We have a lot of customers, and we have a lot of research into our installed base. We also watch industry trends pretty carefully. But in the end, for something this complicated, it's really hard to design products by focus groups. A lot of times, people don't know what they want until you show it to them.
-- Steve Jobs
The Yelp Dataset (https://www.yelp.com/dataset) comes as a series of JSON files, so we will start by downloading and preprocessing the dataset for training a Heterogeneous Graph Neural Network. Although GNNs are popular, a theoretical overview of how to use and create GNNs is needed to understand each step of the tutorial. Then, we will use Neo4j to inspect and visualize the graph. The central part of the tutorial focuses on how to use the DGL library to create a heterogenous Graph Neural Network, preprocess the dataset, and use the graph to produce business recommendations for users. We will use Python, Deep Graph Library (DGL), and Neo4j. The tutorial can be replicated using a Python Notebook and the Neo4j Desktop application.
LARUS R&D Tribe created this project.
It would be hard for a tech company to remain innovative and competitive without a continuous investment in R&D. That is why in LARUS there is an entire tribe dedicated to research and technological development involved in an ongoing solid relationship with relevant Italian and foreign universities, observatories, research institutes, labs but also professors and researchers.
Assuming that you already have Python installed on your local machine.
pip install -r requirements.txt
If you want to perform some graph exploration/visualization you also need to have Neo4j Desktop installed.