August 2018 Summaries

This is a technical help forum where users can ask and answer questions related to specific technical topics. The forum allows users to search for existing answers, add tags to their posts for easier discovery, and share details about their projects or ideas. After leaving Slack due to scalability issues and limitations on message length, the community migrated to this dedicated forum, which aims to reduce technical conversations and foster casual chats. To encourage participation, the first five weeks will feature a contest where randomly selected users who introduce themselves can win a $50 USD coupon. Users are invited to provide feedback and ideas for future features in the Feedback category.

Neo4j is a graph database that enables companies to make real-time decisions based on connected data, providing improved competitiveness, reduced project time, cost, and faster product time to market and better performance. Leading companies such as Walmart, eBay, movie recommendation websites, and Fortune 200 hospitality companies are using Neo4j for various use cases including real-time product recommendations, logistics, and personalization at scale. These companies have experienced significant business benefits including accelerated project completion, improved competitiveness, reduced costs, and increased success rates on projects. By leveraging the connected data capabilities of Neo4j, these organizations can gain valuable insights into customer behavior and preferences, enabling them to make data-driven decisions and drive business growth.

This week in Neo4j has seen the launch of a brand new community site and forum, which will replace technical discussions on Neo4j Users Slack. The new platform allows users to ask and answer questions around the Neo4j Graph Platform, Cypher, Drivers, Integrations, and more. Yisroel Yakovson, CEO at MatchLynx, has been featured as a community member for his work using GraphQL with Neo4j and publishing articles about "The Full Graph Stack". Additionally, Max De Marzi added new posts to his Build a Dating site series, implementing functionality to allow users to High Five and Low Five posts. Other highlights include the release of angular-neo4j, a module for using Neo4j Bolt driver from an Angular application, and a blog post by Rik Van Bruggen on analyzing data from ESCO into Neo4j.

This year's GraphConnect conference features a diverse lineup of must-see presenters, including experts from top companies like Lockheed Martin, Adobe, Nordstrom, and CA Technologies. Ann Grubbs will discuss Product DNA: Master Data Graph Enabling the Digital Transformation, while Brandy Freitas will share insights on Enhancing Machine Learning with Graph Metrics. David Fox will showcase Harnessing the Power of Neo4j for Overhauling Legacy Systems, highlighting the benefits of graph technology in legacy system modernization. Dr. Alexander Jarasch will present Graphs to Fight Diabetes, leveraging knowledge graphs and machine learning techniques to combat diabetes research challenges. Seth Dimick will discuss Graph Recommendations at Nordstrom, applying graph algorithms to enhance personalized fashion retail experiences. Dr. Tatiana Romina Hartinger will engage in a conversation about A Conversation with Graphs, exploring the intersection of graph theory and combinatorics. Gary Stewart and Will Bleker will present Being In Control and Staying Agile with Graph Requires Shifting Left, emphasizing the importance of shifting left in graph adoption. Amy Hodler will discuss 6 Ways Graph Technology Is Changing Artificial Intelligence and Machine Learning, highlighting the impact of graph analytics on AI and machine learning applications. Dr. Alessandro Negro will deliver two sessions on Graph-Based Natural Language Understanding, exploring the capabilities of graph technology in NLP. Pat Patterson will present Ingesting Data into Neo4j for Master Data Management, demonstrating the value of Neo4j in master data management. Finally, Dr. Peng Sun will discuss Accelerating Digital Transformation in CA Technologies with Neo4j, highlighting the benefits of graph technology in accelerating digital transformation.

The GraphConnect 2018 conference in New York City is expected to feature a lineup of top graph experts sharing their experiences on how graph database technology impacted their businesses. The event will showcase the capabilities of Neo4j, a leading graph database platform, and its applications in various industries such as healthcare, finance, and retail. Key speakers at the conference include Ann Grubbs from Lockheed Martin Space Systems, Brandy Freitas from Pitney Bowes, David Fox from Adobe, Dr. Alexander Jarasch from The German Center for Diabetes Research (DZD), Seth Dimick from Nordstrom, Dr. Tatiana Romina Hartinger from Cognitiva, Gary Stewart and Will Bleker from ING, Amy Hodler from Neo4j, Dr. Alessandro Negro from GraphAware, Pat Patterson from StreamSets, and Dr. Peng Sun from CA Technologies. These experts will be presenting on various topics such as machine learning, natural language processing, graph analytics, and data ingestion, among others. Registration is now open for the event, which will take place on September 20-21.

The eBay App for Google Assistant is a chatbot-powered conversational commerce system that uses knowledge graphs to support natural language understanding. The system was built using Neo4j, a graph database, which powers machine learning on the knowledge graph and enables scalable deployment of the system. To scale with traffic increases, the team used Dockerized systems on top of Google Cloud Platform and Kubernetes, leveraging StatefulSet for persistent data storage. The system uses probabilistic inference to determine the next question to ask in a conversation, allowing it to handle complex queries and provide personalized recommendations. With over 160 million active buyers and $11 billion in mobile sales, the eBay App for Google Assistant is a significant example of conversational commerce and graph-powered AI.

The text discusses the differences between imperative and declarative query languages used in database development, particularly in graph databases. Imperative query languages provide detailed control over the execution of tasks but can be limiting, user-unfriendly, and prone to human error. Declarative query languages, on the other hand, let users express what data to retrieve without specifying how it's done, offering speed, productivity, and flexibility. The choice between these two paradigms depends on the specific use case, with imperative languages suitable for projects requiring finer accuracy and control, and declarative languages better suited for applications where speed and productivity matter more.

Graph databases are a natural fit for delivering real-time recommendations in online commerce, as they can easily capture and analyze large amounts of buyer and product data to gain insight into customer needs and product trends. This technology is quickly leaving traditional relational databases behind and is now available off-the-shelf, with Neo4j being the most popular graph database. It supports many named, directed relationships between entities, giving a rich semantic context for the data, and queries are super-fast since there is no JOIN penalty. Graph databases are especially suited to formulating recommendations as they can understand the customer's past purchases, quickly query this data, and match the customer to the people who are the closest match to them both in their social network and in buying patterns. With Neo4j, businesses like eBay have seen significant profit and productivity improvements over traditional relational systems.

This week in Neo4j has seen the release of Graphileon Personal Edition, a new Getting Started with Neo4j YouTube series, and various community projects such as building a dating site's user timeline and importing data from Zendesk into Neo4j. The team is also seeking help to find a new maintainer for the neo4jrb driver library, which has been instrumental in the Ruby community. Additionally, GraphStory has expanded their offering by supporting Enterprise Neo4j in 15 Google Cloud Platform regions. The community has also shared various blog posts and videos on using Neo4j for different use cases, including asset tracking, querying with Neo4j OGM, and loading graph data into Object Graph Mappers.

GraphConnect 2018 is an upcoming event where Emil will be announcing a new release. The event also features various Neo4j training classes, including half-day courses that allow attendees to take two training sessions in one day. A six-question quiz has been created to help attendees choose the most suitable training option for themselves. The event takes place on September 20-21, 2018, in Times Square, New York City, and can be registered for through the GraphConnect website. Attendees can also reach out to the Neo4j team with any questions or concerns about the training classes or event registration.

The Neo4j Community Maven program is designed to support individuals who are enthusiastic about sharing knowledge and ideas within their local communities. These community-driven supernodes, known as maven's, play a crucial role in connecting others and driving engagement through their social skills and ability to communicate. The program aims to encourage and support these individuals in their growth as informational resources, with the goal of fostering a graph-thinking epidemic. By joining this program, participants can experience personal and professional growth, develop strong connections with like-minded people, and become recognized as thought leaders within their community.

A database query language is essential for creating, manipulating and querying data in a graph database, as it allows users to model their data and ask questions about it. The choice of a database query language matters because it can impact the efficiency and effectiveness of data analysis. Graph databases use a variety of query languages, including Cypher, which is designed to be easy-to-learn for developers and business stakeholders alike. Cypher's syntax and semantics are matched to the nature of graph-based problems, making it an ideal choice for tackling connected data challenges. The language also offers various clauses, such as MATCH, RETURN, WHERE, CREATE, and DELETE, that can be used to model and query data. Other graph technologies, including SPARQL and Gremlin, offer alternative querying options, while a vendor-neutral standard, GQL, is being developed to provide a single, unified approach. Understanding the implications of choosing a database query language can help users optimize their data modeling and analysis processes, leading to improved efficiency and effectiveness in resolving data challenges.

The answer to better visibility into deep connections in risk data lies in understanding risk data lineage, which can help financial houses limit their exposure. A connected data foundation not only streamlines financial risk reporting but also supports innovative uses of enterprise data, including 360-degree customer visibility, fraud detection, and proactive credit assessment. The Finance Industry Business Ontology (FIBO) is a standard for investment instruments, business entities, market data, legal obligations, and corporate actions affecting global financial markets, which can be represented in graph databases to provide clarity and consistency across multiple data repositories and silos. FIBO's flexibility makes it an excellent standard for the ever-changing financial industry, allowing it to adapt as financial markets, technology, and regulations evolve. Neo4j is FIBO-ready, enabling organizations to create an enterprise canonical data model that uses the same infrastructure to store risk data lineage and governance metadata. Graph technology also plays a crucial role in addressing compliance with BCBS 239 risk-reporting mandates, allowing forward-looking banks to build federated databases that add centralized metadata control over operational data in existing silos and sources. By using Neo4j to address these requirements, organizations can analyze financial risk faster and with more accuracy, simplify communications and speed development, and build risk reporting applications that bring them into regulatory compliance.

This week in Neo4j saw releases of the APOC and Neo4j JDBC Driver, as well as a paper on deriving socially useful information from public blockchains. The Neo4j community featured Michael Graham, a full stack developer who has been working with GraphQL and Cypher query execution layer for Neo4j and JavaScript GraphQL implementations. There were also webinars on how graphs revolutionize identity and access management, and a refresh of the Neo4j ETL guide. In terms of releases, Tom Sawyer Perspectives version 8.2 was released with model-based engineering enhancements, while Neo4j JDBC Driver version 3.4.0 added support for spatial and temporal data types. The summer release of APOC brought new features such as reverse geocoding and base 64 URL encoding. Additionally, a paper on blockchain analytics and two papers on biological knowledge networks were published, showcasing the versatility of Neo4j in various domains.

GraphConnect 2018 offers attendees a unique opportunity to network and build relationships with fellow graph technology professionals, as well as learn from the makers of Neo4j, who are available for in-person access. The event features a wide range of trainings, workshops, and sessions on various graph-related topics, including AI and machine learning, biotech and healthcare, and digital transformation. Additionally, attendees can participate in the GraphHack hackathon, where they can build new applications using Neo4j and other technologies, and enjoy the post-conference disConnect party, which offers free drinks and snacks. The event also features an amazing speaker lineup, including top innovators in the graph technology ecosystem, and provides a chance to take away valuable ideas and insights to apply to their job or make a difference at their company.

Graph databases may not be the best fit for projects that involve disconnected data with no relationships between transactions, or where optimizing for writing and storing data without querying it is a priority. They also struggle when core data objects or data models are fixed and tabular, or when large amounts of text or BLOBs need to be stored as properties. Additionally, graph databases may not excel in scenarios that involve bulk data scans or queries starting from an unknown data point. However, they can provide significant value in business-critical use cases where users want to understand relationships in their data, create well-rounded customer profiles, and perform complex analysis on connected data.

Building a connected data foundation is essential for innovative uses of enterprise data, including 360-degree visibility of customers and detecting fraud. However, integrating information into a single logical data model can be difficult due to the structure and location of much of the data making it challenging. Many banks have accepted their data will remain in silos and are now embracing a federated approach that leaves the data dispersed while maintaining control using centralized metadata. This approach makes it easier to relate entity identities, maintain data consistency, and describe end-to-end data lineage. To tackle the complexity of risk management and BCBS 239 requirements, graph database technology is necessary, and Neo4j is considered the most popular and successful native graph database that eliminates data consistency problems and provides dependable query performance. By choosing Neo4j for risk reporting compliance, banks get a lot more than just a leading graph database, including global financial terminology standards and professional services to guarantee success.

This week in Neo4j covers various topics including exploring large Knowledge Graphs with MetaExp, analyzing YouTube videos using NLP techniques, and multiple linear regression on graphs. The community is also featured with Michael McKenzie, a new member of the Neo4j community who took over the D.C.-area meetup group. Tips for using Neo4j SDN and OGM are shared by Michael Simons, while resources are provided for beginners to get started with Neo4j and graph databases. Other topics include NLP analysis of YouTube videos, multiple linear regression, and a deep dive into Cypher's OPTIONAL MATCH clause. The next week's events and a tweet of the week are also highlighted.

Cerved, an Italian data-driven company, developed Graph4you, a tool that utilizes Neo4j's graph database to help businesses and institutions make informed decisions. The technology is based on collecting various data sources, including geo-localized information, real estate data, and business ownerships, resulting in over 45 million nodes and 100 million relationships in the Neo4j cluster. Graph4you allows users to query and navigate through the Italian business network to discover connections between entities or get corporate linkages of companies and individuals. The platform also provides a web application with graph visualization capabilities using Ogma.js, REST API integration for automation, and customization options such as weighted shortest-path algorithms developed by LARUS. With Neo4j's native graph database solution and built-in algorithms, Graph4you helps customers accelerate and improve their decision-making process for procurement, fraud detection, and business intelligence problems.

I took over a local Neo4j meetup group in Washington, D.C. after Karin Wolok recommended it to me. I had no idea what I was doing but was excited and full of energy. The first meetup was a success despite some last-minute issues, including the absence of the scheduled speaker due to work commitments. We served pizza and drinks, introduced the group, presented an introduction to graph technology and Neo4j, and had a short workshop on Cypher queries. Despite initial concerns about being rigid and organized, we allowed for dialogue and questions during the presentation, which worked well. The meetup was enjoyable, and I learned that it's essential to have fun, don't put too much pressure on making everything perfect, use available resources, let others help you, and run meetups that interest you.

Airbnb has developed an internal data tool called Dataportal to help employees with data exploration and discovery, addressing the challenge of managing a large amount of data points. The Dataportal is built on top of Neo4j, a graph database that captures relationships between people and data resources, helping guide users to the relevant data they need. It integrates well with Python and Elasticsearch, existing technologies used by Airbnb. The tool provides a curated view of the data ecosystem, allowing users to search for data resources, explore in-depth resource details, and discover metadata. It also helps build trust in data by providing transparency about data and showcasing former employees' contributions. Despite its benefits, building the Dataportal presented challenges, including managing complex dependencies and designing a user-friendly interface for diverse data literacy levels. Future directions include network analysis to identify obsolete nodes, active curation of data resources, and delivering relevant updates through alerts and recommendations.