The search engine uses this dataset from kaggle as its basis. It consists of ~7700 netflix titles.

The application requires elasticsearch to be running. For simplicity the movie dataset can be loaded through Kibana. Thus, make sure you have the following setup:

elasticsearch (we used version 7.12.1)
kibana (we used version 7.12.1)

In order to load the dataset into elasticsearch, do the following:

1. Open Kibana
2. Press Upload a file on the front page
3. Select the csv file
4. Press import
5. Name the index netflix and press import
6. It will say that some documents could not be imported, don't worry about that

Now the index is ready to be queried from the user interface

The application is built using python 3.8 and it requires the following dependecies to run:

tkinter (installation instructions below)
tinydb (pip install tinydb)

Instructions here

The project uses tinydb which stores its data in json file. We store the user profiles there and feel free to use our pre-made users or create one yourself. Either way the code assumes the database is located at ./db/users.json.

When all dependencies are installed, the application can be run with the following command:

python main.py

If you want to start with an empty user intead of using the ones we've created run the following command:

rm ./db/users.json && python add_person.py

The user interface is fairly simple, at the top you will find a list of the users and whenevery you run a query or select a title a user has to be selected. Below that list is the input box, where you can enter a query and search. Below the input are the results and at the bottom of the page is the search history and previous selections made by the user.