Machine Learning Tutorials, Courses and Certifications
Analysing Airport Review DataSets¶
In [1]:
import pandas as pd
In [2]:
#Supress Warnings
import warnings
warnings.filterwarnings('ignore')
In [3]:
airport_data=pd.read_csv('airport.csv')
print(airport_data.shape)
airport_data.head(1)
(17721, 20)
Out[3]:
| airport_name | link | title | author | author_country | date | content | experience_airport | date_visit | type_traveller | overall_rating | queuing_rating | terminal_cleanliness_rating | terminal_seating_rating | terminal_signs_rating | food_beverages_rating | airport_shopping_rating | wifi_connectivity_rating | airport_staff_rating | recommended | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | aalborg-airport | /airport-reviews/aalborg-airport | Aalborg Airport customer review | Klaus Malling | Denmark | 2014-02-11 | A small very effective airport with few flight… | NaN | NaN | NaN | 9.0 | 5.0 | 5.0 | NaN | NaN | NaN | 4.0 | NaN | NaN | 1 |
In [4]:
airport_data.describe()
Out[4]:
| overall_rating | queuing_rating | terminal_cleanliness_rating | terminal_seating_rating | terminal_signs_rating | food_beverages_rating | airport_shopping_rating | wifi_connectivity_rating | airport_staff_rating | recommended | |
|---|---|---|---|---|---|---|---|---|---|---|
| count | 13796.000000 | 12813.000000 | 12815.000000 | 587.000000 | 27.000000 | 630.000000 | 12676.000000 | 412.000000 | 26.000000 | 17721.000000 |
| mean | 4.274355 | 2.747912 | 3.442450 | 2.580920 | 2.592593 | 2.169841 | 2.821631 | 2.405340 | 2.038462 | 0.221206 |
| std | 2.722765 | 1.572520 | 1.337508 | 1.403862 | 1.393923 | 1.534358 | 1.410575 | 1.579452 | 1.248384 | 0.415071 |
| min | 1.000000 | 0.000000 | 0.000000 | 0.000000 | 1.000000 | 0.000000 | 0.000000 | 0.000000 | 1.000000 | 0.000000 |
| 25% | 2.000000 | 1.000000 | 3.000000 | 1.000000 | 1.000000 | 1.000000 | 2.000000 | 1.000000 | 1.000000 | 0.000000 |
| 50% | 4.000000 | 3.000000 | 3.000000 | 2.000000 | 3.000000 | 2.000000 | 3.000000 | 2.000000 | 1.500000 | 0.000000 |
| 75% | 6.000000 | 4.000000 | 5.000000 | 4.000000 | 4.000000 | 3.000000 | 4.000000 | 4.000000 | 3.000000 | 0.000000 |
| max | 10.000000 | 5.000000 | 5.000000 | 5.000000 | 5.000000 | 5.000000 | 5.000000 | 5.000000 | 4.000000 | 1.000000 |
In [5]:
airport_data.info()
<class 'pandas.core.frame.DataFrame'> RangeIndex: 17721 entries, 0 to 17720 Data columns (total 20 columns): airport_name 17721 non-null object link 17721 non-null object title 17721 non-null object author 17721 non-null object author_country 12777 non-null object date 17721 non-null object content 17721 non-null object experience_airport 647 non-null object date_visit 593 non-null object type_traveller 646 non-null object overall_rating 13796 non-null float64 queuing_rating 12813 non-null float64 terminal_cleanliness_rating 12815 non-null float64 terminal_seating_rating 587 non-null float64 terminal_signs_rating 27 non-null float64 food_beverages_rating 630 non-null float64 airport_shopping_rating 12676 non-null float64 wifi_connectivity_rating 412 non-null float64 airport_staff_rating 26 non-null float64 recommended 17721 non-null int64 dtypes: float64(9), int64(1), object(10) memory usage: 2.7+ MB
In [6]:
#Percentage Null Values
round((airport_data.isna().sum()/len(airport_data)*100),2)
Out[6]:
airport_name 0.00 link 0.00 title 0.00 author 0.00 author_country 27.90 date 0.00 content 0.00 experience_airport 96.35 date_visit 96.65 type_traveller 96.35 overall_rating 22.15 queuing_rating 27.70 terminal_cleanliness_rating 27.68 terminal_seating_rating 96.69 terminal_signs_rating 99.85 food_beverages_rating 96.44 airport_shopping_rating 28.47 wifi_connectivity_rating 97.68 airport_staff_rating 99.85 recommended 0.00 dtype: float64
In [7]:
filtered_airport_data=airport_data.drop(['experience_airport','date_visit','type_traveller','terminal_seating_rating','terminal_signs_rating','food_beverages_rating','wifi_connectivity_rating','airport_staff_rating'],axis=1)
In [8]:
round((filtered_airport_data.isna().sum()/len(filtered_airport_data)*100),2)
Out[8]:
airport_name 0.00 link 0.00 title 0.00 author 0.00 author_country 27.90 date 0.00 content 0.00 overall_rating 22.15 queuing_rating 27.70 terminal_cleanliness_rating 27.68 airport_shopping_rating 28.47 recommended 0.00 dtype: float64
In [9]:
airport_name=filtered_airport_data.airport_name.unique()
print("Total Aiports Conisiderd for Analysis:",len(airport_name))
Total Aiports Conisiderd for Analysis: 741
Counting Recommendations¶
In [10]:
# Recommendation Count
recommendation_count=[]
not_recommendation_count=[]
total_reviews=[]
count=0
for i in airport_name:
temp_df=filtered_airport_data[filtered_airport_data['airport_name']==i]
rec=(temp_df['recommended']==1).sum()
total_reviews.append(len(temp_df))
recommendation_count.append(rec)
not_recommendation_count.append(len(temp_df)-rec)
In [11]:
result=zip(airport_name,total_reviews,recommendation_count,not_recommendation_count)
In [ ]:
In [12]:
recommendation_df=pd.DataFrame(result,columns=['airport_name','total_reviews','recommendation_count','not_recommended'])
In [16]:
recommendation_df.sort_values(by='total_reviews',ascending=False)
Out[16]:
| airport_name | total_reviews | recommendation_count | not_recommended | |
|---|---|---|---|---|
| 410 | london-heathrow-airport | 520 | 160 | 360 |
| 411 | london-stansted-airport | 402 | 41 | 361 |
| 437 | manchester-airport | 303 | 65 | 238 |
| 519 | paris-cdg-airport | 301 | 48 | 253 |
| 210 | dubai-airport | 279 | 63 | 216 |
| 420 | luton-airport | 275 | 28 | 247 |
| 409 | london-gatwick-airport | 252 | 89 | 163 |
| 60 | bangkok-suvarnabhumi-airport | 220 | 69 | 151 |
| 242 | frankfurt-main-airport | 218 | 43 | 175 |
| 414 | los-angeles-lax-airport | 199 | 34 | 165 |
| 455 | miami-airport | 191 | 21 | 170 |
| 488 | new-york-jfk-airport | 185 | 43 | 142 |
| 624 | singapore-changi-airport | 181 | 101 | 80 |
| 392 | leeds-bradford-airport | 166 | 26 | 140 |
| 675 | toronto-pearson-airport | 166 | 41 | 125 |
| 31 | amsterdam-schiphol-airport | 166 | 77 | 89 |
| 304 | hong-kong-airport | 162 | 78 | 84 |
| 366 | klia-kuala-lumpur-airport | 160 | 64 | 96 |
| 581 | rome-fiumicino-airport | 155 | 25 | 130 |
| 116 | bristol-airport | 154 | 32 | 122 |
| 450 | melbourne-airport | 149 | 26 | 123 |
| 324 | istanbul-ataturk-airport | 149 | 33 | 116 |
| 472 | mumbai-airport | 149 | 42 | 107 |
| 440 | manila-ninoy-aquino-airport | 138 | 25 | 113 |
| 308 | houston-george-bush-intercontinental-airport | 138 | 18 | 120 |
| 652 | sydney-airport | 132 | 41 | 91 |
| 195 | delhi-airport | 130 | 63 | 67 |
| 91 | birmingham-airport | 129 | 37 | 92 |
| 489 | newark-airport | 129 | 13 | 116 |
| 405 | liverpool-airport | 128 | 30 | 98 |
| … | … | … | … | … |
| 232 | fairbanks-airport | 1 | 0 | 1 |
| 536 | pointe-a-pitre-airport | 1 | 0 | 1 |
| 321 | irkutsk-airport | 1 | 0 | 1 |
| 145 | changzhou-airport | 1 | 0 | 1 |
| 1 | aarhus-airport | 1 | 0 | 1 |
| 550 | prince-rupert-airport | 1 | 0 | 1 |
| 50 | baghdad-airport | 1 | 0 | 1 |
| 49 | bagdogra-airport | 1 | 0 | 1 |
| 48 | bacau-airport | 1 | 0 | 1 |
| 666 | thandwe-airport | 1 | 0 | 1 |
| 47 | ayers-rock-airport | 1 | 0 | 1 |
| 523 | pensacola-airport | 1 | 0 | 1 |
| 541 | port-harcourt-airport | 1 | 0 | 1 |
| 648 | sukhothai-airport | 1 | 0 | 1 |
| 377 | kuantan-airport | 1 | 0 | 1 |
| 542 | port-macquarie-airport | 1 | 0 | 1 |
| 645 | strasbourg-airport | 1 | 0 | 1 |
| 66 | barnaul-airport | 1 | 0 | 1 |
| 279 | hakodate-airport | 1 | 0 | 1 |
| 642 | stockholm-bromma-airport | 1 | 0 | 1 |
| 67 | barra-eoligarry-airport | 1 | 0 | 1 |
| 544 | port-vila-bauerfield-airport | 1 | 0 | 1 |
| 69 | bay-of-islands-kerikeri-airport | 1 | 0 | 1 |
| 637 | st-pete-clearwater-international-airport | 1 | 0 | 1 |
| 278 | hailar-dongshan-airport | 1 | 1 | 0 |
| 382 | la-rochelle-ile-de-re-airport | 1 | 0 | 1 |
| 131 | calcutta-airport | 1 | 0 | 1 |
| 633 | springfield-branson-airport | 1 | 1 | 0 |
| 549 | preveza-airport | 1 | 0 | 1 |
| 370 | kos-airport | 1 | 0 | 1 |
741 rows × 4 columns
Calculating the Polarity of Each Review¶
In [18]:
import numpy as np
import matplotlib.pyplot as plt
from textblob import TextBlob
In [23]:
result=[]
num=1
numlist=[]
for review in filtered_airport_data.content:
analysis=TextBlob(review)
result.append(analysis.polarity)
numlist.append(num)
num=num+1
result=np.array(result)
result
Out[23]:
array([0.24735294, 0.09988529, 0.33133333, ..., 0.15833333, 0.48240741,
0.19375 ])
Plotting the Polarity¶
In [41]:
plt.figure(figsize=(16,12))
plt.style.use('seaborn')
plt.scatter(numlist,result,label='Polarity')
plt.xlabel('Reviews')
plt.ylabel('Polarity')
plt.legend()
Out[41]:
<matplotlib.legend.Legend at 0x7f2a88ff4c88>
In [33]:
average_polarity=result.sum()/len(result)
print('Averge Plority : {}'.format(str(average_polarity)))
Averge Plority : 0.08175742657371485
In [43]:
plt.style.use('seaborn')
plt.figure(figsize=(16,12))
plt.plot(numlist,result,label='Polarity')
plt.xlabel('Reviews')
plt.ylabel('Polarity')
plt.legend()
Out[43]:
<matplotlib.legend.Legend at 0x7f2a88f0dc88>
