Glossary

Introduction

This glossary lists commonly used key terms in AskTable along with their definitions to help you better understand and use AskTable.

A

API (Application Programming Interface)
A software interface that allows communication between different software applications.

API-Key (API Key)
A key used for authentication, which needs to be included in the request header when calling the AskTable API.

AI Data Agent (AI Data Agent)
The core positioning of AskTable, an AI system capable of understanding natural language queries and directly interacting with structured data.

AskTable (察言观数)
A leading enterprise-level AI data agent platform in China, enabling companies to directly interact with data using natural language.

ATS (AskTable Test Set)
AskTable test set, used to systematically verify the performance of large models in data analysis scenarios, ensuring the reliability of the entire process from natural language to data insights.

ATST (AskTable Secure Tunnel)
AskTable secure tunnel, allowing AskTable to securely access databases deployed in the internal network, ensuring the security of data transmission.

Authorization (Authorization)
The process of ensuring a user has the right to access specific resources or perform specific operations.

B

Bot (AI Data Assistant)
A program capable of engaging in natural language conversations with users, which can be customized and configured in AskTable to meet business needs.

Business Knowledge (Business Knowledge)
Commonly used abbreviations and rules within an enterprise, critical information that helps AI improve its understanding, including a terminology library, training sets, proprietary preferences, and metadata.

C

ChatBot Widget (Chatbot Component)
An AI data assistant component that can be embedded into a website or application, providing natural language conversation capabilities.

D

Data Retrieval (Data Retrieval)
One of the core modules of AskTable, responsible for executing generated SQL queries and retrieving data, including intelligent learning and dynamic error correction mechanisms.

Data Source (Data Source)
Refers to a database instance, file, or other storage entity containing data. AskTable accesses data through data sources.

Data Visualization (Data Visualization)
Various data presentation options provided by AskTable, including text replies, table displays, charts (bar charts, pie charts, line charts, etc.), and Excel file downloads.

E

Embedding (Embedding)
A technique that converts text into vector representations, used for semantic search and similarity matching.

L

LLM (Large Language Model)
Large language model, the core technological foundation of AskTable, used to understand natural language queries and generate corresponding data query instructions.

M

Meta Brain (Metadata Graph)
A core module of AskTable, storing all metadata in the database as well as necessary keyword data, a comprehensive metadata graph of the database built automatically by AI.

Meta Retrieval (Metadata Retrieval)
One of the core modules of AskTable, responsible for entity recognition and metadata retrieval, extracting key entities through natural language processing and performing vector searches.

Metadata (Metadata)
"Data about data," describing the background information of data, such as structure, format, and meaning, helping AI more accurately interpret and process data.

N

Natural Language Query (Natural Language Query)
A data query request made by users using natural language. AskTable is able to understand and convert it into corresponding data query instructions.

P

Policy (Policy)
Rules defining data access permissions. Policies can allow or deny access to specific data.

Private Deployment (Private Deployment)
A deployment method where AskTable is deployed on the company's own servers, ensuring that data does not leave the internal network.

R

ReAct (Reasoning and Acting)
The ability to reason and act. Bot not only can answer queries but also can perform corresponding actions based on user requests.

Role (Role)
A role is a collection of access permissions. By assigning roles, users can obtain corresponding access permissions.

Row Level Security (Row Level Security)
A data security mechanism that ensures each row of data is filtered according to permission rules before being returned to the user.

S

SaaS (Software as a Service)
Software as a Service, a cloud service model provided by AskTable.

SDK (Software Development Kit)
A set of tools and libraries that allow developers to create applications and use specific APIs.

Secure Tunnel (Secure Tunnel)
An encrypted connection method that allows AskTable to access data sources within the enterprise's internal network.

SQL Generation (SQL Generation)
A core feature of AskTable, automatically converting natural language queries into corresponding SQL query statements.

T

Token (Token)
A string used for authentication, typically used in API calls, including API-Key and temporary Token.

Training Set (Training Set)
A dataset used to train and optimize AI models. In AskTable, it is used to help AI learn how to generate answers better.

Glossary Library (Terminology Library)
A database storing the unique terminology of an enterprise, helping AskTable AI understand the specialized terminology and conventions within the enterprise.

V

Vector Search (Vector Search)
A search technology based on vector similarity, used to efficiently match relevant data in complex data structures.

Conclusion

This glossary provides commonly used key terms and their definitions in AskTable, hoping to help you understand and use AskTable.