Introduction

Examine the archive of relevant petition information. Victories, the number of those who voted for the petitions, and the countries of the participants were noted. It seems interesting to me to analyze texts of petitions with information using a neural network. Identify the most exciting topics. How do they correlate with the political component and social needs of the country's population? Speed wins. Who were the applicants and to whom were the petitions addressed. The accompanying information background is a mention of the problem and petitions in the media. And the main goal: to help better understand how and why to participate in public life, help each other and change life in the country for the better.

Data set

The data set is created in the form of a CSV, Excel table (two files: Criminal proceedings Russia, Criminal proceedings World p1), which can be found at boot time. You can see the data source on the website https://www.change.org/

1. column - To whom the petition is directed
2. сolumn - Title of the petition
3. column - Petition link web
4. column - Number of people supporting the petition
5. column - Applicant
6. column - Applicant Country
7. column - Applicant's city
8. сolumn - Text of the petition
9. column - User id
10. column - Petition status: publisher, victory, closed
11. сolumn - Petition Opening Date
12. column - Date of victory of the petition.
13. column - Closing date of the petition
14. сolumn - Progress
15. column - Keyword

Inspiration

1. Search for criteria for the most successful petitions.
2. Resonant topics, their epicenters.
3. Relations with political events.
4. Relationship with the state of criminal law.
5. Tracking petitions.
6. Influence, dynamics of distribution of Change.org petitions in different parts of the world.
7. How do requests for social change correlate with petition topics?