Warnings-dropoff-analysis

Repository which contains the scripts to retrive user warnings metrics from a MongoDB database.

The collections should be obtained merging the outcomes of user-metrics and Wikidump repository.

Setup

Clone this repository

git clone https://github.com/WikiCommunityHealth/warnings-dropoff-analysis.git

Create the poetry project virtual environment using your current Python version

make env

Install the required packages and dependencies

make install

Usage

Only after having installed the dependencies in the poetry virtual environment, can you run the following command

poetry run python -m warnings_dropoff_analysis db-name collection-name [--output-compression gzip] extract-user-warnings-metrics [month to consider the drop-off]

Or you can simply edit the Makefile and type

make run

The result will be found in the ouput folder.

Plots

After the data has been retrieved, you can plot some basic statistics by typing

poetry run python plotter/[plotter-stats-file].py [community-lang]

Or, as in the previous case, you can edit the Makefile and run

make plot

The produced figures can be found in the plots folder.

Note

The script tries, by default, to connect to the local MongoDB instance.

So as to connect it to a remote one, you can add a .env file writing the connection string within, following the .env.example template file:

echo '[your mongodb connection string]' > .env

Note about the chart scripts

Be careful with the plot scripts, since pandas memory overhead, to instantiate the DataFrame structure, is extremely RAM consuming.

As a result, make sure you have enough free memory before running one of those scripts.

Data format

The data retrived by the script is in the following format:

{
  "name": "<name>",
  "id": 21,
  "last_edit_month": 2,
  "last_edit_year": 2010,
  "retirement_declared": false,
  "retire_date": null,
  "retirement_parameters": null,
  "retirement_template_name": null,
  "edit_count_after_retirement": null,
  "last_serious_warning": {
    "name": "avvisoimmagine2",
    "date": "200_-0_-1_ 2_:1_:4_+00:00"
  },
  "last_serious_warning_name": "avvisoimmagine2",
  "last_serious_warning_date": "200_-0_-1_ 2_:1_:4_+00:00",
  "last_normal_warning": {
    "date": null,
    "name": null
  },
  "last_normal_warning_name": null,
  "last_normal_warning_date": null,
  "last_not_serious_warning": {
    "date": null,
    "name": null
  },
  "last_not_serious_warning_name": null,
  "last_not_serious_warning_date": null,
  "average_edit_count_before_last_serious_warning_date": 9.333333333333334,
  "average_edit_count_before_last_normal_warning_date": 0,
  "average_edit_count_before_last_not_serious_warning_date": 0,
  "average_edit_count_after_last_serious_warning_date": 0.0036443148688046646,
  "average_edit_count_after_last_normal_warning_date": 0,
  "average_edit_count_after_last_not_serious_warning_date": 0,
  "count_serious_templates_transcluded": 1,
  "count_warning_templates_transcluded": 0,
  "count_not_serious_templates_transcluded": 0,
  "count_serious_templates_substituted": 0,
  "count_warning_templates_substituted": 0,
  "count_not_serious_templates_substituted": 0,
  "edit_history": {
    "2001": {
      "1": 0,
      "2": 0,
      "3": 0,
      "4": 0,
      "5": 0,
      "6": 0,
      "7": 0,
      "8": 1,
      "9": 0,
      "10": 0,
      "11": 0,
      "12": 0
    },
    "2015": {
      "1": 0,
      "2": 0,
      "3": 0,
      "4": 0,
      "5": 0,
      "6": 0,
      "7": 0,
      "8": 0,
      "9": 0,
      "10": 0,
      "11": 0,
      "12": 3
    }
  },
  "warnings_history": {
    "2005": {
      "10": {
        "serious_substituted": 0,
        "not_serious_substituted": 0,
        "not_serious_transcluded": 0,
        "warning_substituted": 0,
        "warning_transcluded": 0,
        "serious_transcluded": 1
      },
    "2021": {
      "1": {
        "serious_substituted": 0,
        "not_serious_substituted": 0,
        "not_serious_transcluded": 0,
        "warning_substituted": 0,
        "warning_transcluded": 0,
        "serious_transcluded": 0
      },
      "12": {
        "serious_substituted": 0,
        "not_serious_substituted": 0,
        "not_serious_transcluded": 0,
        "warning_substituted": 0,
        "warning_transcluded": 0,
        "serious_transcluded": 0
      }
    }
  },
  "sex": null, 
  "banned": false
}

Fields

A brief description of the fields

• name name of the user
• id user id
• last_edit_month month of the last edit date
• last_edit_year year of the last edit date
• retirement_declared is a boolean field representing, whether the user has specified a retirement template or not, on the user page or user talk page
• retire_date the retirement date if specified
• retirement_parameters: parameters specified in the retired template
• retirement_template_name: name of the retired template
• edit_count_after_retirement: activity count after the retirement date
• last_serious_warning: last transcluded serious warning
• last_serious_warning_name: the name of the last high severity warning received
• last_serious_warning_date: the date of the last high severity warning received
• last_normal_warning last transcluded warning of medium severity
• last_normal_warning_name: the name of the last medium severity warning received
• last_normal_warning_date: the date of the last medium severity warning received
• last_not_serious_warning last transcluded non serious warning
• last_not_serious_warning_name: the name of the last low severity warning received
• last_not_serious_warning_date: the date of the last low severity warning received
• average_edit_count_before_last_serious_warning_date average count of actions the user has made before the last serious warning date (in the range of [date - 12 months, date])
• average_edit_count_before_last_normal_warning_date average count of actions the user has made before the last warning date (in the range of [date - 12 months, date])
• average_edit_count_before_last_not_serious_warning_date average count of action the user has made before the last not serious warning date (in the range of [date - 12 months])
• average_edit_count_after_last_serious_warning_date average count of action the user has made after the last serious warning date (in the range of [date, date + 12 months])
• average_edit_count_after_last_normal_warning_date average count of action the user has made after the last warning date (in the range of [date, date + 12 months])
• average_edit_count_after_last_not_serious_warning_date average count of action the user has made after the last not serious warning date (in the range of [date, date + 12 months])
• count_serious_templates_transcluded count of the transcluded serious warnings templates
• count_warning_templates_transcluded count of the transcluded warnings templates
• count_not_serious_templates_transcluded count of the transcluded not serious warnings templates
• count_serious_templates_substituted count of the substituted serious warnings templates
• count_warning_templates_substituted count of the substituted warnings templates
• count_not_serious_templates_substituted count of the substituted not serious warnings templates
• edit_history history of the user's activity per month
• warnings_history history of the user's warnings per month
• sex: the user's sex if specified
• banned: whether the user is banned or not

Code documentation

In order to get the code documentation, you can use pdoc by typing

make doc

Or you can open it directly with your browser using xdg-open

make openDoc

Charts

Here, the plots, produced by the scripts stored in the plotter folder, are listed

Users who have declared their withdraw from Wikipedia

The outcomes of the retired_stats.py.

Considering only the users who have declared their withdrawal from Wikipedia:

• A bar chart which indicates the percentage of users who have stopped editing in Wikipedia, and the percentage of users who have continued to be active.
• A bar chart which shows the number of edits the users have made after having declared their withdrawal.
• A bar chart which illustrates the percentage of men and women who have declared the withdrawal from Wikipedia

Users who have received at least a warning of medium severity

The outcomes of the serious_warnings_stats.py.

Considering only the users who have received at least a serious warning:

• A bar chart which indicates the number of users who have decreased their activity in the community within the time interval of the last serious warning received.
• The graph above but considering only men and women
• A bar chart which compares the percentage of women and men who have decreased their activity
• Considering the five users who have received the highest amount of edits:
  • A line chart which shows the user activity over months with some vertical lines, which indicate one or more warning.
  • A line chart illustrating the z-score of the user history within the time interval of the last serious user warning received

Run

In order to call the scripts on all the MongoDB database collections, it is possibile to run the run bash script.

./run.sh

First of all, be sure you have modified all the readonly variables so as to fit your needs; feel free to change whatever you want.

The dependencies of the previously defined script are

• GNU parallel

Docker

So as to call the entire program in a Docker cotainer, a Dockerfile has been provided.

First, you need to change the content of the run.sh file in order to fill your requirements, such as the files' locations and which operation should be carried out by the script.

Then, you can build the Docker image by typing:

docker build -t warning-dropoff .

Then, make sure you have your local MongoDB instance working with all the required data.

Finally, run the docker image:

docker run --network="host" warning-dropoff