data analysis by python

Project Description:

This project involves analyzing and visualizing e-commerce customer behavior using Python. The dataset consists of two CSV files merged based on the 'Gender' column to create a comprehensive dataset for analysis.

Key Steps:

Data Merging:

Combined two datasets using an outer join to retain all relevant information.

Data Wrangling:

Explored the dataset structure, dimensions, and statistical summaries.

Identified missing values, incorrect data types, and duplicates.

Data Cleaning:

Converted date columns to appropriate formats.

Removed duplicates and negative values in numerical columns.

Handled missing values by filling numerical columns with mean/median and categorical columns with 'Unknown'.

Standardized text data by converting it to lowercase.

Exported the cleaned dataset to a new CSV file.

Data Visualization:

Used Seaborn and Matplotlib to generate insights through various visualizations:

Scatter Plots: Analyzed relationships between age, city, and items purchased.

Count Plots: Examined customer distribution by gender.

Histograms: Explored the distribution of sales, price, and items purchased.

Bar Plots: Investigated sales by membership type, device category, and country.

Time Series Analysis: Created a monthly sales trend visualization.

Pie Chart: Showed the proportion of different membership types.

Heatmap: Displayed the correlation between key numerical variables.

Box Plot: Compared sales distribution by product rating.

Skills Demonstrated:

Python for Data Analysis (Pandas, NumPy)

Data Cleaning & Preprocessing

Data Visualization (Matplotlib, Seaborn)

Exploratory Data Analysis (EDA)

Feature Engineering

Time Series Analysis

Outcome:

Successfully cleaned and visualized e-commerce customer data.

Identified insights on sales trends, customer behavior, and spending patterns.

Created an in-depth report that can help businesses make data-driven decisions.