Descriptive Data Analysis

epfl-ada / ada-2023-project-amonavis

ada-2023-project-amonavis created by GitHub Classroom

0 stars 0 forks source link

Descriptive Data Analysis #5

Open carloscca10 opened 8 months ago

carloscca10 commented 8 months ago

Basically Pandas summary statistics.

Examples:

How many articles are, how many paths
Histograms of the links from each article (for example, how many articles have 20 links, etc)
Histogram of the number of games at each point in time (for example, 30 paths were completed in May 2021).
Classify how many times each word appears in the games (we could also do a histogram with that)

Add whatever shit you think could be cool to plot.

WhimZig commented 7 months ago

Someone please do extra processing, that analyzes and kinda filters the finished paths. Try and get some aggregate data for all existing paths.

In a nutshell, go through finished paths and do the following:

Find all unique source and target pairs. Call this the key for now For each key, do some analysis of the number of times it was explored, average path length, average difficulty, etc.

carloscca10 commented 7 months ago

I can do that.

codecamaru commented 7 months ago

Added a visual representation for degree node (graph drawing). We can compare this figure with the one from pagerank and use this to assess which metric is better.