Improving recommendations with GDS algorithms
In this section, we will look at how we can enhance the graph further to gain more insights into the graph to build a better recommendation system. We will start with the graph database we created in the last chapter. For reference, you can download it from https://packt-neo4j-powered-applications.s3.us-east-1.amazonaws.com/Building+Neo4j-Powered+Applications+with+LLMs+Database+Dump+files.zip.
The Neo4j GDS algorithms (https://neo4j.com/docs/graph-data-science/current/) will help us enhance the graph. This process includes the following steps:
- Calculate the similarity between customers based on the embeddings we have created and create a similar relationship between these customers. For this purpose, we will leverage the K-Nearest Neighbors (KNN) algorithm (https://neo4j.com/docs/graph-data-science/current/algorithms/knn/).
- Run the community detection algorithm to group the customers based on similar relationships...
