Exploratory Data Analysis (EDA) of Raisin Dataset

The analysis includes data cleaning, descriptive statistics, visualizations, correlation analysis, and interpretation of the results.

Project Summary

1. Data Import and Cleaning:

- Import the Raisin dataset from an Excel file.

- Check and handle missing values.

- Inspect the dataset properties and summarize the data.

2. Descriptive Statistics:

- Calculate and display basic statistics (mean, median, standard deviation, minimum, maximum) for each numerical feature.

- Summarize the findings in a formatted table.

3. Visualizations:

- Generate histograms for each numerical feature to visualize their distributions.

- Create box plots to examine the spread and outliers in the data.

- Plot pairwise scatter plots to explore relationships between features.

- Generate a correlation matrix heatmap for both Pearson and Spearman correlations to assess linear relationships between features.

4. Class-Based Analysis:

- Generate box plots and strip plots for each feature, split by class labels, to analyze differences between classes.

- Create class-specific histograms for each feature to understand the distribution of values within each class.

- Plot pairwise scatter plots with class labels to visualize relationships between features across different classes.

5. Interpretation:

- Interpret the correlation matrix, identifying strong, moderate, and weak linear relationships between features.

- Summarize the key insights from the visualizations and correlation analysis, focusing on patterns that may be relevant for further analysis or modeling.