The application helps tourists visit interesting places. The main idea is to use the input text from the user to create the interests vector and therefore recommend the best-fitting places in the visited city. The application not only finds the best-fitting places but also creates a comprehensive visiting plan according to user requirements.
Currently, we are building the proof of concept. However, there is going to be deployed a web application once we prove that the system works well.
The following image shows the places in Krakow sorted by the best-matching ones.
and another example
The application layout todo: paste the app layout image
First of all clone the repository
mkdir Tourister
cd Tourister
git clone https://github.com/kbarszczak/Tourister .
The next step is to install the requirements
pip install -r requirements.txt
Now the application is ready for lunch. Just run the main.py
python app/main.py
There are some things that may be improved. Those are some of them:
- find a better places description to create an interest vector more precisely related to the place
- modify the training dataset and create it by taking more reliable data