Project: Visualizing the History of Nobel Prize Winners

# Loading in required libraries
import pandas as pd
import seaborn as sns
import numpy as np

# Start coding here!

# 1. load dataset nobel.csv and view first few rows

nobel_df = pd.read_csv("data/nobel.csv")
nobel_df.head()

# Exploring dataset info

nobel_df.info()

# Determining most common gender and birth country

top_country = nobel_df[['birth_country','sex']].value_counts() 
top_country

# 1. what is the most commonly awarded gender and birth country

top_gender = nobel_df['sex'].value_counts().index[0]
top_country = nobel_df['birth_country'].value_counts().idxmax()
print(f'Most commonly awarded gender is "{top_gender}" from "{top_country}".')

# 2. create a new US born winners column based on information in an existing column:

# create US-born winners column
nobel_df['US_winner'] = nobel_df['birth_country'].apply (lambda x: 1 if x== 'United States of America' else 0)
nobel_df['US_winner'].value_counts()

# 2. create decade col should end with 0. divide year values by 10 and wrap this in np.floor() You'll then need to multiply by 10 to get the four-digit year value and set the type as int using .astype().

import numpy as np
nobel_df['decade'] = (np.floor(nobel_df['year'] / 10) * 10).astype(int)
nobel_df.head()

# 2.  finding ratio
ratio_decade_usa = nobel_df.groupby('decade', as_index=False)['US_winner'].mean()

# Identify the decade with the highest ratio of US-born winners
max_decade_usa = ratio_decade_usa[ratio_decade_usa['US_winner'] == ratio_decade_usa['US_winner'].max()]['decade'].values[0]
max_decade_usa

# 2. create a relational line plot
import seaborn as sns
import matplotlib.pyplot as plt

plot = sns.relplot(x='decade', y= 'US_winner',data= ratio_decade_usa, kind='line')
plot.fig.suptitle("US-Born Decade Winners")
plt.show()

# 3. Find the decade and category with the highest proportion of female laureates.

# 3.a   filtering for female winners 1 is true 0 is false
nobel_df['female_winners']=nobel_df['sex'].apply(lambda x: 1 if x =='Female' else 0)
nobel_df['female_winners'].value_counts()

# 3.b   Groupby two columns decade and category 
female_decade_winner = nobel_df.groupby(['decade','category'], as_index=False)['female_winners'].mean()

#   identify highest decade and category of female winners
highest_female_decade_winner= female_decade_winner[female_decade_winner['female_winners']  ==                        female_decade_winner['female_winners'].max()][['decade','category']]
highest_female_decade_winner

# 3.c   create a dictionary

max_female_dict = {highest_female_decade_winner['decade'].values[0]: highest_female_decade_winner['category'].values[0]}
print('Max Female Dictionary:',max_female_dict)

print(f'In Decade "{list(max_female_dict.keys())[0]}" Nobel Prize category "{list(max_female_dict.values())[0]}" combination had the highest proportion of female laureates.')

# 3.d   Create a relational line plot with multiple categories.

sns.relplot(x='decade', y='female_winners',data=female_decade_winner, kind='line', hue= 'category')
plt.title('Proportion of Female Nobel Laureates by Decade and Category')
plt.xlabel('Decade')
plt.ylabel('Proportion of Female Winners')
plt.show()

# 4. Who was the first woman to receive a Nobel Prize, and in what category?

nobel_women = nobel_df[nobel_df['sex'] == 'Female']

# find the minimum value in a column year
min_row = nobel_women[nobel_women['year'] == nobel_women['year'].min()]

first_woman_name = min_row['full_name'].values[0]
first_woman_category = min_row['category'].values[0]
print(f'\n The first woman to win a Nobel Prize was {first_woman_name} in the category {first_woman_category}.')