October 2023 Summaries

This week's edition of This Week in Neo4j covers the excitement surrounding NODES2023, where attendees had a great day with many interactions and sessions. The event also saw several product announcements, including Parallel Runtime Capability, Change Data Capture, and Easier Knowledge Graph Creation. Additionally, there are featured nodes speaker Stephen McGowan, parallel runtime for speeding up queries, geospatial app U.S. Flu Vaccine Provider Search, LangChain article, RAG with Vector Search, and a tweet of the week about AD Miner, an Active Directory audit tool that leverages cypher queries to crunch data from BloodHound graph database.

Neo4j has introduced major new capabilities for operational and analytical workloads, enabling cloud customers to drive up to 100x faster analytics and real-time decision-making on a single database. The Parallel Runtime uses morsel-based parallelism to split complex analytical graph queries into concurrent threads across multiple CPU cores, increasing query speeds by up to 100x. Native Change Data Capture automates tracking and notification of database changes for mission-critical decisions in real time. Knowledge Graph Embedding models find missing relationships and infer new connections within knowledge graphs for greater semantic understanding. New Pathfinding Algorithms make complex workflows more efficient by identifying critical paths between graph nodes and determining the best sequence for dependent tasks. These new capabilities are designed to help organizations dramatically accelerate graph analytics and develop the next wave of real-time, mission-critical applications.

The new Parallel Cypher Runtime is a fantastic alternative for executing graph-global read queries that can drastically improve query speed on databases with multiple CPUs available. It's ideal for users who need to process large sections of the graph, such as those engaged in explorative work like data analytics or investigative journalism. However, it may not be suitable for high transactional workloads, write statements, or graph-local queries, where pipelined runtime is still a better choice. Advanced Neo4j users can benefit from running different queries on different runtimes to customize their performance based on specific workload needs. The parallel runtime is optimized for large-scale databases with multiple CPUs, and its performance will likely improve in future releases.

Needle version 2 is a significant leap forward in Neo4j's commitment to design excellence, consistency, and a DesignOps-driven culture. The new brand identity reflects the natural world around us and includes new colors, fonts, and shapes, with a focus on efficiency, consistency, and accessibility. The update aims to empower product teams to iterate and adapt quickly, with improved user interfaces, automated processes for adopting the new brand colors, and comprehensive documentation. A well-functioning design system now bridges the gap between web, marketing, and product teams, fostering collaboration and providing a common language that brings everyone on the same page when it comes to design and branding.

The latest update for Neo4j DevTools includes several significant features aimed at improving the development experience in Workspace. A new Cypher query editor has been introduced, offering improved syntax highlighting and code completion. The editor now uses a rendering similar to that used in Neo4j Browser, making it easier to identify nodes, relationships, and paths in tables. Additionally, the Preview feature has been enhanced to allow users to preview only selected parts of their graph, and a new option to generate Cypher scripts for import logic has been added. Furthermore, Workspace now supports multi-database connections and allows users to switch between databases, enabling more efficient development with Neo4j instances. These updates aim to enhance the overall usability and productivity of the Neo4j DevTools suite.

The project leverages the capabilities of Neo4j Vector Index and GraphCypherQAChain with Mistral-7b to provide a robust system for handling complex data that effectively bridges the gap between voluminous unstructured data and intricate graph knowledge, providing a comprehensive and accurate response to user queries by synthesizing information from both data sources. Utilizing Neo4j for both vector similarity search and graph database retrieval ensures that the responses generated are not only informed by the vast pre-trained knowledge of Mistral-7b but are also contextually enriched and validated by real-time data from the vector and graph databases. The implementation demonstrates a practical application of retrieval-augmented generation, where the synthesized information from diverse data sources is utilized to generate responses that are a harmonious blend of pre-trained knowledge and specific, real-time data, thereby enhancing the accuracy and relevance of the responses to user queries.

This week in Neo4j brings exciting news and updates for developers and graph database enthusiasts. The highly anticipated NODES 2023 event is finally happening this Thursday, featuring over 100 sessions on various topics including Application Building, Visualization, AI/ML, and more. In addition to the live stream of the event, there are numerous upcoming conferences, webinars, and events, as well as a GraphSummit Series that will connect users with graphs. The featured nodes speaker is Elena Kohlwey, who will discuss user-defined procedures for full-graph or big-subgraph traversals. The GenAI Stack has been launched, providing a powerful tool for building AI apps capable of generating text, code, and other creative content. Additionally, there are articles on GraphQL development best practices, the integration of Vectors with knowledge graphs, and the latest update to Data Explorer, which enables users to engage in conversations with their database. The Neo4j team is also excited about Hannah Fry's keynote speech at NODES 2023.

Neo4j recently announced the formation of a GenAI Stack together with LangChain, Ollama, and Docker. A hackathon is currently underway with a $20k prize. Neo4j also introduced a patent chatbot that uses retrieval augmented generation (RAG) to improve accuracy. Additionally, Hannah Fry will deliver a keynote at NODES 2023 titled "The Joy of Data". Guido Thys, a database expert, will introduce the "superconnector" - a knowledge graph designed to consolidate disparate data sources. The GenAI Stack is being showcased through a blog post that demonstrates how to implement a support agent using RAG and other best practices. Building an LLM chatbot on top of Neo4j's patent database was also inspired by presentations at GraphSummit Minneapolis. Furthermore, there are upcoming live streams and conferences, including NODES 2023, All Things Open 2023, and the GraphSummit Series. Guido Thys will be speaking at NODES 2023 about his "superconnector" project. The Neo4j Developer Podcast is also previewing NODES 2023 with Melly Beechwood, a Machine Learning Engineer with a passion for data exploration.

The document discusses various best practices for developing GraphQL APIs. It covers topics such as sharing design early, being minimalist with features, using consistent and symmetric naming conventions, being clear in schema definitions, and avoiding generic fields and runtime logic. The document also explores pagination strategies, including offset, cursor, and Relay connections, and how to implement global identification, nullability, abstract types, static queries, mutations, fine-grained vs coarse-grained mutations, error messages and notifications, security measures such as rate limiting, authentication and authorization, blocking introspection, persisted queries, and provides a conclusion summarizing the key takeaways.

Welcome to This Week in Neo4j, your weekly fix for news from the world of graph databases!` We are now at episode 20 of `Going Meta`, a series that celebrates the latest developments and success stories in the Neo4j community. The upcoming event `NODES 2023` is just around the corner on October 26th, featuring an exciting agenda and the final `Road to NODES` workshop: Geospatial Data. `Spring AI` has now integrated with Neo4j as a Vector Store, while OpenAI GPT-4 is being used in a blog post to demonstrate how to build a conversational agent capable of querying a Neo4j database. Meanwhile, the `Open Ownership` project is gaining traction, showcasing the potential of beneficial ownership using Neo4j and other tools.

Supply chain intelligence is a systematic approach to collecting, analyzing, and acting on data to improve performance across all aspects of the supply chain. This approach enables companies to acquire real-time information that strengthens business decision-making and operational efficiency. Supply chain intelligence helps companies improve risk management by identifying and mitigating risks and rapidly responding to crises. It also enables faster time to market, better customer service, and cost control through improved forecasting and operational efficiency. With the increasing complexity of modern supply chains, adopting a data-driven approach has become crucial for competitiveness and profitability. Supply chain leaders are focusing on end-to-end visibility, forecasting, real-time decision-making, cost control, sustainability, and supplier interdependencies to achieve these objectives. To achieve these goals, they are turning to supply chain intelligence and the data technologies that support it. Graph databases have emerged as a key technology for supply chain management due to their ability to model complex relationships and handle large volumes of interconnected data. By leveraging graph databases, companies can visualize their supply chains, optimize partnerships, and respond to disruptions in real-time, ultimately driving modernization and future-proofing of the supply chain.