Exploratory Data Analysis

Analysis OF PRICE ON INDEPENDENT HOUSES IN HYDERABAD

Data collected from the square yards website.

square yards, is a real estate website that helps you to buy or rent your dream house. In this website you can search for different types of houses or for villas or for independent houses.

import re
import numpy as np
import requests
from bs4 import BeautifulSoup
!pip install selenium
from selenium import webdriver
from selenium import webdriver

import re
url = 'https://www.squareyards.com/sale/independent-houses-for-sale-in-hyderabad'
page = requests.get(url)
page
soup = BeautifulSoup(page.text)
soup

you should get the page response as 200

loc = ['Kukatpally','Gachibowli','Nampalli','Ameerpet','Shankarpalli','Kondapur','Tarnaka','Nizampet','Balanagar','Cherlapally']

bhk = []
location=[]
furnish = []
bathroom = []
possessionstatus = []
area = []
noofimages = []
tenantpre = []
price=[]
for x in loc: 
    url = f'https://www.squareyards.com/sale/independent-houses-for-sale-in-{x}-hyderabad' 
    driver = webdriver.Chrome(r"C:\Users\Bhavitha Cherukuri\Downloads\chromedriver_win32(1)")
    driver.get(url)
    soup = BeautifulSoup(driver.page_source,'html.parser')
    driver.close()

    
    a = soup.find_all("div",class_="tlBdy")#inspecting website and getting class and container code
    for i in a:
        text = i.text
        regex = re.findall('(\d.*)\sBHK',text)
        if regex:
            bhk.append(regex[0])
        else:
            bhk.append(np.nan)
     
    all_loc = soup.find_all("div",class_="tlBdy")
    for i in all_loc:
        text=i.text    
        regex=re.findall("Kukatpally|Gachibowli|Nampalli|Ameerpet|Shankarpalli|Kondapur|Tarnaka|Nizampet|Balanagar|Cherlapally",text)
        #print(regex)
        if regex:
            location.append(regex[0])  
        else:
            location.append('Balanagar')
            
    b = soup.find_all("div",class_="tlBdy")
    for i in b:
        text = i.text
        print(text)
        regex=re.findall("Furnished|Semi-Furnished|Unfurnished",text)
        #print(regex)
        if regex:
            furnish.append(regex[0])
        else:
            furnish.append('Semi-Furnished')
            
    b = soup.find_all("div",class_="tlBdy")
    for i in b:
        text = i.text
        regex=re.findall(",?\s(\d)\s?bathroom|bathrooms",text)#regex for particular value
        #print(regex)
        if regex:
            bathroom.append(regex[0])
        else:
            bathroom.append(2)
            
    
    b = soup.find_all("ul",class_="tlPrjctUl")
    for i in b:
        text = i.text
        print(text)
        regex=re.findall("(Ready To Move| Under Construction)",text)
        #print(regex)
        if regex:
            possessionstatus.append(regex[0])
        else:
            possessionstatus.append('Under Construction')

    for i in b:
        text = i.text
        regex=re.findall("(\d.*)\sSq",text)
        #print(regex)
        if regex:
            area.append(regex[0])
        else:
            area.append('2023')
            
    b = soup.find_all("div",class_="tlBdy")
    for i in b:
        text = i.text
        regex=re.findall("(\d+)\s\s",text)
        if regex:
            noofimages.append(regex[0])
        else:
            noofimages.append(0)
        
    for i in b:
        text = i.text
        regex=re.findall("(?:Bachelors|Bachelors/Family|Family)",text)
        #print(regex)
        if regex:
            tenantpre.append(regex[0])
        else:
            tenantpre.append('Bachelors')
    
    for i in b:
        text=i.text
        regex=re.findall("₹\s(\d.*)",text)
        if regex:
            price.append(regex[0])
        else:
            price.append(0)

Now by using the all the variables used u can print the details related to them.

creating a data frame

data={"location":location,"furnish":furnish,"bathroom":bathroom,"tenantpre":tenantpre,"bhk":bhk,"possessionstatus":possessionstatus,"area":area,"noofimages":noofimages,"price":price}

importing the libraries

houses = pd.DataFrame(data)
houses

houses.info

#converting csv data for rentals dataframe
houses_csv_data = houses.to_csv('houses_project2.csv', index = True)
print('\nCSV String:\n', houses_csv_data)

Reading csv file

df = pd.read_csv('houses_project2.csv')
df

Cleaning the data by removing special symbols and null values

df['bhk']=df['bhk'].str.replace('+',' ')
df

I have shown only one step and there are few more columns to clean the data and Now the data is cleaned as shown in above picture.

Dropping unnecessary columns

df.drop(['Unnamed: 0' ],axis = 1,inplace=True)

checking for outliers

df.isnull().sum()

Removing outliers from data fram using minimum and maximum iqr of numerical values

import seaborn as sns

iqr_df = df['bathroom'].describe().loc['75%'] - df['bathroom'].describe().loc['25%']
iqr_df   

ufbathroom = df['bathroom'].describe().loc['75%'] + iqr_df*1.5
ufbathroom

lfbathroom = df['bathroom'].describe().loc['25%'] - iqr_df*1.5
lfbathroom

df = df[df['bathroom']<ufbathroom]
df.shape

By continuing this process the data is cleaned

df.isnull().sum()

import matplotlib.pyplot as plt
import seaborn as sns

sns.histplot(df['area']) # there is more demand for area = 1500, as the count is 80

sns.kdeplot(df["bathroom"],color='r',shade=True)# there is demand for houses with 2 bathroom 

plt.figure(figsize=(10,4)) # there is more demand for houses in Balanagar
sns.countplot(x='location',data=df)
plt.title("count of locations")
plt.xticks(rotation=90)
plt.show()

sns.countplot(df['tenantpre']) # most of the houses are for families

df['furnish'].value_counts().plot(kind = 'bar') # the houses that are unfurnished are more

sns.boxplot(df['tenantpre'],df['area'],hue=df['bathroom']) # most of the families are preffering 2 bathroom and bachelors 3 bathrooms

sns.heatmap(df.corr(),
            annot=True,
            fmt = '.2f',)

And thus there is more demand for houses in balanagar and most of the houses are semi-furnished. As, these are some insights and go through the story to find out more insights.