February 2025 Summaries

The text discusses using relationship coloring and Neo4j to solve scheduling problems. Relationship coloring is a technique for assigning colors to relationships in a graph, where each color represents a time window, and no node can be connected to multiple relationships of the same color. The Misra and Gries algorithm is used to find a valid coloring for a graph by iteratively selecting colors that are free on key fan nodes. The algorithm starts with an uncolored relationship and finds a maximal Vizing fan, chooses two colors that are not adjacent to the first node in the fan, and then inverts the color of all relationships along the c-d path. Finally, it rotates the colors assigned to the relationships in the rotation fan. The text also discusses how to apply this algorithm to real-world problems such as scheduling football games and network communications. Neo4j is used to implement the algorithm and visualize the results.

The article provides a list of 15 graph visualization tools for Neo4j graph databases, categorized into exploration tools, development tools, dashboarding tools, and embedded tools. The tools are designed to help users visualize their data and gain insights that would be difficult or impossible with traditional tables and business intelligence tools. Each tool has its unique features, strengths, and use cases, ranging from simple and interactive visualizations to more advanced and customizable solutions. The article aims to provide a comprehensive overview of the best graph visualization tools for Neo4j, helping users choose the right tool for their specific needs and use case.

Cypher's conditional queries feature makes it easier to implement Conway's Game of Life, a cellular automaton that simulates the evolution of cells in a grid based on simple rules. In this implementation, Cypher's declarative nature simplifies the process of counting neighbors and updating cell states, making it ideal for this task. The author initially aimed to write a blog post about conditional queries but realized they could be replaced with more efficient case expressions. A Java implementation is provided for comparison, highlighting the verbosity of traditional programming languages in handling similar tasks. The Cypher implementation uses a graph database to model the game board and leverages Cypher's query language to efficiently count neighbors and update cell states. A graphical version using Neo4j Bloom provides an interactive visualization of the game, allowing users to see the evolution of cells over time. Additionally, the Gosper glider gun pattern is implemented in Cypher, demonstrating its capabilities for more complex cellular automata patterns.

Neo4j has joined the Linux Foundation for AI and Data (LFAI), a leading community organization for open-source software artificial intelligence. The company is also hosting a competition to test GraphQL for Neo4j AuraDB in its pre-release beta phase, with a $250 gift card and a Neo4j T-shirt as a prize for the winner. Additionally, users can participate in a user research panel to share their experiences and influence the future of Neo4j products, with a chance to connect directly with product development teams and get paid compensation. The company is also releasing new features such as LlamaIndex, which allows for agentic workflows using multi-step queries, and GraphRAG, a tool that helps build Knowledge Graphs. Users can learn more about these features and others through Neo4j's GenAI courses, webinars, and On-Demand webinar library.

Entity resolution is a critical capability that enables organizations to understand when different data points refer to the same real-world entity, going beyond simple deduplication by considering context and relationships between data points. It's essential for large, complex datasets and can help uncover hidden patterns and relationships in data. Entity resolution matters because poor data quality costs organizations significantly, leading to missed fraud patterns, incomplete customer views, redundant marketing efforts, compliance risks, and supply chain blind spots. Successful entity resolution enables organizations to gain a deeper understanding of their data relationships and make better decisions. Real-world use cases include healthcare, where it solves the challenge of determining when different medical records represent the same patient, and retail, where it resolves product information discrepancies across e-commerce sites, inventory systems, and supplier catalogs. Core entity resolution techniques include deterministic matching, probabilistic matching, and graph-based methods, each with its strengths and limitations. To implement entity resolution effectively, organizations should focus on data quality, scale and performance, privacy and compliance, and emerging technologies that transform how we connect related entities.

Neo4j is offering a $250 gift card as part of a competition to build Conway's Game of Life using GraphQL for Neo4j AuraDB Beta. The game, which involves simple rules creating complex patterns, is inherently a graph and makes Neo4j a perfect fit for storing and computing game states. Participants will learn how to set up GraphQL for AuraDB, define the schema, write custom Cypher queries, and create an interactive UI in a browser through a tutorial series. The challenge requires participants to comply with specific rules, including using Neo4j AuraDB as the back-end database, communicating via GraphQL, and meeting functional scope requirements. The best implementation will win the gift card, and submission deadline is March 31, 2025.

The LLM Knowledge Graph Builder is an open-source tool that enables developers to build retrieval-augmented generation (RAG) experiences to interact with unstructured data. It preprocesses documents, transcripts, and web articles into chunks, computes text embeddings, and connects them to create a lexical graph. The tool also extracts entities and their relationships, which can be used to connect pieces of information and provide relevant context to user questions. In the first release of 2025, the LLM Knowledge Graph Builder introduces new features such as generating community summaries, running multiple retrievers in parallel for question answering, guiding extraction with custom prompt instructions, and supporting read-only database connections. The tool has been tested with various large language models (LLMs) and has shown promising results. It aims to improve the richness of knowledge graphs representing unstructured data and provide a more interactive experience for users.

The LLM Knowledge Graph Builder is an open-source tool developed by Neo4j that enables developers to build retrieval-augmented generation (RAG) experiences for interacting with unstructured data. The tool preprocesses documents, computes text embeddings, and connects them into a lexical graph, which can be used to extract entities and their relationships. This knowledge graph then enables multiple retrievers to fetch data, allowing users to compare the results of different retrievers. The tool also includes features such as community summaries, local entity retrieval, multi-retrievers, retriever evaluation, guided instructions for extraction, user experience improvements, graph consolidation, and more. With its ability to generate context-rich answers and provide explainability, the LLM Knowledge Graph Builder is a valuable resource for developers looking to build more sophisticated AI-powered applications.

The Neo4j Parallel Spark Loader Python package is a solution to common issues that occur during parallel ingestion into a graph database, specifically Neo4j. It eliminates deadlocks and provides enhanced performance at high data volumes by identifying groups of relationships where no nodes are endpoints of relationships in more than one group, which can then be loaded in parallel without the risk of deadlocking. The package requires some basic knowledge of how Spark works as well as an understanding of the underlying graph structure contained in the relationships of interest. It supports three scenarios: predefined components, bipartite data, and monopartite data, each with its own module and function to create batches and groups. The package can improve load time performance around 250k row size for predefined components, while bipartite and monopartite begin to show improvement around 1.5mil rows. Additional improvements include using CREATE instead of MERGE, passing group number to ingest function, and custom property-based groups to reduce preprocessing time. The Neo4j Parallel Spark Loader Python package is useful for those looking to improve large-scale ingestion jobs by alleviating headaches caused by loading relationships in parallel and providing tools for grouping and batching data before ingestion.

Neo4j has released a new ebook, "The Developer's Guide: How to Build a Knowledge Graph", which provides guidance on building knowledge graphs. The company also announced updates to its GraphAcademy Course and showcased various product features, including AI support in Modus. Neo4j is continuing to improve its cloud-first graph technology with developments such as parallel runtime for faster analytical queries and advanced security features. The company has also partnered with Hypermode to integrate Neo4j into its Modus framework, enhancing semantic structure in AI applications. Additionally, Neo4j User Research panel invites users to participate in sharing their experiences with product development teams and influencing future products. Various events are scheduled, including conferences, meetups, and webinars, showcasing Neo4j's presence at industry events.