Scribd
Upload
7 views

Amazon_Sales_Report_Analysis_Presentation

Uploaded by

Samridhii
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
7 views

Amazon_Sales_Report_Analysis_Presentation

Uploaded by

Samridhii
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 8

Project Objective & Scope

• Objective: Analyze the Amazon Sales Report to


identify trends and patterns in sales data.
• Scope: Insights can help optimize inventory
management and predict future sales trends.
Data Description
• Dataset includes various attributes:
• - Order ID
• - Date
• - Status
• - Fulfilment Method
• - Sales Channel
• - Product Category
• - Size, Amount, and Location details.
Methodology
• 1. Import Libraries and Load Data
• 2. Initial Data Exploration and Visualization
• 3. Data Cleaning and Preprocessing
• 4. Implement Models for Analysis and
Prediction
• 5. Evaluate and Compare Model Performance
Data Preprocessing
• 1. Removed missing and duplicate values.
• 2. Handled outliers using scaling techniques.
• 3. Encoded categorical features for machine
learning compatibility.
Models Used
• Implemented the following models for
analysis and prediction:
• 1. K-Means Clustering
• 2. K-Nearest Neighbors (KNN)
• 3. Random Forest
• 4. Linear Regression
Accuracy Comparison
• Compared models based on their
performance metrics:
• - Random Forest performed best in prediction
accuracy.
• - Linear Regression showed acceptable results
for sales trend predictions.
Inference
• 1. Identified popular product categories and
seasonal sales trends.
• 2. Random Forest provided the most accurate
predictions for future sales.
• 3. Useful insights for improving Amazon’s
inventory and sales strategies.
Future Scope of Improvements
• 1. Use larger datasets for enhanced model
training.
• 2. Include more features such as customer
demographics or marketing campaigns.
• 3. Optimize models for real-time sales
predictions.

You might also like