Physics paper recommender

This project is part of blog post on topic of Recommendation System Using Online Node2Vec with Memgraph MAGE.

• An installation of Memgraph Advanced Graph Extensions (MAGE)
• An installation of Memgraph Lab or usage of Memgraph's command-line tool, mgconsole, which is installed together with Memgraph.

Setup

In order to prepare this repo, run:

pip3 install -r requirements.txt

Prechecks

File public/recommender.py assumes existence of node2vec_online module, and calculated node embeddings. If this is not prepared, follow blog post to learn how.

In order to check that you have node2vec_online query module loaded and embeddings ready, run following command inside Memgraph Lab or mgconsole:

CALL node2vec_online.get() YIELD *;

Commands

Position yourself inside public repo

To visualize k-means inertia, run:

python3 recommender.py visualize

To get top 10 similarities over 5 groups, run:

python3 recommender.py similarities --top_n_sim=10 --n_clusters=5