Information visualization of direct messages between two users on Discord. An IPython notebook document is used so that the infographics can easily be modified and extended.
The following types of visualization are presented in the project:
- Number of messages sent per person (pie chart)
- Number of messages sent per day (scatter plot)
- Number of messages sent per month (bar plot)
- Number of messages sent per month per person (bar plot)
- Number of messages sent per day (bar plot)
- Number of messages sent per day per person (bar plot)
- Number of messages sent per hour (bar plot)
- Number of messages sent per hour per person at different times (e.g., GMT+2 and GMT+12) (bar plot)
- Message length per person (< 50 characters)
- Message length per date (scatter plot)
- Type of message sent (pie chart)
- Most used words per person
- Most common used words
- Word usage by absolute frequency of use (possible to add a limit to number of occurrences)
- Word usage by relative frequency of use (possible to add a limit to number of occurrences)
- Word usage by frequency in percentages of use
These instructions will inform you about the prerequisites and how a data set containing direct messages between two users can be obtained.
The following packages are needed and can be installed with e.g. pip:
- pandas
- matplotlib
- numpy
- seaborn
- adjustText
It is possible to use (this)[https://github.com/Tyrrrz/DiscordChatExporter] tool to obtain a data set that captures all direct messages between two users. One can get the a data set containing the following columns using this tool:
- author (e.g., “EXAMPLE#0000”)
- date (e.g., “15-Dec-17 11:33 AM”)
- content (e.g., “Hello”)
- attachments (e.g., “https://cdn.discordapp.com/attachments/“)
- reactions
The project can be run with e.g. a Jupyter Notebook or Google Collaboratory.