April 2022 Summaries

The text delves into various topics related to graph technology and data science, including knowledge graphs, graph data science, and Neo4j. Ali Emre Varol uses Graph Data Science 2.0 to analyze the European gas distribution network, while Jennifer Reif continues building a microservices application in Java. The article also features Neo4j-related content, such as using GDS algorithms to determine critical nodes, creating a Kubernetes cluster on GCP, and exploring the SciGRID_gas dataset. Additionally, it discusses new releases and updates for Neo4j, Structr, and Apache Hop, highlighting improvements in performance, usability, and data import handling.

The COVID-19 pandemic led to a significant increase in cyberattacks as businesses shifted to digital platforms, resulting in a 300% rise in cybercrime reports and a 650% increase in software supply chain attacks. Cybersecurity professionals face challenges in defending against attackers who think in graphs, while defenders traditionally rely on lists, creating blind spots. Graph databases can help defenders build a complete graph of their infrastructure, enriching it over time, and provide a high-fidelity model of the real world, integrating many data sources and revealing dependencies. By modeling infrastructure as a graph, defenders can identify valuable assets, generate alerts for relevant teams, analyze suspicious behavior, and conduct ongoing cyber risk assessment, ultimately gaining predictive, preemptive, and proactive threat identification and management capabilities.

The European Natural Gas Network can be turned into a knowledge graph using Neo4j to analyze its data. The network consists of various components such as pipelines, compressors, LNGs, storages, consumers, and power plants. Each component has a unique structure with specific attributes like node_id, lat, long, country_code, etc. The knowledge graph is created by defining the constraints and then creating the components. Exploratory data analysis (EDA) and graph data science (GDS) are used to analyze the graph. GDS algorithms such as PageRank, Degree Centrality, Betweenness Centrality, and Cluster Detection Via Louvain Modularity are applied to find important nodes and communities in the network. PathFinding is also used to determine the shortest path between two nodes. Finally, NeoDash is used to visualize the graph and create interactive dashboards.

The COVID-19 pandemic accelerated the shift to digital platforms, resulting in a significant increase in cyberattacks in 2020, with a 300% rise in reported cases and a 650% increase in software supply chain attacks. Cybersecurity professionals face challenges in defending against attackers who think of networks as graphs, exploiting vulnerabilities to gain access to more systems and data. Traditional list-based approaches are insufficient, while graph databases can capture the complexity of IT infrastructure and security tools, enabling defenders to build a complete graph of their infrastructure and create a digital twin to analyze connections at a real-world scale. This approach offers advantages in identifying valuable assets, generating alerts, detecting suspicious behavior, and conducting ongoing cyber risk assessment, ultimately providing predictive, preemptive, and proactive threat identification and management.

We recently hosted an exciting webinar featuring four talented women who are making significant contributions to the field of knowledge graphs. The panel, moderated by Dr. Maya Natarajan, a leading expert in knowledge graphs, included Dr. Ashleigh Faith from EBSCO, Shilpa Karkeraa, the CEO and founder of Myraa Technologies, Michele Howard from DXC Technology, and Yan Yee Yap from Deloitte Consulting. These women shared their experiences and insights on how knowledge graphs are driving innovation in various industries, including healthcare, education, and logistics. Their work has the potential to accelerate medical discoveries, optimize employee experience, and improve supply chain efficiency, among other benefits.

Michael Hunger and I collaborated to create a hands-on course that guides users through writing code to build a completed IMDB-clone application with Neo4j and Java, using the SparkJava framework. The data used in the app comes from the MovieLens recommendation dataset and themoviedb.org movie and cast data. The app presents results to the user through a webpage supported by Vue.js for a user-friendly experience, focusing on the backend aspects of Java without agnostically covering other web frameworks like Spring or Quarkus.

The text discusses the importance of choosing a compatible data structure for an application, highlighting that different database management systems store data in various structures. It then introduces the Neo4j Under the Hood series, which explores the features of graph data models and their uniqueness in modern application development. The series aims to provide insights into the nuts and bolts of graph database technology, making it a valuable resource for developers interested in this topic.

Neo4j Bloom is a tool that helps users connect and interact with their graph data. It provides an easy-to-use interface for exploring relationships and properties in the database. The new version of Bloom, 2.2, introduces "Scene Actions," which allow users to execute custom Cypher queries on specific user selections of nodes and relationships. This enables users to perform complex operations and automate repetitive tasks, such as creating new data or updating existing data. Scene Actions can be used for various applications, including retrieving data based on shared property values, aggregating functions, subqueries, and procedures. They can also be applied to relationships and limit the scope of applicability to a single relationship type. The tool is designed to empower users to efficiently manipulate graph data and perform business operations in a safe and controlled manner.

With the latest release of Neo4j Data Importer, users can expect improved handling of empty ID fields, which now treat empty strings as null values and filter out rows with empty IDs. The new feature aims to prevent supernodes from forming due to missing IDs. Additionally, the release provides a setting for customizing null values in load configurations, allowing users to specify string values that should be treated as nulls. The Data Importer also improves its representation of partial mappings and errors, making it clearer when certain fields are not fully mapped or when there are errors in the graph model. Furthermore, the release includes small improvements to load progress displays and error handling for type conversion failures.

The speaker, a Machine Learning Engineer, is discussing the challenges of identifying repeat offenders in counterfeit goods imports into South Africa. The company was using a database to track this information, but found it difficult to manage due to the large amount of unstructured data and complex relationships between entities. They were considering using a SQL or document database, but ultimately chose Neo4j as the solution. Neo4j's graph database structure allowed them to easily identify repeat offenders and visualize their connections in a graph form, which was appealing to their clients. The implementation involved creating a knowledge graph with one type of relation, extracting entities from documents, and building queries to find connections between them. The use of Neo4j AuraDB helped deploy the solution in a cloud environment. The speaker concludes that Neo4j is a powerful tool for solving complex problems like this, and recommends exploring its solutions.

The text discusses the challenges of managing complex supply chains, particularly in a global marketplace. It highlights the need for precise information about inventory, product distribution, and logistics to stay competitive. The author suggests that graph databases, such as Neo4j, can help model complex relationships and identify patterns and insights that would be hidden in traditional tabular data sets. The text then describes a demo application built using Neo4j, focusing on fleet management and warehouse distribution center mechanics. It covers the API development, building the graph data model, project structure, front-end coding, and bringing the code, API, and database backend together. The demo showcases an interactive web application that supports multiple cities (globally distributed) and demonstrates how Neo4j can help with supply chain analytics, cost control, planning, risk management, supplier-partner relationships, and consumer demands.

The Graph Data Science team has released a new version of their toolset called GDS 2.0, which includes a Python client and several new courses in the Neo4j Data Scientist learning path. The release is accompanied by a five-minute video demonstration of the new features, as well as two new free training courses: Introduction to GDS and GDS Fundamentals. Additionally, there are articles and interviews available on various topics related to Neo4j, including a simple recommendation engine implementation using Cypher and an overview of Neo4j technologies from a developer's perspective. The release is also accompanied by a tweet from Soliman ElSaber, encouraging users to retweet the post if they like it.

This week in Neo4j, the company announced Graph Data Science 2.0 and AuraDS releases, which are designed to make data science more accessible to developers. The new feature includes a Python client for Graph Data Science 2.0, a free training program called #GraphAcademy, and courses on building recommendation engines in Neo4j. Additionally, the company featured an interview with Michael Simons about using Neo4j's OGM with Quarkus, as well as articles on implementing simple recommendation engines and Daniel Starns' journey working on Neo4j GraphQL. The week also saw a tweet from Soliman ElSaber, where he shared his experience playing with the Egyptian football league clubs data in Neo4j.

Graph technology, specifically Neo4j, a leading graph database, is emerging as a defensive leader in combating modern cyber threats. Organizations worldwide are leveraging its power to identify vulnerabilities, analyze network health, and visualize unpredictable patterns. A recent event, "Cybersecurity: Combating Modern Threats with Graph Databases", brought together experts and thought-leaders to share insights on the role of graph technology in cybersecurity. The event covered topics such as digital twinning for DevSecOps, the attackers' perspective on CI/CD graphs, graph-based threat modeling, and real-life case studies using Neo4j's graph database. These sessions provided a comprehensive understanding of how graph databases can help organizations create value from their data relationships and protect themselves against cyber threats.

The author, a Developer Experience Engineer at Neo4j, discusses their experience with handling authentication and identity using Auth0 and Neo4j. They initially used express-openid-connect to add authentication to the GraphAcademy site, which is built in TypeScript and backed by a Neo4j Aura database. However, they encountered issues when multiple users had the same email address but different authentication methods, resulting in separate entities with unique IDs. To handle this, they created a :HAS_ALIAS relationship between the two nodes with the same email address. The author also leveraged the *0..1 technique to unify user enrollments at query time by using variable-length paths and Neo4j's ability to expand relationships. Additionally, they mentioned that Auth0 allows for email verification, which ensures that only verified emails are linked to a user's account. The GraphAcademy site is now able to handle authentication and identity with ease, providing a seamless experience for users.

Neo4j Aura, a cloud-native graph service, has introduced the ability to pause database instances when not in use. This feature allows users to pause their instance on demand with just a click and pay only 20 percent of the regular running cost while paused, retaining data, backups, and configurations securely. Pausing a database is as simple as clicking the pause button, and resuming it can be done by clicking the play button after selecting "I understand" in a confirmation dialog. This feature is available in Professional and Enterprise tiers, with Free Tier instances auto-paused, and databases can be paused for up to 30 days before being automatically resumed with the latest updates and patches. Additionally, AuraDB Free databases are paused automatically after 72 hours of inactivity, and if remaining paused for more than 90 days, they will be deleted permanently.

Neo4j Graph Data Science is a comprehensive graph analytics and modeling platform that uses graph algorithms and machine learning to identify patterns and behaviors in data, improving models for use across various applications such as recommendation engines, fraud detection, and customer 360 scenarios. The platform is being developed with five key areas of focus: ease of use, enterprise readiness, graph built for data science, ecosystem integration, and cloud, resulting in new features including machine learning pipelines, best-in-class data science algorithms, and a native Python client. Additionally, the platform offers a managed service AuraDS on Google Cloud Platform, making it easy to deploy and access Neo4j Graph Data Science without requiring extensive technical expertise or IT approval.

The New York Times uses a knowledge graph to store article metadata, which can be accessed through their API. The graph is built using Neo4j AuraDB Free and includes nodes for articles, topics, people, organizations, and locations. The data model is based on the API response, with each node having properties like title, id, url, byline, source, published_date, abstract, des_facet (topics), org_facet (organizations), per_facet (people), and geo_facet (locations). The graph can be imported into Neo4j using the `apoc.load.json` procedure, which returns the response of the API as Cypher data structures. The nodes are then created and relationships are established between them. The graph can be queried to find articles by date, title, or topics, and can also be used for recommendations based on topic overlap. Other metadata like photos, authors, and geolocation can be extracted from the article data and added to the graph. The graph can be further extended with additional entity extraction and data from other publications and sources.

The text discusses various topics related to graph databases and their applications. Moriba Jah leads a research program at the University of Texas at Austin that focuses on space safety, security, and sustainability, where he uses ASTRIAGraph to help solve the problem of space junk. Jesús Barrasa explores how to use SHACL to control the structure of graphs in this episode of Going Meta. Additionally, Chris Gioran discusses scalability challenges in graph databases, while Simon Martinelli visualizes a graph using Vaadin and Neo4j to represent business functions and messages. The text also includes a tweet from Guy Clapperton about an interview podcast discussing space garbage with Prof. Moriba Jah.

Stratazon is a boutique consulting company that utilizes insights and technology to complement companies' existing work, helping them reach their strategic goals. They use graph technology to gain real-time insights into the customer journey and understand how customers interact with products. By analyzing reviews and transactional data, they identify patterns and trajectories in client behavior, providing actionable insights for businesses. Stratazon's approach is hyper-charged from a business perspective, focusing on creating value for clients through transformation and innovation. They partner with Neo4j to create knowledge graphs that can quickly identify key data points and provide real-time analytics and recommendations. By leveraging graph technology, Stratazon helps businesses understand customer behavior, optimize their product portfolios, and drive growth.

Neo4j is a leading graph database platform that offers native graph storage and processing, ensuring the integrity and performance of graph databases. It provides flexible native property graphs, index-free adjacency, and supports Cypher and Gremlin query languages. Neo4j also excels in data ingestion and integration, offering high-speed bulk ingestion, rapid recovery from system failures, and high availability. The platform is designed to scale horizontally and vertically, providing flexibility for deployment on-premises or in the cloud. With a strong open source foundation and community, Neo4j has partnered with numerous business and technology partners to provide a comprehensive ecosystem of tools, connectors, and services. Its vendor credibility and resilience are also noteworthy, with substantial funding, a large engineering team, and a dedicated services team. Overall, Neo4j is well-suited for modern connected data applications, offering significant ROI and unparalleled business and technical benefits.

There's a wealth of fine content and varied perspectives on graphs in this week's newsletter, including articles on transferring unstructured text into knowledge graphs, analyzing data for ML projects with Neo4j, the uses of graph databases in cybersecurity, and Kubernetes deployment with the Neo4j Helm chart. The newsletter also explores how knowledge graphs can represent domains of knowledge, and discusses using Neo4j to analyze COVID-19 treatments and discovering new treatments by combining multiple domain graphs. Additionally, it covers the ability of knowledge graphs to transform unstructured text inputs into a useful knowledge graph through an information extraction pipeline, demonstrates ML on graphs with Neo4j and Python, and showcases how Neo4j can be used for cybersecurity and Kubernetes deployment.

The presentation discusses the challenges of tracking and managing anthropogenic space objects, also known as space debris. There are currently over 26,000 objects in Earth's orbit, with about 4,000 working satellites providing services such as navigation, communication, and weather forecasting. However, most of these objects are dead or de-orbiting, posing a hazard to operational satellites and potentially causing collisions. The speaker argues that it is essential to develop a global system for tracking and managing space debris to prevent further pollution and ensure the long-term sustainability of space exploration. To address this challenge, the presenter introduces ASTRIAGraph, a knowledge graph built on Neo4j, which aggregates information from various sources to provide a comprehensive view of space objects. The graph is used to predict potential collisions, identify areas for improvement in space debris management, and support research into space safety, security, and sustainability.

In a new video series, Jesus Barrasa and Alexander Erdl delve into the world of graph query languages, comparing Cypher and SPARQL in a hands-on session. The first installment explores how these two models and languages relate to and differ from one another. This series aims to provide an intimate deep-dive into graph query languages for those interested in gaining a better understanding of their capabilities and limitations.