SuperSonic

Preview：

Introduce：

SuperSonic is a next-generation business intelligence (BI) platform that integrates Chat BI driven by a Large Language model (LLM) and Headless BI driven by a semantic layer. It ensures that Chat BI has access to the same curated and governed semantic data model as traditional BI. In addition, implementations of both paradigms benefit from integration: Chat BI’s Text2SQL is enhanced by retrieving context from semantic models; Headless BI’s query interface is extended through a natural language API. SuperSonic provides a Chat BI interface that enables users to query data using natural language and visualize the results in appropriate graphs. To achieve this experience, the only thing required is to build a logical semantic model (defining metrics/dimensions/tags and their meanings and relationships) through the Headless BI interface. Meanwhile, SuperSonic is designed to be extensible and composable, allowing custom implementations to be added and configured using Java SPI.

Stakeholders:
SuperSonic’s target audience is data analysts and business users who need a platform that can query and analyze data through natural language. The product suits them because it simplifies the process of data query and analysis, making it easy for non-technical users to access and analyze data, while providing technical users with powerful semantic data model building tools.
Usage Scenario Examples:

• Data analysts use SulierSonic to quickly capture key business metrics through natural language queries.
• Business users use SulierSonic’s visual tools to visualize data analysis results.
• The technical team builds and optimizes semantic data models through SulierSonic’s Headless BI interface.

The features of the tool:

• Built-in Chat BI interface to query data using natural language
• Built-in Headless BI interface for analytics engineers to build semantic data models
• Built-in rule-based semantic parser to improve efficiency in specific scenarios
• Supports automatic input completion, multiple rounds of dialogue, and post-query recommendation
• Supports three levels of data access control: data set level, column level, and row level
• The knowledge base periodically extracts schema information from semantic models and builds dictionaries and indexes to facilitate schema mapping
• The pattern mapper recognizes schema element references in user queries and matches the query text with the knowledge base
• A semantic parser understands user queries and generates semantic query statements, combining rule-based and model-based parsers
• The semantic corrector checks the validity of semantic query statements and corrects and optimizes them if needed
• The semantic translator converts semantic query statements into executable SQL statements

Steps for Use:

• Download the latest prebuilt binaries.
• Run the script ‘assembly/bin/suliersonic-daemon.sh start’ to start the standalone Java service.
• Point your browser at httli://localhost:9080 to start exploring.

Tool’s Tabs: BI,Chat BI