Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

Startup Success Prediction

In [6]: #importing the Dependinces
import pandas as pd
import numpy as np
import matplotlib as mpl
from matplotlib import pyplot as plt
import seaborn as sns
from datetime import date
from scipy import stats
from scipy.stats import norm, skew #for some statistics

import plotly.express as px
import plotly.graph_objects as go
import plotly.figure_factory as ff
from plotly.colors import n_colors
from plotly.subplots import make_subplots

import warnings
warnings.filterwarnings("ignore")
%matplotlib inline
pd.set_option("display.max_columns",None)
pd.set_option("display.max_rows",None)
plt.style.use('seaborn')

from collections import Counter

import datetime
import wordcloud
import json
import pandas as pd
import geopandas as gpd
import matplotlib.pyplot as plt
import shapefile as shp
import sys

In [7]: #Loading the dataset in pandas dataframe

data = pd.read_csv('/content/startup data.csv')

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 1 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [8]: #check first five rows of the dataset

data.head()

Out[8]: Unnamed: Unnamed:

state_code latitude longitude zip_code id city
0 6

San
0 1005 CA 42.358880 -71.056820 92101 c:6669 NaN
Diego

Los
1 204 CA 37.238916 -121.973718 95032 c:16283 NaN
Gatos

San San Diego

2 1001 CA 32.901049 -117.192656 92121 c:65620
Diego CA 92121

Cupertino
3 738 CA 37.320309 -122.050040 95014 c:42668 Cupertino
CA 95014

San
San
4 1002 CA 37.779281 -122.419236 94105 c:65806 Francisco
Francisco
CA 94105

In [9]: #check last five rows of the dataset

data.tail()

Out[9]: Unnamed: Unnamed:

state_code latitude longitude zip_code id city
0

San
918 352 CA 37.740594 -122.376471 94107 c:21343 NaN
Francisco

Burlington
919 721 MA 42.504817 -71.195611 1803 c:41747 Burlington
MA 1803

920 557 CA 37.408261 -122.015920 94089 c:31549 Sunnyvale NaN

San
921 589 CA 37.556732 -122.288378 94404 c:33198 NaN
Francisco

Santa
Santa
922 462 CA 37.386778 -121.966277 95054 c:26702 Clara CA
Clara
95054

In [10]: #check shape of the dataset

data.shape

Out[10]: (923, 49)

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 2 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [11]: #check columns of the dataset

data.columns

Out[11]: Index(['Unnamed: 0', 'state_code', 'latitude', 'longitude', 'zip_c

ode', 'id',
'city', 'Unnamed: 6', 'name', 'labels', 'founded_at', 'clos
ed_at',
'first_funding_at', 'last_funding_at', 'age_first_funding_y
ear',
'age_last_funding_year', 'age_first_milestone_year',
'age_last_milestone_year', 'relationships', 'funding_rounds
',
'funding_total_usd', 'milestones', 'state_code.1', 'is_CA',
'is_NY',
'is_MA', 'is_TX', 'is_otherstate', 'category_code', 'is_sof
tware',
'is_web', 'is_mobile', 'is_enterprise', 'is_advertising',
'is_gamesvideo', 'is_ecommerce', 'is_biotech', 'is_consulti
ng',
'is_othercategory', 'object_id', 'has_VC', 'has_angel', 'ha
s_roundA',
'has_roundB', 'has_roundC', 'has_roundD', 'avg_participants
',
'is_top500', 'status'],
dtype='object')

In [12]: #check more infomation of the dataset

data.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 923 entries, 0 to 922
Data columns (total 49 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 Unnamed: 0 923 non-null int64
1 state_code 923 non-null object
2 latitude 923 non-null float64
3 longitude 923 non-null float64
4 zip_code 923 non-null object
5 id 923 non-null object
6 city 923 non-null object
7 Unnamed: 6 430 non-null object
8 name 923 non-null object
9 labels 923 non-null int64
10 founded_at 923 non-null object
11 closed_at 335 non-null object
12 first_funding_at 923 non-null object
13 last_funding_at 923 non-null object
14 age_first_funding_year 923 non-null float64
15 age_last_funding_year 923 non-null float64
16 age_first_milestone_year 771 non-null float64
17 age_last_milestone_year 771 non-null float64
18 relationships 923 non-null int64

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 3 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

19 funding_rounds 923 non-null int64

20 funding_total_usd 923 non-null int64
21 milestones 923 non-null int64
22 state_code.1 922 non-null object
23 is_CA 923 non-null int64
24 is_NY 923 non-null int64
25 is_MA 923 non-null int64
26 is_TX 923 non-null int64
27 is_otherstate 923 non-null int64
28 category_code 923 non-null object
29 is_software 923 non-null int64
30 is_web 923 non-null int64
31 is_mobile 923 non-null int64
32 is_enterprise 923 non-null int64
33 is_advertising 923 non-null int64
34 is_gamesvideo 923 non-null int64
35 is_ecommerce 923 non-null int64
36 is_biotech 923 non-null int64
37 is_consulting 923 non-null int64
38 is_othercategory 923 non-null int64
39 object_id 923 non-null object
40 has_VC 923 non-null int64
41 has_angel 923 non-null int64
42 has_roundA 923 non-null int64
43 has_roundB 923 non-null int64
44 has_roundC 923 non-null int64
45 has_roundD 923 non-null int64
46 avg_participants 923 non-null float64
47 is_top500 923 non-null int64
48 status 923 non-null object
dtypes: float64(7), int64(28), object(14)
memory usage: 353.5+ KB

In [13]: #check mathamtic describe

data.describe()

Out[13]: Unnamed: 0 latitude longitude labels age_first_funding_year age_last_fundin

count 923.000000 923.000000 923.000000 923.000000 923.000000

mean 572.297941 38.517442 -103.539212 0.646804 2.235630

std 333.585431 3.741497 22.394167 0.478222 2.510449

min 1.000000 25.752358 -122.756956 0.000000 -9.046600

25% 283.500000 37.388869 -122.198732 0.000000 0.576700

50% 577.000000 37.779281 -118.374037 1.000000 1.446600

75% 866.500000 40.730646 -77.214731 1.000000 3.575350

max 1153.000000 59.335232 18.057121 1.000000 21.895900

In [14]:
http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 4 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [14]:
#check corr relation of the dataset
data.corr()

Out[14]:
Unnamed: latitude longitude labels age_first_funding_year
0

Unnamed: 0 1.000000 0.054726 0.023292 -0.068721 -0.004507

latitude 0.054726 1.000000 0.368475 0.046560 -0.046868

longitude 0.023292 0.368475 1.000000 -0.036092 -0.014158

labels -0.068721 0.046560 -0.036092 1.000000 -0.075637

age_first_funding_year -0.004507 -0.046868 -0.014158 -0.075637 1.000000

age_last_funding_year -0.116533 -0.041692 -0.000148 0.073731 0.762382

age_first_milestone_year -0.135614 -0.072000 -0.051674 0.162279 0.593526

age_last_milestone_year -0.131698 -0.054275 -0.087701 0.265871 0.472029

relationships -0.079950 -0.039198 -0.073197 0.360434 -0.187817

funding_rounds -0.118456 -0.000659 0.022447 0.206049 -0.155478

funding_total_usd -0.064169 -0.072941 0.017970 0.040176 0.046350

milestones -0.000338 0.017708 -0.016420 0.328260 -0.295894

is_CA -0.042446 -0.417471 -0.780122 0.077217 -0.010800

is_NY 0.033485 0.205747 0.449871 0.059996 -0.128102

is_MA 0.043021 0.318015 0.441031 0.081735 0.020279

is_TX -0.021463 -0.423888 0.066199 -0.045309 0.032838

is_otherstate 0.002249 0.338590 0.257801 -0.169067 0.081031

is_software 0.001367 -0.001656 0.024857 0.012429 0.116797

is_web 0.007076 -0.009799 -0.022024 -0.000873 -0.166601

is_mobile -0.028279 0.035917 0.013527 0.007312 -0.054658

is_enterprise 0.042640 -0.002291 -0.003244 0.073772 -0.047326

is_advertising -0.075131 0.054575 0.039998 0.044355 -0.071336

is_gamesvideo 0.065020 -0.033160 -0.025569 -0.025893 -0.063787

is_ecommerce -0.026132 0.041628 0.043092 -0.072193 -0.071580

is_biotech 0.004224 0.012956 0.028075 0.000104 0.190653

is_consulting -0.040929 -0.033905 0.021244 0.002373 -0.012596

is_othercategory 0.006243 -0.039656 -0.046560 -0.042408 0.115649

has_VC -0.040057 0.031045 0.024852 -0.056515 0.168140

has_angel 0.134044 0.028891 0.102001 -0.072840 -0.345985

has_roundA -0.076568 -0.033072 -0.066288 0.184307 -0.293081

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 5 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

has_roundB -0.135289 -0.011801 -0.067017 0.208257 -0.060532

has_roundC -0.090922 -0.057762 -0.042309 0.165902 0.033388

has_roundD -0.081123 -0.018825 -0.042854 0.139940 0.121338

avg_participants 0.026713 -0.018176 -0.045191 0.185992 0.114363

is_top500 0.026019 0.032675 -0.091913 0.310652 0.050638

In [15]: #check missing value of the dataset

data.isnull().sum()

Out[15]: Unnamed: 0 0
state_code 0
latitude 0
longitude 0
zip_code 0
id 0
city 0
Unnamed: 6 493
name 0
labels 0
founded_at 0
closed_at 588
first_funding_at 0
last_funding_at 0
age_first_funding_year 0
age_last_funding_year 0
age_first_milestone_year 152
age_last_milestone_year 152
relationships 0
funding_rounds 0
funding_total_usd 0
milestones 0
state_code.1 1
is_CA 0
is_NY 0
is_MA 0
is_TX 0
is_otherstate 0
category_code 0
is_software 0
is_web 0
is_mobile 0
is_enterprise 0
is_advertising 0
is_gamesvideo 0
is_ecommerce 0
is_biotech 0
is_consulting 0
is_othercategory 0
object_id 0

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 6 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

has_VC 0
has_angel 0
has_roundA 0
has_roundB 0
has_roundC 0
has_roundD 0
avg_participants 0
is_top500 0
status 0
dtype: int64

In [16]: # Total Missing Values kolom "Unnamed: 6"

totalNull = data['Unnamed: 6'].isnull().sum()

print('Total Missing Values Kolom "Unnamed: 6": ', totalNull)

Total Missing Values Kolom "Unnamed: 6": 493

In [17]: #Handling Missing Value closed_at

data['closed_at'] = data['closed_at'].fillna(value="31/12/2013")

In [18]: totalNull = data['closed_at'].isnull().sum()

print('Total Missing Values Kolom "closed_at": ', totalNull)

Total Missing Values Kolom "closed_at": 0

In [19]: #Handling Missing Value age_first_milestone_year and age_last_milestone

data[['age_first_milestone_year','age_last_milestone_year','milestones'

Out[19]: age_first_milestone_year age_last_milestone_year milestones

0 4.6685 6.7041 3

1 7.0055 7.0055 1

2 1.4575 2.2055 2

3 6.0027 6.0027 1

4 0.0384 0.0384 1

In [20]: data['age_first_milestone_year'] = data['age_first_milestone_year'].

data['age_last_milestone_year'] = data['age_last_milestone_year'].filln

In [21]: #Handling Missing Value state_code.1

for index, row in data.iterrows():
if row['state_code']!=row['state_code.1']:
print(index, row['state_code'], row['state_code.1'])

515 CA nan

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 7 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [22]: data.drop(["state_code.1"], axis=1, inplace=True)

In [23]: null=pd.DataFrame(data.isnull().sum(),columns=["Null Values"])

null["% Missing Values"]=(data.isna().sum()/len(data)*100)
null = null[null["% Missing Values"] > 0]
null.style.background_gradient(cmap='viridis',low =0.2,high=0.1)

Out[23]: Null Values % Missing Values

Unnamed: 6 493 53.412784

Graphic Approach
In [24]: #Correlation heatmap
data['age_first_milestone_year'] = data.age_first_milestone_year.astype
data['age_last_milestone_year'] = data.age_last_milestone_year.astype

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 8 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [25]: features = ['age_first_funding_year','age_last_funding_year','age_first

plt.figure(figsize=(30,20))
ax = sns.heatmap(data = data[features].corr(),cmap='YlGnBu',annot=True

bottom, top = ax.get_ylim()

ax.set_ylim(bottom + 0.5,top - 0.5)

Out[25]: (31.5, -0.5)

Scatter plot

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 9 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [26]: fig, ax = plt.subplots()

_ = plt.scatter(x=data['age_first_funding_year'], y=data['age_last_fund
_ = ax.set(xlabel="age_first_funding_year", ylabel="age_last_funding_ye

In [27]: fig, ax = plt.subplots()

_ = plt.scatter(x=data['age_first_milestone_year'], y=data['age_last_mi
_ = ax.set(xlabel="status", ylabel="milestones")

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 10 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

Box plots
In [28]: featuresNum = ['age_first_funding_year','age_last_funding_year','age_fi

plt.figure(figsize=(15, 7))
for i in range(0, len(featuresNum)):
plt.subplot(1, len(featuresNum), i+1)
sns.boxplot(y=data[featuresNum[i]], color='green', orient='v')
plt.tight_layout()

Dataset collection founded years

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 11 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [29]: cdf = data["founded_at"].apply(lambda x: '' + x[:2]).value_counts() \

.to_frame().reset_index() \
.rename(columns={"index": "year", "founded_at": "No_of_star

fig, ax = plt.subplots()
_ = sns.barplot(x="year", y="No_of_startup", data=cdf,
palette=sns.color_palette(['#003f5c', '#ffa600'], n_col
_ = ax.set(xlabel="Year", ylabel="No. of startup")

In [30]: data["founded_at"].apply(lambda x: '20:' + x[:2]).value_counts(normaliz

Out[30]: 20:1/ 563

20:6/ 43
20:8/ 42
20:10 38
20:5/ 36
20:9/ 35
20:7/ 31
20:3/ 30
20:4/ 30
20:2/ 29
20:11 23
20:12 23
Name: founded_at, dtype: int64

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 12 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [31]: data["founded_at"].apply(lambda x: '20:' + x[:2]).value_counts(normaliz

Out[31]: 20:1/ 0.609967

20:6/ 0.046587
20:8/ 0.045504
20:10 0.041170
20:5/ 0.039003
20:9/ 0.037920
20:7/ 0.033586
20:3/ 0.032503
20:4/ 0.032503
20:2/ 0.031419
20:11 0.024919
20:12 0.024919
Name: founded_at, dtype: float64

In [32]: data["closed_at"].apply(lambda x: '20:' + x[:2]).value_counts(normalize

Out[32]: 20:31 0.637053

20:1/ 0.069339
20:6/ 0.041170
20:7/ 0.037920
20:2/ 0.035753
20:5/ 0.033586
20:8/ 0.027086
20:10 0.020585
20:3/ 0.020585
20:11 0.020585
20:4/ 0.019502
20:12 0.018418
20:9/ 0.018418
Name: closed_at, dtype: float64

How many Startup are acquired or closed

have?
In [33]: df_acquired = data[(data["status"] == True)]
df_acquired.shape

Out[33]: (0, 48)

In [34]: df_closed = data[(data["status"] == False)]

df_closed.shape

Out[34]: (0, 48)

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 13 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [35]: sns.countplot(data['status'])

Out[35]: <matplotlib.axes._subplots.AxesSubplot at 0x7f88cc80cb50>

Which category has the largest number of

startup

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 14 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [36]: fig, ax = plt.subplots(figsize=(12,8))

_ = sns.countplot(x="category_code", hue="status", data=data, palette

order=data.category_code.value_counts().index)

_ = ax.set_xticklabels(ax.get_xticklabels(), rotation=90)
_ = ax.set(xlabel="Category", ylabel="No. of startup")
plt.legend(bbox_to_anchor=(0.945, 0.90))

Out[36]: <matplotlib.legend.Legend at 0x7f88cc7480d0>

Which category having most number of total

funding

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 15 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [37]: funding_sorted_category = pd.pivot_table(data,

index=['category_code'],
values=['funding_total_usd'],
aggfunc=['sum']
).reset_index()
funding_sorted_category.columns = ['category_code', 'funding_total_usd'
funding_sorted_category = funding_sorted_category.sort_values(['funding
funding_sorted_category.head(10)

Out[37]: category_code funding_total_usd

18 mobile 7263750881

30 software 2657598865

34 web 1729035436

3 biotech 1723699484

8 enterprise 1338882096

4 cleantech 1300284730

28 semiconductor 1105156970

0 advertising 918619012

11 games_video 844643530

12 hardware 773938873

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 16 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [38]: fig, ax = plt.subplots(figsize=(15,7))

_ = sns.barplot(x="category_code", y="funding_total_usd", data=funding_
palette="nipy_spectral", ax=ax)
_ = ax.set_xticklabels(ax.get_xticklabels(), rotation=90)
_ = ax.set(xlabel="Category", ylabel="Total Funding USD")

Which State having most number of

Startup

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 17 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [39]: fig, ax = plt.subplots(figsize=(12,8))

_ = sns.countplot(x="state_code", hue="status", data=data, palette=

order=data.state_code.value_counts().index)

_ = ax.set_xticklabels(ax.get_xticklabels(), rotation=90)
_ = ax.set(xlabel="state_code", ylabel="No. of startup")
plt.legend(bbox_to_anchor=(0.945, 0.90))

Out[39]: <matplotlib.legend.Legend at 0x7f88cc3f10d0>

In [40]: trending_statea = data.groupby(['state_code']).size().rename('num_start

most_trending_statea = trending_statea[trending_statea.groupby('state_c
most_trending_statea = most_trending_statea.sort_values('num_startup'
most_trending_statea

Out[40]: state_code num_startup

2 CA 488

23 NY 106

12 MA 83

32 WA 42

29 TX 42

3 CO 19

9 IL 18

26 PA 17

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 18 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

31 VA 13

7 GA 11

20 NJ 7

13 MD 7

25 OR 7

18 NC 7

6 FL 6

24 OH 6

16 MN 5

5 DC 4

4 CT 4

15 MI 3

28 TN 3

27 RI 3

30 UT 3

22 NV 2

19 NH 2

1 AZ 2

14 ME 2

11 KY 2

10 IN 2

17 MO 2

33 WI 1

0 AR 1

21 NM 1

8 ID 1

34 WV 1

Which State having most number of acquired

Startup per category

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 19 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [41]: trending_statea = df_acquired.groupby(['state_code','category_code']).

most_trending_statea = trending_statea[trending_statea.groupby('state_c
most_trending_statea = most_trending_statea.sort_values('num_startup'
most_trending_statea.head(10)

Out[41]: state_code category_code num_startup

Which State having most number of closed

Startup per category
In [42]: trending_statec = df_closed.groupby(['state_code','category_code']).

most_trending_statec = trending_statec[trending_statec.groupby('state_c
most_trending_statec = most_trending_statec.sort_values('num_startup'
most_trending_statec

Out[42]: state_code category_code num_startup

Which city having most number of acquired

Startup per category
In [43]: trending_categorya = df_acquired.groupby(['city','category_code']).

most_trending_categorya = trending_categorya[trending_categorya.groupby
most_trending_categorya = most_trending_categorya.sort_values('num_star
most_trending_categorya

Out[43]: city category_code num_startup

Which city having most number of closed

Startup per category

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 20 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [44]: trending_categoryc = df_closed.groupby(['city','category_code']).size

most_trending_categoryc = trending_categoryc[trending_categoryc.groupby
most_trending_categoryc = most_trending_categoryc.sort_values('num_star
most_trending_categoryc

Out[44]: index city category_code num_startup

Which city having most number of total

funding
In [45]: funding_sorted_city = pd.pivot_table(data,
index=['city'],
values=['funding_total_usd'],
aggfunc=['sum']
).reset_index()
funding_sorted_city.columns = ['city', 'funding_total_usd']
funding_sorted_city = funding_sorted_city.sort_values(['funding_total_u
funding_sorted_city = funding_sorted_city.head(10)
funding_sorted_city

Out[45]: city funding_total_usd

91 Kirkland 5718914576

174 San Francisco 1673487129

135 New York 1231405734

125 Mountain View 985553322

181 Santa Clara 839050274

176 San Jose 733181780

13 Austin 706317317

173 San Diego 614475001

163 Redwood City 521330100

110 Loveland 510000000

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 21 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [46]: fig, ax = plt.subplots(figsize=(10,7))

_ = sns.barplot(x="city", y="funding_total_usd", data=funding_sorted_ci
palette="nipy_spectral", ax=ax)
_ = ax.set_xticklabels(ax.get_xticklabels(), rotation=90)
_ = ax.set(xlabel="No of State", ylabel="Number of Start Up")

In [47]: df_what_in_kirkland = data[(data["city"] == 'Kirkland')]

df_what_in_kirkland.shape

Out[47]: (2, 48)

In [48]: df_what_in_kirkland.head()

Out[48]: Unnamed: Unnamed:

state_code latitude longitude zip_code id city
0 6

Kirkland
98033- WA
62 332 WA 47.675489 -122.191667 c:19861 Kirkland
6314 98033-
6314

364 86 WA 30.632480 -86.984345 98033 c:13219 Kirkland NaN

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 22 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [49]: #How many Startup have has_VC?

fig, ax = plt.subplots(figsize=(10,5))

_ = sns.countplot(x="has_VC", hue="status", data=data, palette="nipy_sp

order=data.has_VC.value_counts().index)

_ = ax.set_xticklabels(ax.get_xticklabels(), rotation=90)
_ = ax.set(xlabel="Has_VC", ylabel="No. of startup")
plt.legend(bbox_to_anchor=(0.945, 0.90))

Out[49]: <matplotlib.legend.Legend at 0x7f88cc2e5c50>

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 23 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [50]: #How many Startup have is_top500?

fig, ax = plt.subplots(figsize=(10,5))

_ = sns.countplot(x="is_top500", hue="status", data=data, palette="nipy

order=data.is_top500.value_counts().index)

_ = ax.set_xticklabels(ax.get_xticklabels(), rotation=90)
_ = ax.set(xlabel="is_top500", ylabel="No. of startup")
plt.legend(bbox_to_anchor=(0.945, 0.90))

Out[50]: <matplotlib.legend.Legend at 0x7f88cc4d1490>

In [51]: #How many Startup have both 'acquired' status and is_top500?
len(data[(data["status"] == True) & (data["is_top500"] == True)].index

Out[51]: 0

In [52]: #How many Startup have both 'closed' status and is_top500?
len(data[(data["status"] == False) & (data["is_top500"] == False)].

Out[52]: 0

In [53]: df_acquired["is_top500"].value_counts(normalize=True)

Out[53]: Series([], Name: is_top500, dtype: float64)

In [54]: #How many years on average the company closes

df_closed.founded_at=pd.to_datetime(df_closed.founded_at)
df_closed.closed_at=pd.to_datetime(df_closed.closed_at)

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 24 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [55]: #which relationship related to acquired or closed startup?

fig, ax = plt.subplots(figsize=(17,10))

sns.countplot(x="relationships", hue="status", data=data, palette="nipy

order=data.relationships.value_counts().index)
plt.legend(bbox_to_anchor=(0.945, 0.90))

Out[55]: <matplotlib.legend.Legend at 0x7f88cc86f3d0>

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 25 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [56]: #which funding_rounds related to acquired or closed startup?

fig, ax = plt.subplots(figsize=(12,8))

sns.countplot(x="funding_rounds", hue="status", data=data, palette=

order=data.funding_rounds.value_counts().index)

Out[56]: <matplotlib.axes._subplots.AxesSubplot at 0x7f88ccbf37d0>

Mapping area startup

In [57]: 'geopandas' in sys.modules

Out[57]: True

In [58]: new_data = gpd.GeoDataFrame(data, geometry=gpd.points_from_xy(data.

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 26 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [59]: new_data.head()

Out[59]: Unnamed: Unnamed:

state_code latitude longitude zip_code id city
0 6

San
0 1005 CA 42.358880 -71.056820 92101 c:6669 NaN
Diego

Los
1 204 CA 37.238916 -121.973718 95032 c:16283 NaN
Gatos

San San Diego

2 1001 CA 32.901049 -117.192656 92121 c:65620
Diego CA 92121

Cupertino
3 738 CA 37.320309 -122.050040 95014 c:42668 Cupertino
CA 95014

San
San
4 1002 CA 37.779281 -122.419236 94105 c:65806 Francisco
Francisco
CA 94105

In [60]: age=["age_first_funding_year","age_last_funding_year","age_first_milest

for a in range(len(age)):
print("Is there any negative value in '{}' column : {} ".format

Is there any negative value in 'age_first_funding_year' column :

True
Is there any negative value in 'age_last_funding_year' column : T
rue
Is there any negative value in 'age_first_milestone_year' column
: True
Is there any negative value in 'age_last_milestone_year' column :
True

In [61]: df=data.drop(data[data.age_first_funding_year<0].index)
df=data.drop(data[data.age_last_funding_year<0].index)
df=data.drop(df[data.age_first_milestone_year<0].index)
df=data.drop(data[data.age_last_milestone_year<0].index)

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 27 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [62]: for a in range(len(age)):

print("Is there any negative value in '{}' column : {} ".format

Is there any negative value in 'age_first_funding_year' column :

True
Is there any negative value in 'age_last_funding_year' column : T
rue
Is there any negative value in 'age_first_milestone_year' column
: True
Is there any negative value in 'age_last_milestone_year' column :
False

Outliers
In [63]: featuresNumfinal = ['age_first_funding_year','age_last_funding_year'

plt.figure(figsize=(15, 7))
for i in range(0, len(featuresNumfinal)):
plt.subplot(1, len(featuresNumfinal), i+1)
sns.boxplot(y=df[featuresNumfinal[i]], color='green', orient='v'
plt.tight_layout()

In [64]: #Log-transformation of the funding and milestone year variable

df["age_first_funding_year"] = np.log1p(df["age_first_funding_year"])
df["age_last_funding_year"] = np.log1p(df["age_last_funding_year"])
df["age_first_milestone_year"] = np.log1p(df["age_first_milestone_year"
df["age_last_milestone_year"] = np.log1p(df["age_last_milestone_year"
df["funding_total_usd"] = np.log1p(df["funding_total_usd"])

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 28 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [65]: featuresNumfinal = ['age_first_funding_year','age_last_funding_year'

plt.figure(figsize=(15, 7))
for i in range(0, len(featuresNumfinal)):
plt.subplot(1, len(featuresNumfinal), i+1)
sns.boxplot(y=df[featuresNumfinal[i]], color='green', orient='v'
plt.tight_layout()

Feature Engineering
In [66]: #New Column has_RoundABCD
df['has_RoundABCD'] = np.where((df['has_roundA'] == 1) | (df['has_round
df.head()

Out[66]: Unnamed: Unnamed:

state_code latitude longitude zip_code id city
0 6

San
0 1005 CA 42.358880 -71.056820 92101 c:6669 NaN
Diego

Los
1 204 CA 37.238916 -121.973718 95032 c:16283 NaN
Gatos

San San Diego

2 1001 CA 32.901049 -117.192656 92121 c:65620
Diego CA 92121

Cupertino
3 738 CA 37.320309 -122.050040 95014 c:42668 Cupertino
CA 95014

San
San
4 1002 CA 37.779281 -122.419236 94105 c:65806 Francisco
Francisco
CA 94105

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 29 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [67]: #New Column "has_Investor

df['has_Investor'] = np.where((df['has_VC'] == 1) | (df['has_angel']
df.head()

Out[67]: Unnamed: Unnamed:

state_code latitude longitude zip_code id city
0 6

San
0 1005 CA 42.358880 -71.056820 92101 c:6669 NaN
Diego

Los
1 204 CA 37.238916 -121.973718 95032 c:16283 NaN
Gatos

San San Diego

2 1001 CA 32.901049 -117.192656 92121 c:65620
Diego CA 92121

Cupertino
3 738 CA 37.320309 -122.050040 95014 c:42668 Cupertino
CA 95014

San
San
4 1002 CA 37.779281 -122.419236 94105 c:65806 Francisco
Francisco
CA 94105

In [68]: len(df[(df["has_RoundABCD"] == 1)].index)

Out[68]: 674

In [69]: len(df[ (df['has_RoundABCD'] == 1) & (df['status'] == 1) ].index)

len(df)

Out[69]: 911

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 30 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [70]: #New Column "has_Seed"

df['has_Seed'] = np.where((df['has_RoundABCD'] == 0) & (df['has_Investo
df.head()

Out[70]: Unnamed: Unnamed:

state_code latitude longitude zip_code id city
0 6

San
0 1005 CA 42.358880 -71.056820 92101 c:6669 NaN
Diego

Los
1 204 CA 37.238916 -121.973718 95032 c:16283 NaN
Gatos

San San Diego

2 1001 CA 32.901049 -117.192656 92121 c:65620
Diego CA 92121

Cupertino
3 738 CA 37.320309 -122.050040 95014 c:42668 Cupertino
CA 95014

San
San
4 1002 CA 37.779281 -122.419236 94105 c:65806 Francisco
Francisco
CA 94105

In [71]: df['has_Seed'] == 1

Out[71]: 0 True
1 False
2 False
3 False
4 True
5 False
6 False
7 False
8 False
9 False
10 False
11 False
12 True
13 False
14 False
15 True
16 False
17 False
18 False
19 False

Model belding

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 31 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [72]: #Cek categorical

cat_feature = df.select_dtypes(include='object')
cat_feature.head()

Out[72]: Unnamed:
state_code zip_code id city name founded_at closed_at
6

San
0 CA 92101 c:6669 NaN Bandsintown 1/1/2007 31/12/2013
Diego

Los
1 CA 95032 c:16283 NaN TriCipher 1/1/2000 31/12/2013
Gatos

San San Diego

2 CA 92121 c:65620 Plixi 3/18/2009 31/12/2013
Diego CA 92121

Cupertino Solidcore
3 CA 95014 c:42668 Cupertino 1/1/2002 31/12/2013
CA 95014 Systems

San
San
4 CA 94105 c:65806 Francisco Inhale Digital 8/1/2010 10/1/2012
Francisco
CA 94105

In [73]: df = data.drop(['state_code'],axis=1)
df = df.drop(['id'],axis=1)
df = df.drop(['Unnamed: 6'],axis=1)
df = df.drop(['category_code'],axis=1)
df = df.drop(['object_id'],axis=1)
df = df.drop(['zip_code'],axis=1)
df = df.drop(['founded_at'],axis=1)
df = df.drop(['closed_at'],axis=1)
df = df.drop(['first_funding_at'],axis=1)
df = df.drop(['last_funding_at'],axis=1)
df = df.drop(['city'],axis=1)
df = df.drop(['name'],axis=1)
df = df.drop(['Unnamed: 0'],axis=1)
df = df.drop(['latitude','longitude'],axis=1)
df = df.drop(['geometry'],axis=1)
#df = df.drop(['age_closed_startup'],axis=1)
df = df.drop(['relationships'],axis=1)

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 32 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [74]: df.columns

Out[74]: Index(['labels', 'age_first_funding_year', 'age_last_funding_year'

,
'age_first_milestone_year', 'age_last_milestone_year', 'fun
ding_rounds',
'funding_total_usd', 'milestones', 'is_CA', 'is_NY', 'is_MA
', 'is_TX',
'is_otherstate', 'is_software', 'is_web', 'is_mobile', 'is_
enterprise',
'is_advertising', 'is_gamesvideo', 'is_ecommerce', 'is_biot
ech',
'is_consulting', 'is_othercategory', 'has_VC', 'has_angel',
'has_roundA', 'has_roundB', 'has_roundC', 'has_roundD',
'avg_participants', 'is_top500', 'status'],
dtype='object')

In [75]: #del data['Unnamed: 6']

del data['Unnamed: 0']

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 33 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [76]: df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 923 entries, 0 to 922
Data columns (total 32 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 labels 923 non-null int64
1 age_first_funding_year 923 non-null float64
2 age_last_funding_year 923 non-null float64
3 age_first_milestone_year 923 non-null float64
4 age_last_milestone_year 923 non-null float64
5 funding_rounds 923 non-null int64
6 funding_total_usd 923 non-null int64
7 milestones 923 non-null int64
8 is_CA 923 non-null int64
9 is_NY 923 non-null int64
10 is_MA 923 non-null int64
11 is_TX 923 non-null int64
12 is_otherstate 923 non-null int64
13 is_software 923 non-null int64
14 is_web 923 non-null int64
15 is_mobile 923 non-null int64
16 is_enterprise 923 non-null int64
17 is_advertising 923 non-null int64
18 is_gamesvideo 923 non-null int64
19 is_ecommerce 923 non-null int64
20 is_biotech 923 non-null int64
21 is_consulting 923 non-null int64
22 is_othercategory 923 non-null int64
23 has_VC 923 non-null int64
24 has_angel 923 non-null int64
25 has_roundA 923 non-null int64
26 has_roundB 923 non-null int64
27 has_roundC 923 non-null int64
28 has_roundD 923 non-null int64
29 avg_participants 923 non-null float64
30 is_top500 923 non-null int64
31 status 923 non-null object
dtypes: float64(5), int64(26), object(1)
memory usage: 230.9+ KB

In [77]: X = df.drop('status', axis=1)

Y = df['status']

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 34 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [78]: print(X)
print(Y)

labels age_first_funding_year age_last_funding_year \

0 1 2.2493 3.0027
1 1 5.1260 9.9973
2 1 1.0329 1.0329
3 1 3.1315 5.3151
4 0 0.0000 1.6685
5 0 4.5452 4.5452
6 1 1.7205 5.2110
7 1 1.6466 6.7616
8 1 3.5863 11.1123
9 1 1.6712 4.6849
10 1 4.6274 9.4493
11 0 1.0849 5.3370
12 0 4.9041 4.9041
13 1 0.0192 2.4356
14 1 4.6658 8.9973
15 0 6.6082 6.6082
16 0 2.5863 6.7644
17 1 4.5918 7.1726
18 1 0.7425 1.5808

Type Markdown and LaTeX: 𝛼 2

In [79]: from sklearn.model_selection import train_test_split

X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size

In [81]: print(X_train.shape, X_test.shape, Y_train.shape, Y_test.shape)

(738, 31) (185, 31) (738,) (185,)

In [82]: from sklearn.preprocessing import StandardScaler

scler = StandardScaler()
X_train = scler.fit_transform(X_train)
X_test = scler.transform(X_test)

In [84]: from sklearn.linear_model import LogisticRegression

model = LogisticRegression()

In [85]: model.fit(X_train, Y_train)

Out[85]: LogisticRegression()

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 35 of 36
Untitled2 - Jupyter Notebook 10/11/22, 3:14 AM

In [87]:

#accuracy on training data

from sklearn.metrics import accuracy_score
X_train_prediction = model.predict(X_train)
training_data_accuracy = accuracy_score(X_train_prediction, Y_train)

In [88]: print('Accuracy on training data : ', training_data_accuracy)

Accuracy on training data : 1.0

In [89]: #accuracy on test data

X_test_prediction = model.predict(X_test)
test_data_accuracy = accuracy_score(X_test_prediction, Y_test)

In [90]: print('Accuracy on test data : ', test_data_accuracy)

Accuracy on test data : 1.0

http://localhost:8888/notebooks/Downloads/Untitled2.ipynb Page 36 of 36