April 2023 Summaries

The new graph database plugin for IntelliJ provides useful developer features such as syntax highlighting and autocompletion for Cypher, allowing developers to work efficiently with graph databases. The plugin supports Neo4j 3.4+ (Bolt), including AuraDB, and is maintained by Alberto Venturini. This week also marks the start of various exciting events and projects related to graph databases, including Neo4j Live, GraphSummit Series, and LangChain's integration with a graph database. Additionally, Kineviz shares insights on GPT4-powered querying in GraphXR, allowing developers to ask questions about their data and get relevant results in seconds.

Apna, one of India's largest jobs and professional networking platforms, has leveraged Neo4j's graph database and analytics offering to create a more personalized and efficient job search experience for its users. By utilizing the power of Neo4j, Apna addresses unique challenges faced by India's emerging workforce, providing features such as personalized connections, skill-based communities, and job recommendations. This approach improves user engagement, enhances trust and safety, and ensures scalability and performance, ultimately transforming the job search landscape in India.

The text covers various topics related to Neo4j and its ecosystem. A tutorial by Sofia Vei and her team demonstrates how to analyze Twitter data with reusable Python utility functions, making it useful for anyone doing similar ETL work. Tomaz Bratanic's blog post showcases a chatbot that uses GPT-4 endpoint to generate Cypher statements, while using the Neo4j database to avoid chatbot hallucinations. Additionally, Friedrich Lindenberg discusses his open-source project OpenSanctions, which tracks company ownership registrations and is of interest to both investigative journalists and law enforcement. The text also mentions upcoming events such as Neo4j Live, Node.js 2022 presentation by Christophe Willemsen, and conferences like GIDS 2023.

Today marks a special occasion to celebrate the beauty and elegance of graphs, honoring Leonhard Euler's birthday. Swiss mathematician Euler was fascinated by the Seven Bridges of Königsberg puzzle, which he analyzed from a mathematical standpoint rather than focusing on geographical layout. His innovative approach led him to discover that it is impossible to cross all seven bridges without retracing steps due to each land mass having an odd number of bridges connected to it. Euler's work in graph theory paved the way for processing and analyzing complex data, revolutionizing modern times. To celebrate this occasion, join fun activities, giveaways, test your knowledge with a crossword, and participate in an ebook raffle to win comprehensive guides on graph databases and analytics.

Neo4j is partnering with AtomicJar to offer preconfigured implementations of various dependencies through Testcontainers modules, making it easier for developers to write tests. The team is seeking feedback from users of Neo4j drivers in multiple programming languages and invites developers to present topics at local conferences and meetups. Several upcoming events and tutorials are highlighted, including a presentation on data governance and a tutorial on building a simple web app with Neo4j. Additionally, a developer has created a tool called GraphInsight that uses GPT4 to automatically create materials knowledge graphs.

GraphSummit 2023 is taking place with 16 events across 14 locations in EMEA, offering opportunities for professionals to meet and learn from graph experts and explore new use cases. The series features diverse organizations such as Kerry Group, EY, Volvo Cars, and GSK sharing their experiences with graph technology, including applications in customer 360, fraud detection, trend aggregation, recommendation engines, clinical reporting, and network analysis. Notable speakers include Katariina Kari discussing knowledge graphs, Ali Khalili and Jeroen Vermeulen on OXYGEN Data Insights Monitor, Joakim Sandroos on explainability and causality in ML Ops, Noel Yuhanna delivering the opening keynote on typical data challenges solved by graphs, and Jérémy Grignard presenting pharma data exploitation using a Knowledge Graph. The event is designed for business and technical professionals at all levels of graph experience, with exciting speaker lineups and live events available across EMEA locations.

Neo4j has launched its Graphs4APAC initiative to provide Asia Pacific professionals with free training, materials, and tools to learn and enhance their graph technology skills. This is part of Neo4j's commitment to giving back to the regions it serves through an open source approach. The initiative aims to equip graph-skilled professionals with the knowledge, skills, and certifications needed to succeed in a rapidly growing field. Graph databases are becoming increasingly important for rapid organizational decision-making, and the demand for graph-skilled professionals is on the rise. Neo4j's Graphs4APAC initiative targets specific countries/regions, including Indonesia and Singapore, and offers comprehensive online training courses, certification opportunities, and access to free resources like books and documentation. The initiative's goal is to enable, empower, and engage the next generation of APAC graph-skilled professionals through a range of activities, including training, meetups, events, and webinars.

Friedrich Lindenberg has created an open-source project called OpenSanctions, which tracks sanctions and financial restrictions on assets and individuals. The project aims to provide a centralized hub for journalists and researchers to discover connections between people and organizations involved in sanctions, as well as uncover hidden relationships and illicit activities. With the increasing use of data leaks, graphs, and international collaboration, OpenSanctions is designed to bridge the gap between data availability and analysis, making it easier for users to explore and visualize complex financial restrictions. The project's goal is to empower investigative journalists and law enforcement agencies by providing a powerful tool to uncover hidden relationships and track sanctions targets.

The chatbot uses OpenAI's Chat Completion endpoint with GPT-4 to generate Cypher statements based on user input. The model is trained on a knowledge graph database, which provides complete and total control over the answers provided by the chatbot. The chatbot can answer follow-up questions that rely on information provided in the conversation history, making the whole discussion smoother. It has excellent value for answering follow-up questions and can make the user feel like they are talking to someone who understands their meaning. However, it's recommended to use GPT-4 when possible, especially for generating Cypher statements, as it is better at following instructions and understanding the given task. Additionally, the model should only return Cypher statements, but it might override system instructions and return previously provided database results in some cases. The code is available on GitHub.

What’s happening this week? We have the launch of Neo4j AuraDS Enterprise, a fully managed graph data science cloud service available on Microsoft Azure. This allows users to integrate and analyze their graph data science pipelines using Azure Machine Learning and Azure Synapse Analytics. Additionally, we can connect Neo4j AuraDB instances with Google BigQuery for seamless integration. Furthermore, developers can now use Neo4j AuraDB as a source and sink for AWS Managed Streaming for Kafka (MSK), enabling more flexibility in their data management. The company is also promoting community engagement through various events and training series, including Neo4j Live, GraphSummit Series, and Spring Data Neo4j training series. These initiatives aim to provide users with more opportunities to learn and share knowledge about graph databases and related technologies.

The Neo4j Hackathon's Cypher Code-Golf challenge encouraged developers to solve three questions in increasing difficulty levels, with the efficiency of their queries measured by "Database Hits" and query length. Over 5,000 queries were submitted by more than 1,000 participants, and 18 winners received cash prizes. The challenges ranged from finding the maximum number of tags for a question in the Cypher area to identifying the top 10 tags most frequently co-occurring with Lua, requiring users to aggregate data, filter results, and use various techniques like parameterized queries and regular expressions. Despite some issues with invalid queries and lack of capacity to provide explanations, the event provided valuable learnings and a platform for developers to showcase their skills.

Neo4j's Aura platform is enhancing its security features with the release of User Management, introducing a new tenant entity that allows for logical segregation of resources across different environments. This feature provides an increased level of autonomy for security-focused operations and enables users to create a single tenant that spans multiple products within Aura. The introduction of roles, such as Member and Admin, also streamlines user management, allowing admins to manage core aspects of user management themselves. Additionally, the tenant switcher component enables authorized users to jump between tenants without re-entering credentials, promoting a strong security posture and best practices regarding production environment segregation. With these enhancements, AuraDB and AuraDS Enterprise users can now enjoy an increased level of flexibility and security.

Sudhir Hasbe has been appointed as the Chief Product Officer at Neo4j, bringing his extensive experience in data analytics and machine learning from his previous role at Google Cloud to drive the company's product innovation and market leadership in graph database technology. Sudhir is enthusiastic about the transformative potential of graph technology, which can address complex data challenges across various applications, such as fraud detection and recommendation systems. He aims to make graph technology more accessible and expand Neo4j's product portfolio. Additionally, Philip Rathle has been promoted to Chief Technology Officer, where he will focus on market and customer engagement, further strengthening Neo4j’s position in the industry. Both appointments are part of Neo4j's strategic efforts to leverage connected data structures in solving pressing business challenges while fostering a culture of innovation and collaboration.

The updated Neo4j-Java-Driver provides a powerful API for querying and manipulating graph data. It introduces an `executableQuery` method that allows developers to create and execute Cypher queries, which can be enriched with parameters and used as an eager result. The driver also supports Java's Collectors API, enabling client-side grouping and counting tasks. The new API is suitable for scripts, simple Java programs, and projects without external transaction boundaries. It offers features such as transactional functions, retries, and native support for the Collectors API, making it a valuable tool for developers working with Neo4j graph databases.

The author uses GPT-4 as an information extraction model to extract relevant entities and relationships from a video transcript. The transcript is obtained from a nature documentary, specifically about the deep sea ecosystem. The author groups the transcript into sections based on pauses in narration and then uses GPT-4 to extract entities and relationships. The extracted entities are related to biology, chemistry, or archeology, and include animals, locations, and biological entities. The model is able to generalize across different domains and provide consistent results. However, the author notes that the model's performance can vary depending on the input data and that it may hallucinate information such as citations. The extracted relationships are provided in a triple format (Head, Relationship, Tail). The author also uses GPT-4 for entity disambiguation by asking the model to identify which values reference the same entity. The results show that the most mentioned animals are moray eels, lionfish, and brittle stars. The knowledge graph is used as a search engine to retrieve timestamps of sections where relevant entities are mentioned. The author concludes that GPT-4 is a powerful tool for exploring and analyzing different datasets to extract relevant information.

The newsletter covers various topics related to Neo4j graph database, including ingesting big data into Neo4j, describing property graph data models, and using graph neural networks for word embedding. It also mentions upcoming events such as Neo4j Live, GraphStuff.FM, and Meetups. Additionally, it highlights a recent release of Graphy McGraphface, a tool that simplifies data analysis, and shares a tweet from Christophe Willemsen about the new security features in Neo4j 5.6. The newsletter encourages sharing and learning within the community by inviting readers to present at Meetups and local conferences.