Best practices

更新时间:
复制 MD 格式

This topic summarizes best practices for Graph Database.

Description

Reference

Get started with PolarDB Graph Database. Learn its core concepts and how to use it.

Quick Start

Use the Graph Database plugin in PolarDB to quickly import hundreds of millions of nodes and edges. This method avoids performance bottlenecks caused by queries during edge insertion.

Graph analytics based on PolarDB: Quickly import data to a graph from a table

Use a Data Transfer Service (DTS) task to create a real-time synchronization link that synchronizes data from other databases to PolarDB Graph Database.

Graph analytics based on PolarDB: Synchronize data tables from other databases to a PolarDB graph using DTS

Use the Graph Database plugin in PolarDB to run graph queries on a public insurance dataset. This example shows how to identify abnormal claims and fraud rings in an insurance claims scenario.

Graph analytics based on PolarDB: A practical guide to insurance data analytics

Use the Graph Database plugin in PolarDB to identify relationships between fraudulent transactions with graph queries. Calculate the Jaccard similarity between transactions to trigger fraud alerts.

Graph analytics based on PolarDB: A practical guide to graph analytics in the banking and finance sector

Build a retrieval-augmented generation (RAG) system with PolarDB, Qwen, and LangChain. This system combines a knowledge graph and vector retrieval to improve question-and-answer quality.

GraphRAG: Best practices for knowledge graphs and LLMs based on PolarDB, Qwen, and LangChain

PolarDB uses the Mem0 framework to integrate vector and graph database engines. This integration enables an AI agent to store and retrieve user preferences and history across sessions, delivering a true long-term memory experience. This design addresses the common limitation of Large Language Models (LLMs) forgetting conversation history due to context limits, thereby improving service continuity.

Graph analytics based on PolarDB: A one-stop solution for AI agent long-term memory