DataSciencePortfolio
Data Analytics Project: Retail Sales Forecasting
As a data scientist, I utilized predictive modeling and statistical analysis to forecast retail sales, enabling the client to optimize inventory management and increase profitability. Click here to view details of the project.
Machine Learning Project: Fraud Detection System
In this project, I developed a machine learning-based fraud detection system that significantly reduced false positives, saving the client millions in potential losses. Click here to explore the technical aspects and outcomes.
Data Engineering Project: Real-time Data Processing
I spearheaded the design and implementation of a real-time data processing architecture, empowering the client to make informed decisions based on up-to-the-minute insights. Click here to delve into the project's infrastructure and impact. SQL code attached
--Top Customers by Purchase Frequency
SELECT CustomerID, COUNT(SalesID) AS PurchaseCount
FROM sales_data
GROUP BY CustomerID
ORDER BY PurchaseCount DESC;
--Top Customers by Total Spending
SELECT CustomerID, SUM(SaleAmount) AS TotalSpent
FROM sales_data
GROUP BY CustomerID
ORDER BY TotalSpent DESC
LIMIT 10;
-- total revenue per category:
SELECT P.Category, SUM(S.SaleAmount) AS TotalCategoryRevenue
FROM sales_data S
JOIN product_data P ON S.ProductID = P.ProductID
GROUP BY P.Category
ORDER BY TotalCategoryRevenue DESC;
--Top Products Purchased in a Specific Month
SELECT P.ProductID, P.ProductName, SUM(S.SaleAmount / P.Price) AS TotalUnitsSold
FROM sales_data S
JOIN product_data P ON S.ProductID = P.ProductID
WHERE strftime('%Y-%m', S.SaleDate) = '2023-01' -- Specify the month (e.g., January 2023)
GROUP BY P.ProductID, P.ProductName
ORDER BY TotalUnitsSold DESC;
--Monthly Sales Trends for Products
SELECT P.ProductID, P.ProductName, strftime('%Y-%m', S.SaleDate) AS SaleMonth, SUM(S.SaleAmount / P.Price) AS TotalUnitsSold
FROM sales_data S
JOIN product_data P ON S.ProductID = P.ProductID
GROUP BY P.ProductID, P.ProductName, SaleMonth
ORDER BY SaleMonth DESC;
--Best-Selling Products
SELECT P.ProductID, P.ProductName, SUM(S.SaleAmount / P.Price) AS TotalUnitsSold
FROM sales_data S
JOIN product_data P ON S.ProductID = P.ProductID
GROUP BY P.ProductID, P.ProductName
ORDER BY TotalUnitsSold DESC
LIMIT 10;
--Products with the Highest Revenue
SELECT P.ProductID, P.ProductName, SUM(S.SaleAmount) AS TotalRevenue
FROM sales_data S
JOIN product_data P ON S.ProductID = P.ProductID
GROUP BY P.ProductID, P.ProductName
ORDER BY TotalRevenue DESC
LIMIT 10;
--Top Customers by revenue
SELECT S.CustomerID, SUM(S.SaleAmount) AS TotalRevenue
FROM sales_data S
GROUP BY S.CustomerID
ORDER BY TotalRevenue DESC
LIMIT 10;
--sales trend per product category
SELECT P.Category, strftime('%Y-%m', S.SaleDate) AS SaleMonth, SUM(S.SaleAmount) AS TotalRevenue
FROM sales_data S
JOIN product_data P ON S.ProductID = P.ProductID
GROUP BY P.Category, SaleMonth
ORDER BY SaleMonth, P.Category;
--to findSlow-Moving Products
SELECT P.ProductID, P.ProductName, P.StockQuantity, SUM(S.SaleAmount / P.Price) AS TotalUnitsSold
FROM product_data P
LEFT JOIN sales_data S ON P.ProductID = S.ProductID
GROUP BY P.ProductID, P.ProductName, P.StockQuantity
HAVING TotalUnitsSold IS NULL OR TotalUnitsSold < 5 -- Define the threshold for "slow-moving"
ORDER BY TotalUnitsSold ASC;
--top selling months
SELECT strftime('%Y-%m', S.SaleDate) AS SaleMonth, SUM(S.SaleAmount) AS TotalRevenue
FROM sales_data S
GROUP BY SaleMonth
ORDER BY TotalRevenue DESC
LIMIT 10; -- Shows top 10 selling months