Graph Data Science Using Neo4j
Amy E. Hodler, Mark D. Needham
Abstract
This chapter focuses on the application of graph data science in business using Neo4j graph technology to illustrate examples because of its commercial popularity. It explores the steps of graph data science adoption, starting with knowledge graphs and graph analytics, moving to graph feature engineering and graph embedding, and ending with graph networks for graph native learning. Graph Statistics provides basic measures about our graph, such as the number of nodes and distribution of relationships. Graph analytics builds upon graph statistics by answering specific questions and gaining insights from connections in existing or historical data. One machine learning scenario would be to extract graph features for use in a binary classifier to predict mules. The world is driven by connections and revealing the meaning behind these relationships drives breakthrough across industries.