Neo4j - Tokyo User Group Meet Up #50 (Tokyo)

50th Neo4j Users Group Tokyo Meetup at Tokyo, Sep. 11, 2024

Venue
SHIFT Inc. Shinjuku Office
Shinjuku Bunka Quint Building
9th Floor, Cafe Space
3-22-7 Yoyogi, Shibuya-ku, Tokyo 151-0053, Japan

Sessions

GenAI researcher - Building knowledge graphs with LLMs

• In this talk, we will explore the process of building knowledge graphs using Large Language Models (LLMs). We will delve into the techniques and tools necessary to extract, organize, and connect information from diverse data sources. Additionally, we will share practical learnings and real-world examples to illustrate how these methods can be effectively applied to enhance data-driven decision-making and innovation.
  Tomaz Bratanic has extensive experience with graphs, machine learning, and generative AI. He has written an in-depth book about using graph algorithms in practical examples. Nowadays, he focuses on generative AI and LLMs by contributing to popular frameworks like LangChain and LlamaIndex and writing blog posts about LLM-based applications.
• 発表者 (Speaker)
  Tomaz Bratanic
• https://neo4j.com

Powering Clinical Research with Knowledge Graphs

• In healthcare, data is abundant but often siloed. That makes it difficult to extract meaningful insights. Traditional data management systems are not equipped to handle the complexity of healthcare data, which includes clinical records, medical knowledge, omic results, and more. As a consequence, doctors, researchers, and patients are unable to take full advantage of this valuable resource.
  Enter the Knowledge Graphs. They represent and connect diverse data sources. Recently, GPT makes it even easier to construct and query knowledge graphs. In healthcare, we use knowledge graphs to capture the relationships among patients, diseases, treatments, and outcomes. For examples, for a given set of symptoms, doctors can get a complete list of possible causes, ranked by the odds. Knowledge graphs can explain medical concepts and test results in a layman-friendly way so that the patients can understand them more easily.
  In this talk, I will explore the potential of knowledge graphs for healthcare. And we will also discuss the challenges of building and maintaining knowledge graphs, including data quality, interoperability and privacy.
• 発表者 (Speaker)
  Sixing Huang
  https://www.geminidata.com

Knowledge GraphとGraphRAGを用いたリポジトリ全体の概念獲得とその応用
Concept Acquisition and Applications Using Knowledge Graph and GraphRAG for an Entire Repository

• In this presentation, I will introduce the paper "CODEXGRAPH: Bridging Large Language Models and Code Repositories via Code Graph Databases" (https://arxiv.org/pdf/2408.03910) and demonstrate its practical application. Knowledge Graph and GraphRAG are innovative technologies designed to extract concepts from an entire software repository and apply them to various development processes. The Knowledge Graph clarifies relationships between source code and related documents, visually representing dependencies and connections between functions, which helps deepen the understanding of the codebase structure. Meanwhile, GraphRAG (Graph Retrieval-Augmented Generation) leverages this graph structure to enhance the process of retrieving and generating information based on specific contexts. This allows for a comprehensive analysis of the repository’s structure, enabling tasks such as automatic code generation, test case automation, and refactoring suggestions. By combining these technologies, knowledge management in software development is improved, leading to increased developer productivity and better code quality.
• 発表者 (Speaker)
  青山 宙樹 (Hiroki Aoyama)
  Airitech株式会社
  https://www.airitech.co.jp

GQL Standard (ISO/IEC 39075:2024)

• In April 2024, GQL (ISO/IEC 39075:2024) was published. Like SQL for RDB, it is the ISO standard query language for Graph Databases. What are the differences between Cypher and GQL? I will introduce the current status of GQL, which is being gradually incorporated into Neo4j 5.
• 発表者 (Speaker)
  Koji Annoura
  https://annoura.com

【How to join the Neo4j Users Group Meetup】

• Go to this URL
• https://jp-neo4j-usersgroup.connpass.com/event/328200/