Launch of 'Data Modeling Training' to learn practical know-how for data infrastructure

Kazaneba Co., Ltd. (Headquarters: Chuo-ku, Tokyo; Representative Partner: Sho Yokoyama) announces the general availability of its 'Data Modeling Training', where IT departments and those involved in data utilization can learn practical know-how regarding layering and table design in data infrastructure.

■ Background and Purpose Data infrastructure is essential for data utilization, DX promotion, and generative AI. When proceeding with an IT system construction project, the importance of organizing the contents of data becomes clear, as issues arise such as "just gathering data in one place is meaningless" and "there are 10 types of similar sales data, causing users and AI agents to make mistakes."

Challenges surface, such as "I don't have the opportunity to systematically learn modeling for data infrastructure, and I want to acquire it consistently from basics to practice," "Table design depends on individuals, and design intentions are not shared within the team," and "I cannot imagine how to translate current issues and requirements into a data model."

Kazaneba receives many such consultations, and recognizes that one of the factors hindering data utilization and DX promotion in Japanese companies is the "lack of practical programs to systematically learn data modeling." Therefore, 'Yuzutaso (@yuzutas0)', translator of the book 'Agile Data Modeling', took the lead in developing this training program.

■ Program Offered This is a practical program to systematically learn 'data modeling', which is indispensable for data analysis and data infrastructure design, over two days.

### Value provided by this course - You can learn everything from basic design to practical application through dimensional modeling and BEAM✲ tables. - Through deliverable reviews and presentations, you will acquire the skills to share and improve design intentions within your team. - Through hands-on exercises using your own company's data, you can produce training results that can be immediately applied in the field.

### [Day 1] Basic Lecture You will learn the basics of dimensional modeling, up to implementation using SQL, Python, and Excel. 1. Overview of dimensional modeling 2. Fact design 3. Dimension design 4. BEAM✲ tables, event matrix 5. Overview of layering 6. Data provisioning formats (wide tables, summary tables) 7. Data preprocessing 8. Data flow design 9. Implementation procedures for SQL (e.g., dbt), Python (e.g., AWS Glue, Google Colab), and Excel

### Practical Exercises You will work on designing and implementing tables over a period of several weeks. By using the data you handle in your work as the subject, you can gain learning that is more directly connected to your practical duties. - Design documents (e.g., online whiteboard tools like Miro) - Implemented tables (e.g., preview screens of Snowflake or BigQuery, dbt configuration files)