July 2022 Summaries

The developer community is being invited to explore inspiring talks about graphs at GraphConnect, with presentations on business cases and in-depth tech talks available for viewing. The Neo4j NODES conference is scheduled for November 16th and 17, 2022, and there is still time to submit a talk until August 20. A newsletter features articles such as Journeys in Java, which showcases building a Spring Boot application with Neo4j data, and Graph Data Science for User Segmentation, where Tomaz Bratanic applies the Neo4j GDS library to identify node roles. The community is also encouraged to participate in the Neo4j Code Golf challenge, a Cypher coding contest with prizes totaling $27,000. A featured community member, Mark Heckler, is recognized for his contributions and expertise in graph technology. His work on Spring Boot: Up & Running has been published, and he co-presented a session at GraphConnect with Jennifer Reif. Additionally, there are various resources available, including training guides, implementation tutorials, and operational tips for Neo4j users.

Graph technology has been gaining traction in recent years, and Brandon Campbell, Ontologist & Chief Architect at Northrop Grumman, is a key figure in its adoption. As a software engineer who taught himself about graphs, Brandon has seen the benefits of graph technology firsthand. He attributes his interest in Neo4j to discovering that it aligns with his personal idea of how to model anything and resonates with him on a personal level. Brandon notes that using graphs allows for flexibility and capture of emergence, which was surprising to him coming from a relational world. He also appreciates the beauty of graph structures and thinks they will become increasingly important in the future. For those just getting started with Neo4j, Brandon advises to "get your hands dirty" and learn by doing, as his own experience shows that it's essential to overcome existing knowledge and think in graphs. Ultimately, Brandon believes that graph technology will take a bigger share of the market due to its better user experience for developers and its potential in data science.

Cybersecurity threats are becoming increasingly complex and dynamic, requiring advanced data solutions to correlate and analyze connections at a real-world scale. A knowledge graph creates a digital twin of an organization's environment, enabling cybersecurity analysts to query and take action on it. The graph can be analyzed by data scientists to build models detecting malicious activities. Network infrastructures can be modeled as graphs, allowing for actionable insight and analysis. Knowledge graphs can be populated with metadata API services from cloud providers or third-party tools like Cartography and Bloodhound. This enables the creation of a digital twin of an organization's IT environment, which has tactical advantages in assessing changes to the environment, detecting potential security threats, and predicting attack paths. Graph databases like Neo4j provide a powerful platform for modeling, managing, and transforming ever-changing cybersecurity landscapes. They offer security at the database level through role-based access control and include developer tools for efficient query writing and visualization. With Neo4j's graph data science capabilities, organizations can analyze and gain insights from their graph datasets, identifying vulnerabilities and proactively applying relevant fixes.

The author of a Neo4j extension for Liquibase aims to provide graph-specific refactorings, including mergeNodes, which fuses a sequence of nodes together. The implementation poses several challenges, such as repeated node matching and dynamic property writes in Cypher. To overcome these hurdles, the author relies on retrieving node IDs in a user-defined order and using them to match nodes for subsequent queries. The final API is designed with three merge policy options: keep first, keep last, or combine all values. The implementation demonstrates the use of Cypher's dynamic property reads and concise set clause overwrites. Despite its challenges, the author believes that graph refactoring can be achieved in a way that scales well for Neo4j deployments.

Neo4j has introduced a new tool called Data Importer to help users easily import their flat file data into its graph database without writing code. The tool provides a web interface where users can input their flat file data, model nodes and relationships visually, and map the files to their model. In the latest release, Neo4j added a "Preview" button that allows users to validate their data model before importing it, saving time and increasing productivity. However, there are some limitations to the preview feature, such as only showing up to 300 relationships in total and not previewing mapped properties or isolated nodes. The tool also has some styling changes and minor bug fixes to make it tidier and more pleasant. The latest version of Neo4j Data Importer is available for free on the Neo4j AuraDB console, and users can provide feedback through the community.

Data lineage is the ability to track how data flows through an enterprise, allowing users to understand where data comes from, where it goes to, and what happens to it along the way. This is critical for gaining insights into assets and their impact on pipelines in organizations. However, achieving data lineage is difficult due to various reasons such as business vs technical view, level of detail, scope, and company evolution. Data lineage is often manual and time-consuming, making it challenging to meet regulatory deadlines and prove its accuracy. It plays a crucial role in governance and regulatory compliance by providing trust in data, enabling the identification of data quality issues, and facilitating prioritization of fixes. Companies like MANTA provide end-to-end lineage solutions that visualize data flow across various systems, including mainframe sources, operational systems, and data warehouses.

The text discusses user segmentation in a peer-to-peer payment network using graph-based features. The authors use Neo4j, a graph database, and the Graph Data Science library to analyze the network and segment users based on their roles and positions within the network. They define six features that describe user roles, including average transaction amount, years since first transaction, weighted in-degree, weighted out-degree, betweenness centrality, and closeness centrality. The authors use these features to cluster users into groups or communities using the K-means algorithm, which is a widely used unsupervised machine learning technique. The clustering results show that some users are power users who have sent vast amounts of currency to other users, while others are older accounts with smaller transaction amounts. The visualization of the clusters reveals that the split between clusters isn't very distinct, likely due to the dimensionality reduction algorithm used. Overall, the article demonstrates how graph-based features and machine learning algorithms can be used to segment users in a peer-to-peer payment network.

The Neo4j Data Warehouse Connector is a new release that simplifies interoperability between Neo4j, Spark, and popular data warehouse technologies. It provides a simple, high-level API for moving data between Neo4j and data warehouses, making it easier to integrate analytics solutions into the data ecosystem. The connector supports various data warehouse technologies such as Snowflake, Google BigQuery, Amazon RedShift, Azure Synapse, and can be used with PySpark or Scala Spark, as well as with `spark-submit` out of the box. It is compatible with Neo4j AuraDB and Enterprise Edition, including Community Edition, and can be easily integrated into workflows involving data movement between Data Warehouses and Neo4j.

The author of the article is a Developer Experience Engineer at Neo4j, who rebuilt the GraphAcademy platform on top of a Neo4j Aura database. The new platform features a revamped data model, which includes a course hierarchy and a system for storing user progress and attempts at quizzes and challenges. The author used TypeScript as the programming language and the Neo4j JavaScript Driver to interact with the database. The platform also uses Express.js and Pug templates, and is built without a front-end framework. The author praises Neo4j Aura for its ease of use and scalability, and plans to continue developing the platform and integrating existing Neo4j tools. The article concludes by inviting readers to explore the possibilities of graph databases and reach out to the author for guidance.

Graph technology is playing an increasingly important role in various fields such as cryptocurrency, supply chain management, fraud detection, and digital twin. Graph databases can encode blockchain data in ways that make it safer, more transparent, and monetizable for financial institutions. Additionally, graph-base digital twin knowledge graphs can help brands optimize their supply chains by providing a trackable and granular picture of all products, suppliers, and facilities. Graphs can also be effective in detecting fraud and anomalies by leveraging connections to stay ahead of new forms of crime. Furthermore, graph data science is moving towards mainstream adoption, with experts emphasizing the importance of understanding how to work with graphs beyond just data analysis.

The Neo4j Ops Manager (NOM) is a new tool released in June 2022, designed to boost the productivity of ops teams. It's built as a Java application and leverages Spring Data Neo4j (SDN) to persist metadata into a Neo4j database. The NOM server hosts the UI and agent APIs, and requires Java17 to run. It's part of the ongoing reactive story and aims to be fully compatible with an immutable domain-object-approach. The NOM team reworked SDN to achieve this goal, which also involved replacing SDN5+OGM in April 2021. The tool uses the reactive variants of Spring Data Neo4j, making it harder to implement retries properly. However, Project Reactor's Retry Operator can be combined with Spring's TransactionalOperator to make reactive flows fully transactional. This allows for more control over retry mechanics and mitigations.

The Neo4j Online Developer Expo and Summit (NODES) 2022 is a free, two-day virtual conference happening on November 16-17, 2022, featuring technical presentations by developers and data scientists on graph solutions. Building on last year's success with 15,000 attendees and 85 sessions, this year's event expands with additional community speakers and accommodates global time zones. Participants are encouraged to submit talks by August 31, 2022, focusing on educational content related to graph and Neo4j technologies, covering topics such as graph-powered machine learning, algorithms, and application development. The conference will include 30-minute talks, 10-minute lightning talks, and 2-hour hands-on workshops, with live presentations and interactive Q&A sessions. Attendees can explore past talks from previous NODES events for inspiration and are invited to engage with the community through forums for feedback and collaboration opportunities.

We are excited to announce NODES 2022, Neo4j’s Online Developer Expo and Summit, on November 16th and 17th! The call for papers is open now through August 20, 2022. This year's conference will feature highly technical presentations by developers and data scientists solving problems with graphs, and we want to hear from you, so submit your talk! NODES 2022 is a two-day event that includes tutorials, articles, and community spotlights, such as the featured community member Alexander Jarasch, who has expertise in Neo4j graph databases and diabetes research. Additionally, there are resources available for improving machine learning models using Neo4j Graph Data Science, including a tutorial on Event Graphs and a project by Tomaz Bratanic that demonstrates the use of graph algorithms to increase the accuracy of machine learning models. The conference also includes live coding sessions, such as Code-Golf with Michael, and tutorials like "Becoming a Knowledge Graph Jedi" and "Event Graphs".

Our team at Neo4j is excited to share a new video from our recent GraphConnect event, where M. David Allen showcases the visual power of graphs in Cypher, allowing us to spot patterns within complex systems. Using graph visualization techniques, we can create pretty pictures that represent real-world data, making it easier to understand and analyze. This quick and lighthearted talk is a great introduction to the world of graph technology, and we encourage you to check out our YouTube channel for more videos updated weekly.

Graph Data Science for Supply Chains – Part 2: Creating Informative Metrics and Analyzing Performance in Python discusses how to use Neo4j Graph Data Science and the Python client to calculate graph centrality and community metrics. The authors demonstrate how to run algorithms such as degree, betweenness, eigenvector, and Louvain community detection on a supply chain dataset. They interpret the results of these algorithms and show how they can be used in downstream statistical modeling to estimate their association with the probability of delays in business processes. The study finds that centrality scores are significant predictors of delays, with out-degree centrality having a positive effect and in-degree centrality having a negative one. The authors conclude by highlighting the potential of Neo4j Graph Data Science for supply chain analytics and look forward to exploring route optimization and what-if scenarios in the next part of the series.

The Neo4j GraphQL Toolbox is a user interface that allows users to write and execute GraphQL queries and mutations against their Neo4j database with minimal setup, making it easier for new users to get started with GraphQL. The toolbox includes features such as auto-generated type definitions based on the data in the Neo4j graph database, a built-in GraphQL query builder, and a server setup guide. It is built using Typescript and React, and is open-source, allowing users to provide feedback and feature requests. The tool is designed to help users explore and try out GraphQL with Neo4j quickly, and can eventually be used to build powerful GraphQL APIs.

The Neo4j Code Golf contest, announced recently, aims to optimize Cypher queries for performance, with participants competing to write the most efficient code. The contest offers a total of $27,000 in prizes and is open to beginners and experienced golfers alike. Graph databases are highlighted as an important area of development, with various resources available, including articles on machine learning and graph analytics. Additionally, there are featured community members, such as Ghlen Nagels, who share their expertise and projects related to web development and PHP. Various events, including a Berlin GraphDB meetup and GraphSummit, are also mentioned, providing opportunities for developers to connect and learn from each other.

The Usha Renal Care Foundation is working on a project called NephroBlocks, which aims to connect fragmented data in the healthcare industry using graph databases and blockchain technology. The goal is to create a knowledge repository of nephrology data that can be accessed by patients and medical institutions, making it easier for them to find relevant clinical trials and improve healthcare outcomes. The foundation has already populated over 50,000 clinical trials related to renal care in their database, and they plan to use this platform to connect patients with the right clinical trials at the right time, while also enabling researchers to collaborate more effectively. The project is motivated by a personal story of loss and a desire to leverage technology to save lives, and it has the potential to create a safer and more secure ecosystem in the healthcare industry.

The text discusses using Neo4j Graph Data Science in Python to improve machine learning models. The author provides a simple demonstration of how graph-based features can increase the accuracy of a machine learning model, specifically in a fraud detection scenario. They use an anonymized dataset from a P2P payment platform and train a baseline classification model based on non-graph-based features. Then, they explore graph-based features such as PageRank centrality and community detection to improve the model's accuracy. The author finds that incorporating these features results in a more accurate model, with improved performance metrics including higher AUC scores and reduced misclassification rates. They conclude by emphasizing the importance of exploring relationships between data points in datasets to extract predictive graph-based features for downstream machine learning tasks.

The article analyzes the Roland Garros and US Open tennis tournaments using Neo4j graph database. The author, a data scientist and certified Neo4j professional, explores various aspects of the tournaments, including player performance, tournament winning streaks, and champions' routes to victory. The analysis reveals interesting insights into the players' performances, such as Rafael Nadal's dominance in men's singles and Serena Williams' dominance in women's singles. The author also highlights the benefits of using Neo4j for analyzing sports competitions and tournaments, citing its ability to efficiently process large amounts of data and provide simple yet powerful queries. Overall, the article showcases the potential of Neo4j as a tool for exploring complex data and uncovering hidden patterns and relationships.

The Neo4j Ops Manager is a new product that makes it easy to monitor and administer all Neo4j databases, instances, and clusters from a central user interface. It allows users to instantly view the status of their databases, inspect instance health, and set up role-based access for each user. This product aims to increase productivity and improve the reliability of Neo4j deployments. Additionally, there are various resources available, including a new GraphAcademy course on building Neo4j applications with .NET, a blog post about the product, and information about local GraphSummits and meetups. The community is also highlighted through featured members, such as Michael McKenzie, who is a graph database developer and a long-time member of the Neo4j community.