The Data Analyst Assistant is an artificial intelligence assistant in Globant Enterprise AI that allows the end user to interact with a dataset in a similar way as a professional data analyst would.
This assistant offers functions such as data summaries, calculations of maximum, minimum and average values, data organization in tables and report generation.
To define an assistant of this type, you must provide the following information:
- Metadata: A description of each dataset, its columns (including data types and possible values), and considerations for the coder and the interpreter (see Define Metadata (JSON)).
- Glossary: A list of terms used within the company and/or the domain of the end user that would be relevant for the LLM to understand the questions.
- Datasets: A set of CSV files that follows the descriptions added as metadata.
When an end user interacts with the Data Analyst Assistant through the Frontend, Globant Enterprise AI performs a series of tasks to identify the purpose of the question, select the appropriate dataset, and generate code to extract the relevant information and present it in a friendly way.
This process is based on the following two components that work together to achieve the full functionality of the Assistant:
- Dispatcher: The Dispatcher is the component in charge of interacting with the end user and managing the selection of data needed to process the query. Its main functions are to:
- Select the appropriate dataset: The Dispatcher analyzes the end user's query and chooses the appropriate dataset according to the question asked.
- Generate SQL code: Once the dataset is selected, the Dispatcher creates a SQL query that extracts the necessary information from the database.
To perform these tasks, the Dispatcher relies on the following elements:
- Metadata: To understand the structure and content of the datasets and generate accurate queries.
- Glossary: To correctly interpret the terminology used in the query and avoid confusion.
- Thinker: The Thinker is the component that receives the retrieved data and processes it to generate an understandable response for the end user. Its main functions are to:
- Process the data: It analyzes the data obtained through the SQL query generated by the Dispatcher.
- Generate the final answer: From the processed data, the Thinker elaborates a clear and useful answer for the end user.
To perform these tasks, the Thinker depends on:
- Metadata: To understand the structure of the selected dataset and process the data appropriately.
- SQL Query: To know which subset of data is available for analysis and work with the correct information.
To define or update a Data Analyst Assistant, you must use the Globant Enterprise AI Backoffice and follow the steps outlined in How to create a Data Analyst Assistant.