LLM Query Benchmarks: Cypher vs. SQL with “Spotify Indonesian Artist” dataset

Session Track: App Dev

Session Time: 04:45 am - 05:00 am November 6

Session description

In this session, we will explore and benchmark the capabilities of leading LLMs — ChatGPT 4.1, Gemini 2.5 Flash, and Claude Sonnet 4 — in translating natural language questions into both SQL and Cypher queries. Using a real-world “Spotify Indonesian Artist” dataset, we will compare how each LLM performs across a variety of query types, from simple lookups to complex filters and pattern-based questions. Join us to discover which LLM is best suited for SQL or Cypher query generation and to gain practical insights on leveraging AI for data analytics in music datasets.

Speakers

Anas Ghifari

Undergraduate student, Sepuluh Nopember Institute of Technology

Anas Ghifari is an undergraduate student at Sepuluh Nopember Institute of Technology, majoring in Information Systems. Through his academic journey with Luthfan Aryananda, they have gained hands-on experience in various aspects of IT, with a strong focus on databases and graph technologies. Their background in information systems equips them with a solid understanding of both theoretical concepts and practical applications—skills that are directly relevant to the benchmarking project we’re presenting today.

Luthfan Aryananda

Undergraduate student, Institut Teknologi Sepuluh Nopember

Luthfan Aryananda is an undergraduate student majoring in information systems at Institut Teknologi Sepuluh Nopember. Luthfan is in his 6th semester and currently taking a Graph Knowledge course.