcuskbrandt / socialgraphs Goto Github PK

We want to create the communities and comment on how they are compared with the categoeries that we pull from the wiki

We want a column in our dataset with the main category of each character and another column containing a list of all the categories for the character

We want to do sentiment analysis for both communities and categories. It will be fun to see how this is different.

It would be nice to have some better-styled wordclouds.
We also have some names left in the wordclouds since it is not links but mentions in text.

Looking at the sentiment analysis for all characters through all seasons.

I have seen a few places that you can plot your network more interactive. More nodes around, place name of character, maye some stats etc.

Do what we did in the first assignment. This way we will get all the stats for the network.
This will have to be showed on the webpage.

We want to see how the main persons evolve through the seasons. We will do that by looking at the tf-idf score, for epsisodes in the seasons. This can maybe be done more easily by only taking the first episode from a season or last.

Use the code from exercise 6, to create plots of Zipf's law from the dialogue texts.

The page should say clearly what the dataset is and give the reader some idea of its most important properties (kind of Project Assignment A-style).

Use some of the code from assignment 1. Maybe try to look at some of the other groups who did a great job. You have 3 other groups assignments on peergrade.

As Michael Berger says (An engineer steels with arms and legs)

Make dem nice plots

Plot network of communities with ForceAtlas
Plot network of categories with ForceAtlas
Histograms of community sizes

We want to look at tf-idf for both communities and categories. It will be interesting to see how they are different.