November 2017 Summaries

This week in Neo4j highlights the power of graph databases in software analytics, data integration, and various applications. Markus Harrer, a featured community member, presented on software analytics with Jupyter Notebooks and Py2Neo to query and visualize data from software project structures, commits, and source code. The Paradise Papers sparked activities, including releasing updated Neo4j ICIJ Sandbox data and publishing blog posts on graph analytics and map visualizations. Releases include Neo4j 3.2.8 with small fixes and the next alpha of Neo4j 3.4.0 scheduled for next week. Various articles were published on topics such as DevOps, data architecture, and healthcare data, showcasing Neo4j's capabilities in different domains. Cool projects like TableTop-Generation and RepoYouMayLike were shared, and good answers were given to StackOverflow questions related to Cypher and graph algorithms. The FOSDEM 2018 Graph Developer Room submissions are due this weekend, and the tweet of the week features Manuel Villa's experience with the ICIJ Data Fellow program.

The Graph DevRoom will take place at FOSDEM 2018, an annual conference focused on Open Source technology. The event is part of the Developer Rooms series and aims to share ideas, meet contributors, collaborate, and feature Belgian beer. The room will focus on various aspects of graph processing and databases, including graph databases, query languages, distributed processing frameworks, semantic graphs, and real-life applications. The conference is free to attend, and attendees can submit proposals for talks or demos until November 26th. The event promises to be an exciting experience with the support of Neo4j and university partners.

This week in Neo4j highlights the latest developments and community news from the world of graph databases. The featured community member is Nicolas Mervaillie, a Senior Consultant at GraphAware who has been working on various Neo4j projects for several years. He recently presented his project on analysing till receipts in Neo4j and discussed it on the Graphistania podcast. Additionally, he is a core developer on Spring Data Neo4j and Neo4j OGM products, and consults with clients on Neo4j projects. The online meetup covered the recent release of Spring Data Neo4j 5.0 and its new features. Will Lyon has been working on applying Natural Language Processing techniques to analyse Russian Twitter trolls' tweets, and Michael Hunger wrote a blog post on importing the Relato Business Graph into Neo4j. Other notable mentions include David Allen's work on NLP in Cypher and Máté Thurzó's cloud-based natural language processing framework using Neo4j and Spring Boot. The upcoming events include a talk on eBay ShopBot at GraphConnect NYC 2017, and the video from Will Lyon's ReactNYC talk is now available.

The International Consortium of Investigative Journalists (ICIJ) has released over 13 million leaked documents, emails, and database records related to offshore finance, which were analyzed using text analysis, full-text- and faceted-search, graph visualization, and graph-based search. The data is stored in a Neo4j graph database, allowing for powerful analysis of the relationships between entities. Graph querying techniques can be used to find connections between nodes, including indirect paths between public figures like Rex Tillerson and the Queen of England. The data also includes geocoded addresses, enabling geographic analysis and visualization of the Paradise Papers' geographic distribution. By exploring the data, researchers can gain insights into the offshore finance industry, identify enablers, and answer questions about jurisdiction preferences by country of residence.

As part of the Neo4j Developer Relations team, they aim to integrate Neo4j with other technologies and frameworks, ensuring developers can use Neo4j with their favorite tools. They have seen a surge in popularity of Kubernetes, an open-source system for automating deployment, scaling, and management of containerized applications. A Neo4j causal cluster is composed of servers playing two roles: Core Servers, which safeguard data using the Raft protocol, and Read Replicas, which scale out graph workloads by acting as caches for the data protected by Core Servers. They have created Kubernetes templates in the kubernetes-neo4j repository to help developers get started with deploying Neo4j clusters. Helm is a tool that streamlines installing and managing Kubernetes applications, and they have merged the Neo4j Helm package into the charts incubator, allowing easy deployment of Neo4j clusters using Helm. Developers can install and use Neo4j with their favorite technologies, including Kubernetes, by following the provided steps and feedback channels.

The International Consortium of Investigative Journalists (ICIJ) has released the Paradise Papers, a large dataset of leaked documents from law firm Appleby and trust company Asiaciti. The dataset contains over 13 million documents, emails, and database records that have been analyzed using various tools, including text analysis, full-text search, and graph visualization. The data reveals widespread use of shell companies in tax havens to hide and move large amounts of money without fiscal oversight. Examples include Nike, Apple, the Queen's investment group, Russian investments to politicians like Wilbur Ross, and companies like Facebook and Twitter. The dataset has been integrated into a Neo4j graph database, which allows for interactive visualization and querying. The distribution of entities and relationships shows that there are more officers than addresses in the dataset, with some individuals having almost 90 shell companies registered. The most popular offshore jurisdictions used by people in other countries include Bermuda and the Cayman Islands. The dataset also reveals connections between Wilbur Ross and several offshore entities, including those connected to his second-degree network. Additionally, the Duchy of Lancaster – Queen Elizabeth II's private estate and portfolio – appears in the Paradise Papers dataset, with an investment made in a Cayman entity that had not been previously reported.

The Neo4j community has been actively sharing updates and new developments in the world of graph databases. A significant release this week was the Paradise Papers, a global investigation revealing offshore activities of powerful individuals and multinational companies. The International Consortium of Investigative Journalists (ICIJ) utilized Neo4j and visualisation tool Linkurious to explore and share data from the investigation. Manuel Villa, the first-ever Neo4j Connected Data Fellow at the ICIJ, was featured as a community member for his work on the Paradise Papers. Additionally, online meetups and articles discussed topics such as spatio-semantic comparison of large 3D city models in CityGML using Neo4j, data import into Neo4j using Python, and fraud detection in financial datasets. The Neo4j Knowledge Base also released an article about a common exception when using mismatching versions of APOC and Neo4j. Upcoming events include Graphs in the real world: Philips Lighting and Graphlr tell their story and Data to Value.

The International Consortium of Investigative Journalists (ICIJ) has expanded its Offshore Leaks Database using Neo4j, a graph database, to include 13.4 million documents and 1.4 TB of data from various sources, including trust company Asiaciti and Appleby, a 100-year-old offshore law firm. This new data allows journalists to analyze the connections between wealthy individuals and their financial dealings more easily than ever before. The use of Neo4j enables the ICIJ to manage highly connected data and complex queries, making it an essential tool for investigative journalism. The database's structure, which uses nodes, properties, and relationships, allows for intuitive exploration and visualization of the data, even for non-technical users. Graph databases like Neo4j are well-suited to handle large volumes of interconnected data, making them a crucial component in uncovering the truth behind complex datasets.

Graph visualization is a powerful tool for exploring complex data, leveraging the brain's ability to spot patterns and applying it to virtually any dataset. KeyLines' graph visualization capabilities were used to explore the Neo4j GitHub community, showcasing its ability to cache data locally for faster response times, integrate with GraphQL, and provide automated layouts that reveal distinct communities and structural features of the network. The tool also allows for filtering the graph by time, social network analysis, and decluttering the chart using combos functionality, revealing trends and patterns that would otherwise be hidden.

Neo4j powers personalized promotion and product recommendation engines by connecting masses of complex buyer and product data to gain insight into customer needs and product trends. Traditional relational database technology is insufficient for real-time recommendations due to the complexity and speed required, whereas graph databases like Neo4j quickly query customers' past purchases and capture new interests in real time. This enables personalized promotion and recommendation algorithms that utilize a customer's past and present choices to offer timely suggestions. With a connected inventory, supply chain, and customer data system, retailers can implement dynamic pricing and competing promotions in real-time, making complex rules simple with Neo4j. Walmart and a top 10 US-based retailer have successfully implemented Neo4j in their production systems, improving performance, simplicity, and the customer experience.

This week in Neo4j saw several announcements and developments, including the public alpha release of Cypher for Apache Spark, which enables the execution of Cypher queries on property graphs stored in an Apache Spark cluster. The community also welcomed Rik Van Bruggen as this week's featured member, who has been a part of the Neo4j community for over 5 years and has written several popular resources, including "Learning Neo4j". Additionally, there were various meetups and online sessions, such as user path analysis in Neo4j, asset management, traversal timeouts, and an introduction to APOC. The knowledge base also updated with a new article on setting breakpoints in Cypher statements. Upcoming events include the InterActor meetup and a talk by Dr. Alessandro Negro on using metadata management and analysis in graph technology for data governance.

The text discusses various topics related to GraphQL, a relatively new paradigm for building APIs. The DevRel Engineering team has built integrations around GraphQL and Neo4j, which were showcased at the GraphQL Summit conference. The conference featured talks on using GraphQL for unifying API landscapes, solving problems, and exploring new tools like Apollo Client 2.0, Engine, and Link. Additionally, the team announced the launch of GRANDstack, a combination of technologies that enable full-stack development with synergies between GraphQL, React, Neo4j Database, and other tools. GRANDstack provides a prescriptive approach to building full-stack applications by combining the power of the graph-based data model of Neo4j with the expressiveness of the GraphQL schema and queries. The team also invites developers to participate in a launchpad challenge and offers online training classes to learn more about graph databases and master the world of graph technology.

Cypher for Apache Spark is a new language toolkit that enables big data analysts and data scientists to incorporate graph querying into their workflows, making it easier to leverage graph algorithms. The system allows for the execution of Cypher queries on property graphs stored in an Apache Spark cluster, providing both query execution and a more programmatic API for working with graphs. It supports multiple named graphs, query composition, and dynamic construction of new graphs, enabling complex data processing pipelines across multiple heterogeneous data sources to be constructed incrementally. The system provides an extensible API for integrating additional data sources and is the first open source implementation of Cypher in a distributed memory / big data environment outside of academia.