Skip to content

Latest commit

 

History

History
74 lines (53 loc) · 2.57 KB

EDA_report.md

File metadata and controls

74 lines (53 loc) · 2.57 KB

Exploratory Data Analysis (EDA) Report

Cook-Me-Up Project: Simplifying Culinary Delights with Data

Overview

This report outlines the exploratory data analysis process performed on the dataset containing information about Indian foods. The dataset includes attributes such as ingredients, diet type, preparation time, cooking time, flavor profile, course, state, and region.

Data Cleaning

The dataset was downloaded from Kaggle and processed to remove inconsistencies:

  • Missing Values: All missing values were filled with Unknown.
  • Duplicates: Duplicate entries were removed.
  • Final Output: Cleaned data was saved as cleaned_data.csv.

Key Statistics

Metric Value
Total Records 255
Unique Dishes 250
Missing Values (Post-Cleaning) 0

EDA Level 1: Initial Exploration

1. Distribution of Key Numerical Features

  • Preparation Time: Median prep time is 20 minutes, with most dishes requiring less than 30 minutes.
  • Cooking Time: Distribution is right-skewed, with outliers requiring over 2 hours.

Visualization: Cooking Time Distribution Cooking Time Distribution

2. Analysis of Categorical Features

Diet Types:

Diet Type Count
Vegetarian 200
Non-Vegetarian 55

Visualization: Diet Type Distribution Diet Type Distribution

Regional Cuisine:

The dataset highlights cuisines from all regions of India, with North India having the highest representation.

EDA Level 2: Advanced Exploration

Using Seaborn and clustering techniques, we identified patterns in the dataset:

  1. Correlation Analysis:

    • Strong correlation between prep and cook times.
    • Negative correlation between number of ingredients and flavor complexity.

  2. Clustering:

    • Recipes were grouped into three clusters based on cooking time and complexity.

  3. Predictive Insights:

    • Cooking time is heavily influenced by the number of ingredients and preparation method.

SQL Insights

  • Diet Distribution: Vegetarian dishes dominate the dataset.
  • Regional Highlights: North Indian cuisine includes the most unique ingredients.
  • Fastest Dishes: Identified top 10 fastest dishes based on total preparation time.

Next Steps:

  • Integrate insights with Looker Studio for interactive visualization.
  • Incorporate predictive modeling to recommend recipes based on user preferences.