Oshinit

Get Code from GitHub

Analysis of Restaurants in Portsmouth

Objective:
To conduct a comprehensive analysis of Portsmouth's dining scene to:

  1. Identify key dining areas and their geographic distribution.
  2. Understand customer preferences regarding service types, price ranges, and cuisines.
  3. Assess the competitive landscape and identify potential market gaps.
  4. Analyze the correlation between geographic location and restaurant ratings.
  5. Visualize the spatial distribution of dining establishments.
  6. Provide actionable insights to improve offerings and customer satisfaction.

This project involves Data cleansing, Exploration and visualization in Python.

1. Data Collection, Preparation and Processing

Dataset was obtained by web scrapping.

In [1]:
# Import package
          import pandas as pd
          import seaborn as sns
          import matplotlib.pyplot as plt
          import re
          import folium

Import the datasets

In [2]:
# Import Data
          df = pd.read_csv('./data/portsmouth_restaurants_google_maps_data.csv')
          df_location = pd.read_csv('./data/portsmouth_restaurants_location.csv')
          
          
          # df_location.info()
          # df_location.describe()

Preview datasets

In [3]:
print('Preview Restaurant Data')
          df.head()
          
Preview Restaurant Data
Out[3]:
name rating ratingCount priceRanges category address postcode openingHours phone web serviceInfo
0 Cocked Hat 4.0 2209.0 £10–20 Pub Privett Rd, Gosport PO12 3TR PO12 3TR NaN 023 9258 3422 greeneking.co.uk {"Accessibility":"Wheelchair-accessible car pa...
1 Tap & Tandoor 4.7 579.0 NaN Gastropub Gunwharf Rd, Portsmouth PO1 3TJ PO1 3TJ NaN 023 9381 3150 tapandtandoor.co.uk {"Accessibility":"Wheelchair-accessible car pa...
2 Bella Calabria 4.4 919.0 £10–20 Italian restaurant 45 London Rd, North End, Portsmouth PO2 0BH PO2 0BH {"Sunday":"12 - 10pm","Monday":"12 - 10pm","Tu... 023 9269 4371 bellacalabriaportsmouth.com {"Accessibility":"Wheelchair-accessible entran...
3 Nice Tin Tin Chinese Restaurant 4.4 456.0 NaN Chinese restaurant 28 The Hard, Portsmouth PO1 3DT PO1 3DT {"Sunday":"12 - 10:30pm","Monday":"12 - 10:30p... 023 9217 4590 nicetintin.co.uk {"Accessibility":"Wheelchair-accessible car pa...
4 El Greco 4.7 288.0 £20–30 Greek restaurant 8 Bellevue Terrace, Southsea, Portsmouth, Sou... PO5 3AT {"Sunday":"1:30 - 10pm","Monday":"Closed","Tue... 07594 159161 elgrecoportsmouth.co.uk {"Service options":"Takeaway, Dine-in, Deliver...
In [4]:
df.info()
          df.describe()
          
<class 'pandas.core.frame.DataFrame'>
          RangeIndex: 117 entries, 0 to 116
          Data columns (total 11 columns):
           #   Column        Non-Null Count  Dtype  
          ---  ------        --------------  -----  
           0   name          117 non-null    object 
           1   rating        117 non-null    float64
           2   ratingCount   116 non-null    float64
           3   priceRanges   109 non-null    object 
           4   category      113 non-null    object 
           5   address       117 non-null    object 
           6   postcode      117 non-null    object 
           7   openingHours  95 non-null     object 
           8   phone         114 non-null    object 
           9   web           105 non-null    object 
           10  serviceInfo   117 non-null    object 
          dtypes: float64(2), object(9)
          memory usage: 10.2+ KB
Out[4]:
rating ratingCount
count 117.000000 116.000000
mean 4.410256 764.698276
std 0.323873 729.954248
min 2.800000 2.000000
25% 4.300000 271.500000
50% 4.400000 548.500000
75% 4.600000 1047.750000
max 5.000000 4704.000000
In [5]:
print('Preview Restaurant Location Data')
          df_location.head()
          
Preview Restaurant Location Data
Out[5]:
postcode admin_county admin_district nuts pfa latitude longitude country
0 PO12 3TR Hampshire Gosport Gosport Hampshire 50.794883 -1.162522 England
1 PO1 3TJ NaN Portsmouth Portsmouth Hampshire 50.794328 -1.106968 England
2 PO2 0BH NaN Portsmouth Portsmouth Hampshire 50.814827 -1.079977 England
3 PO1 3DT NaN Portsmouth Portsmouth Hampshire 50.797996 -1.105920 England
4 PO5 3AT NaN Portsmouth Portsmouth Hampshire 50.789083 -1.096187 England
In [6]:
df_location.info()
          df_location.describe()
          
<class 'pandas.core.frame.DataFrame'>
          RangeIndex: 117 entries, 0 to 116
          Data columns (total 8 columns):
           #   Column          Non-Null Count  Dtype  
          ---  ------          --------------  -----  
           0   postcode        117 non-null    object 
           1   admin_county    1 non-null      object 
           2   admin_district  116 non-null    object 
           3   nuts            116 non-null    object 
           4   pfa             117 non-null    object 
           5   latitude        117 non-null    object 
           6   longitude       116 non-null    float64
           7   country         116 non-null    object 
          dtypes: float64(1), object(7)
          memory usage: 7.4+ KB
Out[6]:
longitude
count 116.000000
mean -1.092655
std 0.013125
min -1.162522
25% -1.105185
50% -1.091528
75% -1.082569
max -1.064810

Numbers of rows and columns in Restuarant Dataset

In [7]:
print(f"The number of rows in dataset is {df.shape[0]}")
          print(f"The number of columns in dataset is {df.shape[1]}")
          
The number of rows in dataset is 117
          The number of columns in dataset is 11

Check for duplicates rows and remove.

In [8]:
print(f"The number of duplicate restaurants is {df.duplicated(subset=['name', 'postcode']).sum()}")
          df.drop_duplicates(subset=['name', 'postcode'], inplace=True)
          print(f"The number of duplicate restaurants is {df.duplicated(subset=['name', 'postcode']).sum()}")
          
          print(f"The number of duplicate restaurants postcode is {df_location.duplicated(subset=['postcode']).sum()}")
          df_location.drop_duplicates(subset=['postcode'], inplace=True)
          print(f"The number of duplicate restaurants  postcode is {df_location.duplicated(subset=['postcode']).sum()}")
          
The number of duplicate restaurants is 1
          The number of duplicate restaurants is 0
          The number of duplicate restaurants postcode is 32
          The number of duplicate restaurants  postcode is 0

Check for missing values in restaurants dataset

In [9]:
print(f"Missing values in each restaurant column {df.isnull().sum()}")
          print(" ")
          print(f"Missing values in each restaurant location column {df_location.isnull().sum()}")
          
Missing values in each restaurant column name             0
          rating           0
          ratingCount      1
          priceRanges      8
          category         4
          address          0
          postcode         0
          openingHours    22
          phone            3
          web             12
          serviceInfo      0
          dtype: int64
           
          Missing values in each restaurant location column postcode           0
          admin_county      84
          admin_district     1
          nuts               1
          pfa                0
          latitude           0
          longitude          1
          country            1
          dtype: int64

Merge datasets

In [10]:
restaurants_df = df.merge(df_location, how='left', on='postcode' ) 
          print(f"The number of rows in dataset is {restaurants_df.shape[0]}")
          print(f"The number of columns in dataset is {restaurants_df.shape[1]}")
          restaurants_df.head()
          
The number of rows in dataset is 116
          The number of columns in dataset is 18
Out[10]:
name rating ratingCount priceRanges category address postcode openingHours phone web serviceInfo admin_county admin_district nuts pfa latitude longitude country
0 Cocked Hat 4.0 2209.0 £10–20 Pub Privett Rd, Gosport PO12 3TR PO12 3TR NaN 023 9258 3422 greeneking.co.uk {"Accessibility":"Wheelchair-accessible car pa... Hampshire Gosport Gosport Hampshire 50.794883 -1.162522 England
1 Tap & Tandoor 4.7 579.0 NaN Gastropub Gunwharf Rd, Portsmouth PO1 3TJ PO1 3TJ NaN 023 9381 3150 tapandtandoor.co.uk {"Accessibility":"Wheelchair-accessible car pa... NaN Portsmouth Portsmouth Hampshire 50.794328 -1.106968 England
2 Bella Calabria 4.4 919.0 £10–20 Italian restaurant 45 London Rd, North End, Portsmouth PO2 0BH PO2 0BH {"Sunday":"12 - 10pm","Monday":"12 - 10pm","Tu... 023 9269 4371 bellacalabriaportsmouth.com {"Accessibility":"Wheelchair-accessible entran... NaN Portsmouth Portsmouth Hampshire 50.814827 -1.079977 England
3 Nice Tin Tin Chinese Restaurant 4.4 456.0 NaN Chinese restaurant 28 The Hard, Portsmouth PO1 3DT PO1 3DT {"Sunday":"12 - 10:30pm","Monday":"12 - 10:30p... 023 9217 4590 nicetintin.co.uk {"Accessibility":"Wheelchair-accessible car pa... NaN Portsmouth Portsmouth Hampshire 50.797996 -1.105920 England
4 El Greco 4.7 288.0 £20–30 Greek restaurant 8 Bellevue Terrace, Southsea, Portsmouth, Sou... PO5 3AT {"Sunday":"1:30 - 10pm","Monday":"Closed","Tue... 07594 159161 elgrecoportsmouth.co.uk {"Service options":"Takeaway, Dine-in, Deliver... NaN Portsmouth Portsmouth Hampshire 50.789083 -1.096187 England

Check for duplicates

In [11]:
restaurants_df.duplicated(subset=['name', 'postcode']).sum()
Out[11]:
np.int64(0)

Check for missing values

In [12]:
for col in restaurants_df.columns[restaurants_df.isnull().any()]:
              print(f'Column {col} has {restaurants_df[col].isnull().sum()} missing value')
          
Column ratingCount has 1 missing value
          Column priceRanges has 8 missing value
          Column category has 4 missing value
          Column openingHours has 22 missing value
          Column phone has 3 missing value
          Column web has 12 missing value
          Column admin_county has 115 missing value
          Column admin_district has 1 missing value
          Column nuts has 1 missing value
          Column longitude has 1 missing value
          Column country has 1 missing value

Handling missing values

In [13]:
# Handling of missing values will be done per analysis basis.
          restaurants_df.fillna({'priceRanges': '0-0', 'phone': '000', 'web': 'no website', 'category': 'restaurant'}, inplace=True)
          restaurants_df.drop(columns=['admin_county'], axis=1, inplace=True)

Clean column values

In [14]:
# Clean price range
          def get_price_range(price_range):
            if '£' in price_range:
              return price_range.split('£')[1]
            else:
              return price_range
          
          restaurants_df['priceRanges'] = restaurants_df['priceRanges'].apply(get_price_range)
          
          def remove_non_ascii(address):
              return re.sub(r'[^\x00-\x7F]+', '', address)
          
          # Remove non ascii characters:
          restaurants_df['address'] = restaurants_df['address'].apply(remove_non_ascii)
          restaurants_df['phone'] = restaurants_df['phone'].apply(remove_non_ascii)
          restaurants_df['web'] = restaurants_df['web'].apply(remove_non_ascii)
          
          restaurants_df.head(5)
Out[14]:
name rating ratingCount priceRanges category address postcode openingHours phone web serviceInfo admin_district nuts pfa latitude longitude country
0 Cocked Hat 4.0 2209.0 10–20 Pub Privett Rd, Gosport PO12 3TR PO12 3TR NaN 023 9258 3422 greeneking.co.uk {"Accessibility":"Wheelchair-accessible car pa... Gosport Gosport Hampshire 50.794883 -1.162522 England
1 Tap & Tandoor 4.7 579.0 0-0 Gastropub Gunwharf Rd, Portsmouth PO1 3TJ PO1 3TJ NaN 023 9381 3150 tapandtandoor.co.uk {"Accessibility":"Wheelchair-accessible car pa... Portsmouth Portsmouth Hampshire 50.794328 -1.106968 England
2 Bella Calabria 4.4 919.0 10–20 Italian restaurant 45 London Rd, North End, Portsmouth PO2 0BH PO2 0BH {"Sunday":"12 - 10pm","Monday":"12 - 10pm","Tu... 023 9269 4371 bellacalabriaportsmouth.com {"Accessibility":"Wheelchair-accessible entran... Portsmouth Portsmouth Hampshire 50.814827 -1.079977 England
3 Nice Tin Tin Chinese Restaurant 4.4 456.0 0-0 Chinese restaurant 28 The Hard, Portsmouth PO1 3DT PO1 3DT {"Sunday":"12 - 10:30pm","Monday":"12 - 10:30p... 023 9217 4590 nicetintin.co.uk {"Accessibility":"Wheelchair-accessible car pa... Portsmouth Portsmouth Hampshire 50.797996 -1.105920 England
4 El Greco 4.7 288.0 20–30 Greek restaurant 8 Bellevue Terrace, Southsea, Portsmouth, Sout... PO5 3AT {"Sunday":"1:30 - 10pm","Monday":"Closed","Tue... 07594 159161 elgrecoportsmouth.co.uk {"Service options":"Takeaway, Dine-in, Deliver... Portsmouth Portsmouth Hampshire 50.789083 -1.096187 England

2. Data Analysis

2.1 Distribution of Postcode in Portsmouth

In [15]:
postcode_df = restaurants_df.copy()
          
          def get_first_line_of_postcode(code):
              if code is not None:
                  return code.split(' ')[0]
              else:
                  return code
          
          postcode_df['outward_code'] = postcode_df['postcode'].apply(get_first_line_of_postcode)
          print('Distribution of Postcode')
          print(postcode_df['outward_code'].value_counts())
          
Distribution of Postcode
          outward_code
          PO1     51
          PO5     39
          PO2     10
          PO6      7
          PO4      6
          PO3      2
          PO12     1
          Name: count, dtype: int64
In [16]:
plt.figure(figsize=(15, 5))
          postcode_restaurant_count = postcode_df['outward_code'].value_counts()
          
          plt.subplot(1, 2, 1)
          postcode_restaurant_count.plot(kind='pie', autopct='%1.1f%%') 
          plt.title('Distribution of Restaurants by Outward Code')
          
          plt.subplot(1, 2, 2)
          postcode_restaurant_count.plot(kind='bar')
          plt.xlabel('Postcode')
          plt.ylabel('Restaurant Count')
          plt.title('Distribution of Restaurants across Postcodes')
          
          print('P01 has the highest number of restaurants at 51.')
          print('P012 has the lowest number of restaurants at 1.')
          
P01 has the highest number of restaurants at 51.
          P012 has the lowest number of restaurants at 1.
No description has been provided for this image
In [17]:
print(f'Data type of restaurant type is {postcode_df['category'].dtype}')
          cuisine_distribution = postcode_df['category'].value_counts()
          print('Distribution of cuisine')
          print(cuisine_distribution)
          
Data type of restaurant type is object
          Distribution of cuisine
          category
          Restaurant                   30
          Pub                           7
          Italian restaurant            6
          Chinese restaurant            5
          restaurant                    4
          Hamburger restaurant          4
          Turkish restaurant            4
          Cafe                          3
          Vietnamese restaurant         3
          Gastropub                     2
          Thai restaurant               2
          Japanese restaurant           2
          Portuguese restaurant         2
          Fish and chips takeaway       2
          Kebab Shop                    2
          Cocktail bar                  2
          Pizza restaurant              2
          Greek restaurant              2
          Family restaurant             1
          Brazilian restaurant          1
          Tex-Mex restaurant            1
          Tapas bar                     1
          Bistro                        1
          Vegan restaurant              1
          African restaurant            1
          Modern British restaurant     1
          Pizza Takeaway                1
          Filipino restaurant           1
          Belgian restaurant            1
          Caribbean restaurant          1
          Takeaway                      1
          Malaysian restaurant          1
          British restaurant            1
          French restaurant             1
          Bar & grill                   1
          Tapas restaurant              1
          Indian restaurant             1
          Dim Sum restaurant            1
          Asian restaurant              1
          Lebanese restaurant           1
          Buffet restaurant             1
          Chinese Takeaway              1
          Steak house                   1
          Fast food restaurant          1
          Sushi restaurant              1
          Mexican restaurant            1
          Delivery service              1
          Carvery                       1
          Latin American restaurant     1
          Vegetarian restaurant         1
          Name: count, dtype: int64
In [18]:
def categorize_restaurants(restaurant_type):
              if ('bar' in restaurant_type.lower()) or ('pub' in restaurant_type.lower()):
                  return 'pub'
              elif ('takeaway' in restaurant_type.lower()):
                  return 'takeaway'
              elif ('delivery' in restaurant_type.lower()):
                  return 'delivery'
              elif ('restaurant' in restaurant_type.lower()):
                  return 'restaurant'         
              else:
                  return 'others'
              
          
          postcode_df['category_by_primary_service'] = postcode_df['category'].apply(categorize_restaurants)
In [19]:
category_by_primary_service = postcode_df['category_by_primary_service'].value_counts()
          category_by_primary_service_per_postcode = postcode_df.groupby(['outward_code', 'category_by_primary_service']).size().unstack()
          
          plt.figure(figsize=(15, 7.5))
          
          ax1 = plt.subplot(1, 2, 1)
          category_by_primary_service.plot(kind='bar', ax=ax1, color='skyblue')
          ax1.set_xlabel('Primary Service of Restaurants')
          ax1.set_ylabel('Count')
          ax1.set_title('Distribution of Restaurants by Primary Service')
          
          ax2 = plt.subplot(1, 2, 2)
          category_by_primary_service_per_postcode.plot(kind='bar', stacked=True, ax=ax2, colormap='viridis')
          ax2.set_xlabel('Postcode')
          ax2.set_ylabel('Count')
          ax2.set_title('Distribution of Restaurants by Primary Service by Postcode')
          
          plt.tight_layout()
          plt.show()
No description has been provided for this image
In [20]:
plt.figure(figsize=(15, 5))
          
          plt.subplot(1, 3, 1)
          postcode_df['outward_code'].value_counts().plot(kind='bar')
          plt.xlabel('Postcode')
          plt.ylabel('Restaurant Count')
          plt.title('Distribution of Restaurants across Postcodes')
          
          plt.subplot(1, 3, 2)
          cuisine_distribution.head(10).plot(kind='bar', color='orange')
          plt.xlabel('Cuisine')
          plt.ylabel('Cuisine Count')
          plt.title('Distribution of 10 most common Cuisines')
          
          plt.subplot(1, 3, 3)
          cuisine_distribution.tail(10).plot(kind='bar', color='green')
          plt.xlabel('Cuisine')
          plt.ylabel('Cuisine Count')
          plt.title('Distribution of 10 least common Cuisines')
Out[20]:
Text(0.5, 1.0, 'Distribution of 10 least common Cuisines')
No description has been provided for this image

2.2 Geolocation of Restaurant in Portsmouth

In [21]:
postcode_df = postcode_df.dropna(subset=['latitude', 'longitude'])
          location_correlation = postcode_df[['rating', 'latitude', 'longitude']].corr()
In [22]:
location_data = postcode_df[['name', 'rating', 'address', 'latitude', 'longitude']]
          location_data
          
          m = folium.Map([50.79643851477156, -1.1058239725553403], zoom_start=12)
          
          for loc in location_data.iterrows():
              folium.Marker(
                  location=[loc[1]['latitude'], loc[1]['longitude']],
                  tooltip="Click me!",
                  popup=f"<div style='width: 200px;'>{loc[1]['name']}<br>{loc[1]['rating']} rating<br>{loc[1]['address']}</div>",
                  icon=folium.Icon(icon="cloud"),
              ).add_to(m)
          
          print('Visualizing the locations of restaurants in Portmouths on a map')
          m
          
Visualizing the locations of restaurants in Portmouths on a map
Out[22]:
Make this Notebook Trusted to load map: File -> Trust Notebook
In [23]:
# Print Correlation between rating and location
          plt.figure(figsize=(12, 10))
          sns.heatmap(location_correlation)
          
          # No correlation between ration and geolocation
Out[23]:
<Axes: >
No description has been provided for this image

2.3 Distribution of Ratings

In [24]:
rating_df = restaurants_df.copy()
          print(f'Data type of restaurant ratings is {rating_df['rating'].dtype}.')
          ratings_count = rating_df['rating'].value_counts()
          
          print(f'The highest frequency rating for restaurants is {rating_df['rating'].mode()[0]}.')
          
          # Prepare the distribution data
          income_distribution_df = ratings_count.reset_index()
          income_distribution_df.columns = ['Rating', 'Count']
          
          # Create side-by-side plots
          fig, axes = plt.subplots(nrows=1, ncols=2, figsize=(15, 5))
          
          # Plot 1: Distribution of Ratings
          sns.histplot(rating_df['rating'], color='skyblue', kde=True, ax=axes[0])
          axes[0].set_title("Distribution of Ratings")
          axes[0].set_xlabel("Ratings")
          axes[0].set_ylabel("Number of Restaurants")
          
          # Plot 2: Boxplot of Ratings
          sns.boxplot(x=rating_df['rating'], color='skyblue',  ax=axes[1])
          axes[1].set_title("Boxplot of Ratings")
          axes[1].set_xlabel("Ratings")
          
          # Adjust layout for clarity
          plt.tight_layout()
          plt.show()
          
Data type of restaurant ratings is float64.
          The highest frequency rating for restaurants is 4.6.
No description has been provided for this image
In [25]:
print('Summary statistics of rating')
          print(rating_df['rating'].describe())
          
Summary statistics of rating
          count    116.000000
          mean       4.408621
          std        0.324793
          min        2.800000
          25%        4.300000
          50%        4.400000
          75%        4.600000
          max        5.000000
          Name: rating, dtype: float64

2.4 Distribution of Price Range

In [26]:
postcode_df['priceRanges'] = postcode_df['priceRanges'].astype(str)
          postcode_df_filter = postcode_df[~(postcode_df['priceRanges'].str.lower().isin(['nan', 'moderate', 'expensive']))]
          
          # Count occurrences of each price range
          price_range_counts = postcode_df_filter['priceRanges'].value_counts()
          
          # Assuming 'df' contains price_range and count data for the plot
          plt.bar(price_range_counts.index, price_range_counts.values)  
          
          # Customize the plot
          plt.xlabel('Price Range')
          plt.ylabel('Count')
          plt.title('Frequency of Price Ranges')
          plt.xticks(rotation=45)
          plt.show()
No description has been provided for this image
In [27]:
average_rating_per_price_range = postcode_df_filter.groupby('priceRanges')['rating'].agg(['mean', 'min', 'max', 'count'])
          average_rating_per_price_range
Out[27]:
mean min max count
priceRanges
0-0 4.412500 3.2 5.0 8
10–20 4.386957 2.8 5.0 46
10–30 3.900000 3.9 3.9 1
1–10 4.333333 4.1 4.6 3
1–20 4.500000 4.5 4.5 1
20–30 4.423684 3.6 4.8 38
20–40 4.600000 4.4 4.7 6
30–40 4.433333 4.2 4.8 3
In [28]:
most_popular_based_on_ratingCount = postcode_df_filter.groupby('category')['ratingCount'].agg(['size', 'min', 'max', 'mean', 'sum']).sort_values(by='sum', ascending=False)
          most_popular_based_on_ratingCount
Out[28]:
size min max mean sum
category
Restaurant 27 14.0 2392.0 743.703704 20080.0
Pub 7 772.0 2803.0 1567.000000 10969.0
Italian restaurant 6 214.0 2117.0 788.666667 4732.0
Carvery 1 4704.0 4704.0 4704.000000 4704.0
Vietnamese restaurant 3 501.0 1782.0 931.666667 2795.0
Brazilian restaurant 1 2754.0 2754.0 2754.000000 2754.0
Hamburger restaurant 3 229.0 1592.0 796.333333 2389.0
Mexican restaurant 1 2273.0 2273.0 2273.000000 2273.0
Turkish restaurant 4 410.0 788.0 555.000000 2220.0
Steak house 1 2018.0 2018.0 2018.000000 2018.0
Portuguese restaurant 2 359.0 1532.0 945.500000 1891.0
Cafe 2 630.0 1053.0 841.500000 1683.0
Japanese restaurant 2 1475.0 1475.0 1475.000000 1475.0
restaurant 3 267.0 579.0 466.000000 1398.0
Gastropub 2 579.0 785.0 682.000000 1364.0
Cocktail bar 1 1281.0 1281.0 1281.000000 1281.0
Asian restaurant 1 1265.0 1265.0 1265.000000 1265.0
Pizza restaurant 2 353.0 842.0 597.500000 1195.0
British restaurant 1 1121.0 1121.0 1121.000000 1121.0
Lebanese restaurant 1 1085.0 1085.0 1085.000000 1085.0
Belgian restaurant 1 1046.0 1046.0 1046.000000 1046.0
Tex-Mex restaurant 1 1020.0 1020.0 1020.000000 1020.0
Chinese restaurant 5 77.0 456.0 197.800000 989.0
Buffet restaurant 1 976.0 976.0 976.000000 976.0
Fish and chips takeaway 2 441.0 513.0 477.000000 954.0
Greek restaurant 2 288.0 657.0 472.500000 945.0
Tapas bar 1 781.0 781.0 781.000000 781.0
Indian restaurant 1 696.0 696.0 696.000000 696.0
Bar & grill 1 663.0 663.0 663.000000 663.0
Tapas restaurant 1 599.0 599.0 599.000000 599.0
Family restaurant 1 513.0 513.0 513.000000 513.0
Sushi restaurant 1 401.0 401.0 401.000000 401.0
Latin American restaurant 1 394.0 394.0 394.000000 394.0
Caribbean restaurant 1 382.0 382.0 382.000000 382.0
Takeaway 1 340.0 340.0 340.000000 340.0
Chinese Takeaway 1 243.0 243.0 243.000000 243.0
Vegan restaurant 1 242.0 242.0 242.000000 242.0
Kebab Shop 2 82.0 127.0 104.500000 209.0
Dim Sum restaurant 1 197.0 197.0 197.000000 197.0
Malaysian restaurant 1 138.0 138.0 138.000000 138.0
Thai restaurant 1 118.0 118.0 118.000000 118.0
Fast food restaurant 1 117.0 117.0 117.000000 117.0
Filipino restaurant 1 115.0 115.0 115.000000 115.0
African restaurant 1 77.0 77.0 77.000000 77.0
Modern British restaurant 1 59.0 59.0 59.000000 59.0
Pizza Takeaway 1 56.0 56.0 56.000000 56.0
Delivery service 1 17.0 17.0 17.000000 17.0
Vegetarian restaurant 1 2.0 2.0 2.000000 2.0

3 Summary of the Analysis

3.1. Distribution of Postcodes in Portsmouth¶

  1. Distribution by Outward Code (Pie Chart)
    The pie chart shows the distribution of dining establishments across Portsmouth's postal areas. PO1 accounts for 44% and PO5 for 33.6% of the total, making these the most restaurant-dense areas and key central hubs. Other areas, including PO2, PO4, PO6, PO3, and PO12, have significantly smaller shares, ranging from 0.9% to 8.6%.
    Note: The term "restaurants" is used broadly to include various types of dining services categorized by outward code.

  2. Count of Restaurants by Postcode (Bar Chart)
    The bar chart reinforces that PO1 and PO5 have the highest counts of dining establishments, while PO3 and PO12 have the fewest. This further confirms the centrality of PO1 and PO5 as primary dining hotspots.

  3. Primary Service Distribution (Bar Chart)
    Restaurants dominate as the leading service type, with over 80 establishments, far surpassing the secondary categories such as pubs and others, which range between 15 and 20 establishments each. Delivery services are sparsely represented, suggesting that the local market strongly favors dine-in experiences over delivery or takeaway services.

  4. Primary Service by Postcode (Bar Chart)
    PO1 and PO5 host the highest variety and number of services, with restaurants being the predominant category. Pubs and other service types contribute meaningfully in these areas but are sparse in postcodes like PO3 and PO12, where fewer dining establishments and limited service diversity are observed.

3.2. Geolocation of Restaurants in Portsmouth¶

  1. Geolocation Map
    The geolocation map reveals a dense clustering of dining establishments in the southern and southwestern parts of Portsmouth, aligning with the PO1 and PO5 regions. This geographic pattern corresponds closely with the postal code distribution charts, confirming these areas as dining and activity hubs.

  2. Correlation Heatmap

    • Weak Correlation with Ratings: The heatmap reveals a negligible or weak correlation between ratings and geographic location (latitude or longitude). This indicates that restaurant quality, as perceived through ratings, is not significantly influenced by geographic factors, reaffirming that other attributes, such as service quality or cuisine, might play a more critical role.

3.3. Distribution of Ratings¶

  1. Rating Histogram (Left Panel)
    The histogram shows that most restaurants have ratings clustered between 4.0 and 4.5, with a peak around 4.5. The slight left skew suggests high ratings are the norm, with very few establishments receiving scores below 4.0.

  2. Rating Boxplot (Right Panel)
    The boxplot confirms a median rating close to 4.5, with an interquartile range (IQR) between 4.0 and 4.5. Outliers below 4.0 are minimal, indicating a high degree of consistency in customer satisfaction across restaurants in Portsmouth.

3.4. Distribution of Price Ranges¶

  1. Price Range Frequency (Bar Chart)
    The "10-20" and "20-30" price ranges dominate, reflecting a strong customer preference for mid-range dining options. Lower price ranges ("0-10") and higher ranges ("30-40") are less common, suggesting limited demand for extreme budget or premium dining experiences.

Summary of Visualizations¶

  1. Distribution by Service Type and Postcode

    • Restaurants dominate across all postcodes, especially in PO1 and PO5, which are primary dining hubs.
    • Other service types, like pubs and others, are present in smaller numbers and are less evenly distributed. Delivery services remain minimal across all areas.

  2. Geographic Trends

    • Dining establishments are heavily clustered in the southern and southwestern parts of Portsmouth, aligning with PO1 and PO5.
    • Customer ratings are generally high and consistent, with little geographic variation, highlighting quality dining experiences irrespective of location.

  3. Customer Preferences

    • Restaurants are the clear preference among service types, with ratings concentrated between 4.0 and 4.5.
    • Mid-range pricing ("10-30") is the most common, catering to a broad audience, while niche or premium pricing categories are less represented.

  4. Cuisines and Niche Offerings

    • The top 10 cuisines, such as Italian and Chinese, dominate, while niche offerings like vegetarian or buffet restaurants are rare, indicating potential gaps in the market.

Key Insights¶

  1. Central Dining Hubs:
    PO1 and PO5 are the epicenters of restaurant activity, offering the greatest variety and number of dining establishments. Businesses looking to capitalize on foot traffic and customer demand should prioritize these areas.

  2. Dominance of Restaurants:
    Traditional dine-in restaurants significantly outnumber all other service types, reflecting a strong preference for sit-down dining over takeaway or delivery services.

  3. High Customer Satisfaction:
    The majority of establishments achieve ratings above 4.0, with a median close to 4.5, suggesting consistently high service quality across the city.

  4. Pricing Alignment:
    The dominance of mid-range price options ("10-30") indicates affordability is a key consideration for customers, while demand for budget or luxury dining is limited.

  5. Limited Diversity in Services and Cuisines:
    Niche categories like delivery services and less common cuisines, such as vegetarian or buffet restaurants, are underrepresented, suggesting an opportunity to address untapped market segments.

  6. Independent Ratings:
    The lack of correlation between ratings and location suggests that restaurant quality is distributed evenly across different areas and does not necessarily depend on proximity to the central hubs.

Conclusion¶

The analysis reveals that PO1 and PO5 serve as Portsmouth's primary dining hubs, hosting the majority of establishments and showcasing diverse services and cuisines. Restaurants dominate as the preferred service type, supported by consistently high customer ratings and a focus on mid-range pricing. While the market caters well to mainstream preferences, opportunities exist to introduce niche cuisines and expand services like delivery, especially in underrepresented areas like PO3 and PO12.

Recommendations:

  • Businesses should target PO1 and PO5 for expansion while maintaining mid-range pricing and high-quality service.
  • Entrepreneurs might consider filling gaps in niche services, such as delivery and vegetarian options, to capture underserved customer segments.