Data Quality Agent > Data quality cleanse > Sample conversation
  

Sample conversation

Let's take a look at a sample conversation to help you get started on your Data Governance and Catalog data quality cleanse journey in CLAIRE GPT.
As a data steward working in banking and financial services, you want to ensure that the credit card records that you are working on are updated with the correct activation status.
To achieve the goal, perform the following steps:
  1. 1Log in to CLAIRE GPT.
  2. 2Start a conversation to identify what the data quality agent can do with an unclean data set.
    3. Enter the following prompt:
    I have an unclean data set. What can you do with it?
    The following image shows the response displaying a detailed plan on how the agent can help you cleanse your data.The response displays the plan on how the data quality cleanse skill can help you clean your data set.
    CLAIRE GPT lists the steps in a plan to analyze your data set, identify data quality issues in the data, apply data cleanse rules to resolve the data quality issues, generate a profile to give you insights on the structure and quality of data, and preview your data before and after cleansing.
  3. 3Upload the credit_card.xlsx file that contains the credit card records.
    4. The response includes a preview and a profile of the data.
    The following image shows the response displaying a detailed analysis and insights on the data along with a preview of the data in the credit_card.xlsx file. The response displays a detailed analysis and insights on the data in the credit_card.xlsx file along with a card named credit_card Data Preview.
  4. 4Click the credit_card Data Preview card to quickly preview 15 records in the table. You can download the table as a CSV file or copy the table to another file.
    5. The following image shows the canvas with a data preview of the credit card information.The response displays 15 records from the data in credit_card.xlsx.
  5. 5To view the profile of the credit card data, click the Data Profile card.
    6. The following image shows the canvas with profiling statistics of the credit card data.The response displays profiling statistics of the data in credit_card.xlsx.
    You can download the table as a CSV file or copy the table to another file.
  6. 6After the agent profiles your data, the agent suggests cleansing rules for your data.
    7. The following image shows the proposed plan for analyzing data quality issues and suggesting cleansing rules for your data.The response displays the proposed plan of the data quality cleanse skill. The plan includes inferring metadata attributes, identifying critical data elements, computing data quality metrics and so on.
  7. 7If you are happy with the proposed cleansing plan, enter the following prompt:
    8. Yes, proceed with the plan
    The agent identifies critical columns in the credit_card.xlsx file, assesses functional dependencies among critical data elements, updates data with the correct activation status, computes data quality metrics, suggests remediations, and generates cleansed data and a mapplet.
    The following image shows a sample of the cleansed data.The response displays cleansed credit card data. The response also displays the before and after values for the card activation status in the CRD_STAT column.
  8. 8To view the generated cleansing rules report, click the Agent Log of clean data for credit_card card.
    9. The following image shows the canvas with a view of the clean data log.The response displays the cleansed data report, which includes an overview of total, fixable, and non-fixable rules, violations fixed, a data quality dimension summary, and detailed information on the rules extracted by the agent.
  9. 9To view the generated mapplet, click the Card Data Cleaning Mapplet card.
    10. The following image shows the canvas with a view of the data cleansing mapplet that the agent generated.The response displays a cleansing mapplet with rule names, descriptions, and python cleanse codes.
    You can view the mapplet in Data Integration, and reuse the mapplet to cleanse data in similar data sets. You can download the mapplet as a PNG file or copy the mapplet to another file.
Using the data quality cleanse skill of the data quality agent, CLAIRE GPT has analyzed the credit card data, profiled the data, and generated cleansed data and a mapplet.