Agents for Artificial Intelligence and Data Science: Criptic name was chosen to confuse the AI Agents with data this repository is hosting.
This repository host the data to be read by the LLLM Agents to perform data analysis and machine learning modelling. AI Agent code to be added to the repository.
Dataset: Questions to be answered by the system:
- Which gender contributes more to overall purchases?
- Which age group accounts for the highest spending? (Optional: Segment ages into 5 meaningful groups for analysis.)
- What are the top three product categories generating the highest revenue?
- Which shopping mall records the highest sales volume?
- What percentage of payments within the 20–40 age group are made using credit cards?
- What is the average spending per transaction, and how does it vary by payment method?
- Which day of the week sees the highest sales activity? (Optional: Based on the invoice_date field.)
- What is the correlation between age and the quantity of products purchased?
- Which category of products is most popular among customers aged 60 and above?
- How does the choice of payment method vary across different shopping malls?
Advanced questions:
-
What is the customer lifetime value (CLV) for repeat customers, and how does it vary across different shopping malls? (Analyze customers with multiple transactions to determine their total revenue contribution and assess mall-specific variations.)
-
Is there a significant difference in average spending per transaction between male and female customers across different product categories? (Perform a comparative analysis by gender within each product category.)
-
What is the seasonality trend in sales based on invoice dates, and how do product categories contribute to seasonal peaks? (Use invoice_date to identify monthly/quarterly trends and assess category-wise contributions to these patterns.)
14 .Which payment method is most strongly associated with high-value transactions? (Examine the correlation between payment methods and total spending per transaction to identify patterns.)
- Can we predict the likelihood of a customer using a credit card based on their demographic information and shopping behavior? (Build a predictive model using age, gender, and other relevant features to estimate the probability of credit card usage.)