In addition to people who for some reason want to move to another place, the target audience of the project are also tourists, travelers and digital nomads.
Motivation
Sometimes it is difficult for people to choose one thing from several options. Even the question of what kind of coffee to drink in the morning causes difficulties for many. Needless to say, how difficult it is to choose a place to travel or to move for a long time. Moreover, if a trip or relocation is planned with the family, then the analysis of options and the choice of the only one become much more difficult. This is where our project appears, designed to help with the choice of a place for relocation, recommending cities according to their key parameters. Using our service, it will be much easier for the user to choose a suitable city for himself, taking into account his priorities.
Main idea
The main idea is that a user thinking about moving, using a Telegram bot, can get a series of recommendations. In this case, noting the most acceptable options for themselves, the user will receive more and more relevant suggestions. In order to form a basic opinion about the proposed city, the user is able to display basic information about the city in the form of charts.
Site numbeo.com provides crowd-sourced information on the key parameters of life in the cities of the world. For the project, parameters such as cost of living, cost of housing, quality of life, as well as data on crime, health, pollution and traffic were parsed. A more detailed description of the data and the methodology for collecting them can be found on the website. The method of parsing the received data is given in the notebook "numbeo_parser.ipynb".
The website of the World Health Organization provides a wide range of datasets related to the health of the population of the countries of the world. The project used a dataset of air quality in cities around the world. The remaining datasets provided general information about the countries of the world, which is not enough for the task.
Run the app in development mode
Pull and run postgres docker image.
Add environment variables for src/db/app/app.py and run it. The application needs following envs:
DB_NAME is name of DB in postgres container;
DB_ADDRESS and DB_PORT is path and port to get DB access (for local development is localhost and 5432);
DB_USER and DB_PASSWORD are credential to get DB access;
Change listening port of recsys service (src/recsys/app.py) on any value, different from 8000 (for example, 8001),
because port 8000 is busy by db service. Run the application.
Set environment variable BOT_TOKEN for src/tg/app/app.py and run it.
