[ Practiced different approaches and systems in project02.ipynb ]

Statistical Analysis on a Global Food Dataset

This project performs a comprehensive statistical analysis on a dataset of various food items from around the world. The primary goal is to uncover patterns, distributions, and relationships within the data through descriptive statistics, data visualization, and hypothesis testing.

Dataset

The analysis is based on the cleaned_food_data.csv file. This dataset contains information about various food items, including the following key columns used in this notebook:

• Country/Region

• Calories

• Preparation Time

• Category

• Popularity Rank

Analysis Performed

The Jupyter Notebook (Satistical_Analysis_on_Dataset.ipynb) conducts the following analyses:

1. Descriptive Statistics

• Frequency Count: Calculation of the number of unique food items from each country/region to understand the geographical distribution of the data.

• Grouped Analysis: Calculation of the average calories and preparation time for food items, grouped by their country of origin.

2. Cross-Tabulation

• A contingency table was created to analyze the distribution of different food categories (e.g., Dessert, Main Course, Snack) across various countries.

3. Data Visualization

Several plots were generated to visually represent the findings:

• Bar Chart: Food item counts by country.

• Bar Charts: Average calories and average preparation time by country.

• Stacked Bar Chart: Distribution of food categories within each country.

• Scatter Plot: Relationship between the number of calories and the popularity rank of food items, including a regression line to visualize the trend.

• Histogram: Distribution of calories across all food items to observe its central tendency and spread.

4. Hypothesis Testing

• A one-sample Z-test was conducted for each country's food items to test the null hypothesis (H_0) that the mean preparation time is 30 minutes.

  • Null Hypothesis (H_0): mu=30 minutes.

  • Alternative Hypothesis (H_a): muneq30 minutes.

  • A significance level (alpha) of 0.05 was used.

Key Findings

• Geographical Distribution: India, Japan, and the United States are the most represented countries in the dataset.

• Nutritional Insights: Foods from Singapore and the United Kingdom have the highest average calories.

• Preparation Time: Foods from Morocco and Korea have notably long average preparation times compared to other countries.

• Hypothesis Test Result: The Z-test showed that for most countries, there is not enough statistical evidence to reject the claim that the average preparation time is 30 minutes.

  • A significant exception is Korea, where the null hypothesis was rejected. The average preparation time for Korean foods in the dataset (325 minutes) is statistically different from 30 minutes.

Technologies Used

• Python

• Pandas: For data manipulation and analysis.

• NumPy: For numerical operations.

• Matplotlib & Seaborn: For data visualization.

Open the Satistical_Analysis_on_Dataset.ipynb file and run the cells sequentially to see the analysis and visualizations.