Data Analytics Program

 

Data Analytics Program:

Build a simple linear regression model for Salary Prediction from years of experience.(Download Salary dataset from kaggle). Find the accuracy of the model.

In [ ]:

1

​

In [61]:

1

​

2

import pandas as pd

3

​

In [62]:

1

​

2

data = pd.read_csv("Salary.csv")

3

data

4

​

Out[62]:

	experience	salary
0	1	100
1	2	200
2	3	300
3	4	405
4	5	499
5	7	700
6	8	800
7	9	900

In [63]:

1

​

2

data.isnull()

3

​

Out[63]:

	experience	salary
0	False	False
1	False	False
2	False	False
3	False	False
4	False	False
5	False	False
6	False	False
7	False	False

In [64]:

1

​

2

data.describe()

3

​

Out[64]:

	experience	salary
count	8.000000	8.00000
mean	4.875000	488.00000
std	2.900123	289.79648
min	1.000000	100.00000
25%	2.750000	275.00000
50%	4.500000	452.00000
75%	7.250000	725.00000
max	9.000000	900.00000

In [65]:

1

​

2

data.head()

3

​

Out[65]:

	experience	salary
0	1	100
1	2	200
2	3	300
3	4	405
4	5	499

In [66]:

1

​

2

data.tail()

3

​

Out[66]:

	experience	salary
3	4	405
4	5	499
5	7	700
6	8	800
7	9	900

In [67]:

1

​

2

data.sample()

3

​

Out[67]:

	experience	salary
3	4	405

In [68]:

1

​

2

data.dtypes

3

​

Out[68]:

experience    int64
salary        int64
dtype: object

In [69]:

1

​

2

corr = data.corr()

3

corr

4

corr.style.background_gradient(cmap='rainbow')

Out[69]:

	experience	salary
experience	1.000000	0.999980
salary	0.999980	1.000000

In [70]:

1

y = data['salary']

2

x = data.drop(['salary'],axis = 1)

3

y

Out[70]:

0    100
1    200
2    300
3    405
4    499
5    700
6    800
7    900
Name: salary, dtype: int64

In [71]:

1

​

2

x

3

​

Out[71]:

	experience
0	1
1	2
2	3
3	4
4	5
5	7
6	8
7	9

In [72]:

1

​

2

import matplotlib.pyplot as plt

3

​

In [73]:

1

plt.scatter(x,y)

2

plt.title("Salary vs Experience (Training Dataset)") 

3

plt.xlabel(" Experience in Year") 

4

plt.ylabel("Salary(In Rupees)") 

5

plt.show()

6

​

In [74]:

1

from sklearn.model_selection import train_test_split

2

x_train,x_test,y_train,y_test = train_test_split(x,y,test_size=0.25)

In [75]:

1

​

2

print(x_train)

3

​

   experience
4           5
7           9
0           1
3           4
5           7
1           2

In [76]:

1

​

2

print(y_train)

3

​

4    499
7    900
0    100
3    405
5    700
1    200
Name: salary, dtype: int64

In [77]:

1

​

2

print(x_test.shape)

3

​

(2, 1)

In [78]:

1

​

2

from sklearn.linear_model import LinearRegression

3

slr=LinearRegression()

4

​

In [79]:

1

​

2

slr.fit(x_train,y_train)

3

​

Out[79]:

LinearRegression()

In [80]:

1

​

2

y_pred=slr.predict(x_test)

3

y_pred

4

​

Out[80]:

array([300.80147059, 800.39705882])

In [81]:

1

df=pd.DataFrame({'Actual':y_test,'predicted':y_pred})

2

df

Out[81]:

	Actual	predicted
2	300	300.801471
6	800	800.397059

In [82]:

1

​

2

print(slr.predict([[10]]))

3

​

[1000.23529412]

In [83]:

1

plt.scatter(x_test, y_test, color="green") 

2

plt.plot(x_test, y_pred, color="red") 

3

plt.title("Salary vs Experience (Testing Dataset)") 

4

plt.xlabel("Years of Experience") 

5

plt.ylabel("Salary(In Rupees)") 

6

plt.show() 

In [ ]:

1

​

In [ ]:

1

​

In [ ]:

1

​

In [ ]:

1

​

In [ ]:

1

​

Q2. Use the house price prediction dataset to build a multiple linear regression model for predicting purchases. Identify independent and target variable. Split the variables into training and testing sets and print them. Find the accuracy of the model.

In [1]:

1

​

2

import pandas as pd

3

​

In [88]:

1

​

2

data = pd.read_csv("HousePrice.csv")

3

data

4

​

Out[88]:

	area	bedrooms	age	price
0	1000	3.0	20	550000
1	1004	4.0	21	56000
2	1200	5.0	22	60000
3	1300	NaN	23	70000
4	2000	6.0	30	80000
5	3000	8.0	34	90000
6	4000	9.0	40	99000

In [89]:

1

​

2

data.describe

3

​

Out[89]:

<bound method NDFrame.describe of    area  bedrooms  age   price
0  1000       3.0   20  550000
1  1004       4.0   21   56000
2  1200       5.0   22   60000
3  1300       NaN   23   70000
4  2000       6.0   30   80000
5  3000       8.0   34   90000
6  4000       9.0   40   99000>

In [90]:

1

data.dtypes

Out[90]:

area          int64
bedrooms    float64
age           int64
price         int64
dtype: object

In [91]:

1

​

2

data.head()

3

​

Out[91]:

	area	bedrooms	age	price
0	1000	3.0	20	550000
1	1004	4.0	21	56000
2	1200	5.0	22	60000
3	1300	NaN	23	70000
4	2000	6.0	30	80000

In [92]:

1

​

2

data.isna()

3

​

Out[92]:

	area	bedrooms	age	price
0	False	False	False	False
1	False	False	False	False
2	False	False	False	False
3	False	True	False	False
4	False	False	False	False
5	False	False	False	False
6	False	False	False	False

In [93]:

1

​

2

data.isna().sum()

3

​

Out[93]:

area        0
bedrooms    1
age         0
price       0
dtype: int64

In [94]:

1

​

2

data['bedrooms'] = data['bedrooms'].fillna(data['bedrooms'].mean())

3

data

4

​

Out[94]:

	area	bedrooms	age	price
0	1000	3.000000	20	550000
1	1004	4.000000	21	56000
2	1200	5.000000	22	60000
3	1300	5.833333	23	70000
4	2000	6.000000	30	80000
5	3000	8.000000	34	90000
6	4000	9.000000	40	99000

In [95]:

1

​

2

from sklearn.linear_model import LinearRegression

3

​

In [96]:

1

#  Set dependent and independent variables

2

y = data['price']

3

x = data.drop(['price'], axis=1)

4

x

Out[96]:

	area	bedrooms	age
0	1000	3.000000	20
1	1004	4.000000	21
2	1200	5.000000	22
3	1300	5.833333	23
4	2000	6.000000	30
5	3000	8.000000	34
6	4000	9.000000	40

In [97]:

1

​

2

y

3

​

Out[97]:

0    550000
1     56000
2     60000
3     70000
4     80000
5     90000
6     99000
Name: price, dtype: int64

In [98]:

1

​

2

from sklearn.model_selection import train_test_split

3

x_train,x_test,y_train,y_test = train_test_split(x,y,test_size=0.25)

In [99]:

1

​

2

x_train

3

​

Out[99]:

	area	bedrooms	age
1	1004	4.000000	21
4	2000	6.000000	30
2	1200	5.000000	22
0	1000	3.000000	20
3	1300	5.833333	23

In [100]:

1

​

2

y_train

3

​

Out[100]:

1     56000
4     80000
2     60000
0    550000
3     70000
Name: price, dtype: int64

In [101]:

1

​

2

print(x_test.shape)

3

​

(2, 3)

In [102]:

1

​

2

from sklearn.linear_model import LinearRegression

3

slr=LinearRegression()

4

​

In [103]:

1

​

2

slr.fit(x_train,y_train)

3

​

Out[103]:

LinearRegression()

In [104]:

1

​

2

y_pred=slr.predict(x_test)

3

y_pred

Out[104]:

array([2578443.38316721, 1507365.03322259])

In [105]:

1

​

2

df=pd.DataFrame({'Actual':y_test,'predicted':y_pred})

3

df

4

​

Out[105]:

	Actual	predicted
6	99000	2.578443e+06
5	90000	1.507365e+06

In [ ]:

1

​

2

slr.predict([[10]])

3

​

In [ ]:

1

​

Create ‘User’ Data set having 5 columns namely: User ID, Gender, Age, EstimatedSalary and Purchased. Build a logistic regression model that can predict whether on the given parameter a person will buy a car or not. Display confusion matrix.

In [20]:

1

import pandas as pd

2

import numpy as np

3

from sklearn.model_selection import train_test_split

4

from sklearn.linear_model import LogisticRegression

5

import matplotlib.pyplot as plt

In [21]:

1

df = pd.read_csv("car.csv")

2

df

Out[21]:

	Gender	Age	Salary	Purchased
0	male	44	72000	No
1	male	27	48000	Yes
2	female	30	54000	No
3	male	38	61000	No
4	female	40	30000	Yes
5	female	35	58000	Yes
6	male	37	52000	No
7	male	48	79000	Yes
8	male	50	83000	No
9	female	37	67000	Yes

In [22]:

1

from sklearn.preprocessing import LabelEncoder

2

le=LabelEncoder()

3

df['Gender']=le.fit_transform(df['Gender'])

4

df

Out[22]:

	Gender	Age	Salary	Purchased
0	1	44	72000	No
1	1	27	48000	Yes
2	0	30	54000	No
3	1	38	61000	No
4	0	40	30000	Yes
5	0	35	58000	Yes
6	1	37	52000	No
7	1	48	79000	Yes
8	1	50	83000	No
9	0	37	67000	Yes

In [23]:

1

df['Purchased']=le.fit_transform(df['Purchased'])

2

df

Out[23]:

	Gender	Age	Salary	Purchased
0	1	44	72000	0
1	1	27	48000	1
2	0	30	54000	0
3	1	38	61000	0
4	0	40	30000	1
5	0	35	58000	1
6	1	37	52000	0
7	1	48	79000	1
8	1	50	83000	0
9	0	37	67000	1

In [24]:

1

​

2

df.notnull()

3

​

Out[24]:

	Gender	Age	Salary	Purchased
0	True	True	True	True
1	True	True	True	True
2	True	True	True	True
3	True	True	True	True
4	True	True	True	True
5	True	True	True	True
6	True	True	True	True
7	True	True	True	True
8	True	True	True	True
9	True	True	True	True

In [25]:

1

x = df.iloc[:, [0,1,2]].values

2

y = df.iloc[:, 3].values

3

​

In [26]:

1

​

2

x

3

​

Out[26]:

array([[    1,    44, 72000],
       [    1,    27, 48000],
       [    0,    30, 54000],
       [    1,    38, 61000],
       [    0,    40, 30000],
       [    0,    35, 58000],
       [    1,    37, 52000],
       [    1,    48, 79000],
       [    1,    50, 83000],
       [    0,    37, 67000]], dtype=int64)

In [27]:

1

​

2

y

3

​

Out[27]:

array([0, 1, 0, 0, 1, 1, 0, 1, 0, 1])

In [28]:

1

​

2

# Splitting the dataset into training and test set. 

3

from sklearn.model_selection import train_test_split 

4

x_train, x_test, y_train, y_test= train_test_split(x, y, test_size= 0.50, random_state=

5

0) 

In [29]:

1

#feature Scaling 

2

from sklearn.preprocessing import StandardScaler 

3

st_x= StandardScaler() 

4

x_train= st_x.fit_transform(x_train) 

5

x_test= st_x.fit_transform(x_test) 

6

x_test

7

x_train

8

​

Out[29]:

array([[ 0.5       , -0.70226353, -1.26927717],
       [ 0.5       ,  1.56976553,  1.49447151],
       [ 0.5       , -0.49571543, -0.34802761],
       [ 0.5       ,  0.74357315,  0.77794407],
       [-2.        , -1.11535972, -0.6551108 ]])

In [30]:

1

#Fitting Logistic Regression to the training set 

2

from sklearn.linear_model import LogisticRegression 

3

classifier= LogisticRegression(random_state=10) 

4

classifier.fit(x_train, y_train)

5

​

Out[30]:

LogisticRegression(random_state=10)

In [31]:

1

#Predicting the test set result 

2

y_pred= classifier.predict(x_test) 

3

y_pred

4

​

5

​

Out[31]:

array([0, 0, 0, 1, 0])

In [39]:

1

#Creating the Confusion matrix 

2

from sklearn.metrics import confusion_matrix 

3

import seaborn as sn

4

confusion_matrix = pd.crosstab(y_test, y_pred, rownames=['Actual'], colnames=['Predicted'])

5

sn.heatmap(confusion_matrix, annot=True)

6

​

Out[39]:

<AxesSubplot:xlabel='Predicted', ylabel='Actual'>

In [51]:

1

confusion_matrix

2

from sklearn.metrics import confusion_matrix 

3

​

In [ ]:

1

​

2

print('Accuracy: ',metrics.accuracy_score(y_test, y_pred))

3

​

In [ ]:

1

​

2

Accuracy: 0.6

3

In [ ]:

1

​

Frequent itemset and Association rule mining

#dataset creation in python transactions = [['bread','milk'],['bread', 'diaper''milk','eggs'],['milk', 'diaper','beer','coke'], ['bread','milk','diaper','beer'], ['bread','milk','diaper','coke']]

In [4]:

1

​

2

transactions

3

​

Out[4]:

[['bread', 'milk'],
 ['bread', 'diapermilk', 'eggs'],
 ['milk', 'diaper', 'beer', 'coke'],
 ['bread', 'milk', 'diaper', 'beer'],
 ['bread', 'milk', 'diaper', 'coke']]

In [ ]:

1

​

2

pip install mlxtend

3

​

In [ ]:

1

​

2

import pandas as pd

3

from mlxtend.frequent_patterns import apriori, association_rules

4

​

In [ ]:

1

from mlxtend.preprocessing import TransactionEncoder

2

te=TransactionEncoder()

3

te_array=te.fit(transactions).transform(transactions)

4

df=pd.DataFrame(te_array, columns=te.columns_)

5

df

6

​

In [ ]:

1

beer bread coke diaper diapermilk eggs milk

2

0 False True False False False False True

3

1 False True False False True True False

4

2 True False True True False False True

5

3 True True False True False False True

6

4 False True True True False False True

In [ ]:

1

​

2

from sklearn.preprocessing import LabelEncoder

3

df.apply(LabelEncoder().fit_transform)

4

​

In [ ]:

1

beer bread coke diaper diapermilk eggs milk

2

0 0 1 0 0 0 0 1

3

1 0 1 0 0 1 1 0

4

2 1 0 1 1 0 0 1

5

3 1 1 0 1 0 0 1

6

4 0 1 1 1 0 0 1

7

​

In [ ]:

1

​

2

Find the frequent itemsets Generate frequent itemsets that have a support value of at least 50%. By default,

3

apriori returns the column indices of the items, For better readability, we can set use_colnames=True to

4

convert these integer values into the respective item names:

5

In [ ]:

1

​

2

freq_items = apriori(df, min_support = 0.5, use_colnames = True)

3

print(freq_items)

4

​

In [ ]:

1

​

2

support itemsets

3

0 0.8 (bread)

4

1 0.6 (diaper)

5

2 0.8 (milk)

6

3 0.6 (bread, milk)

7

4 0.6 (diaper, milk)

8

​

In [ ]:

1

​

2

Generate the association rules Generate association rules that have a support value of at least 5%

3

​

In [ ]:

1

​

2

rules = association_rules(freq_items, metric ='support', min_threshold=0.05)

3

#rules = rules.sort_values(['support', 'confidence'], ascending =[False,False])

4

print(rules)

5

In [ ]:

6

antecedents consequents antecedent support consequent support support

7

\

8

0 (bread) (milk) 0.8 0.8 0.6

9

1 (milk) (bread) 0.8 0.8 0.6

10

2 (diaper) (milk) 0.6 0.8 0.6

11

3 (milk) (diaper) 0.8 0.6 0.6

12

 confidence lift leverage conviction 

13

0 0.75 0.9375 -0.04 0.8 

14

1 0.75 0.9375 -0.04 0.8 

15

2 1.00 1.2500 0.12 inf 

16

3 0.75 1.2500 0.12 1.6

In [ ]:

1

​

z

z