Content-Based-Filtering-with-ElasticSearch

A Movie recommendation system using Content Based Filtering with ElasticSearch

Built With

• Python
• ElasticSearch
• Sbert
• FastApi
• NLTK
• Pandas

Getting Started

Installation

1. Clone the repo

git clone https://github.com/shulavkarki/Content-Based-Filtering-with-ElasticSearch.git

2. Create Virtual Env.

python -m venv {environmentname}

activate accordingly.

3. Install the requirements.

pip install -r requirements.txt

4. Fire up the elasticsearch and kibana service with docker.

docker-compose up

run docker desktop. Make sure the elasticsearch and kibana is working by going to following ports. localhost:9200 localhost:5601 You should get something like this:

![image]()

5. Index the movie.

For indexing movies to elasticsearch. Run the following.

python create_index.py

6. Run Fastapi

uvicorn foldername.app{filename app.py}:app{name of fastpi method} --reload

In and Out

Pull up:

localhost:8000/docs

Input movie_name: "star wars" no_of_recommendation: 10

Output