Naive-Bayes-Hotel-Price-Classement-Estimation/model

import pandas as pd

data = pd.read_csv('data/hotel_preprocess_v2.csv') data

Nameobject

Capsule Homestay0.9%

OYO 469 Bumi Banyu Bening Syariah0.9%

1151 others98.1%

Ratingfloat64

2.0 - 4.9

Hotel Maxim

3.7

Grandboutique-Inn

4.0

Amaris Hotel Pancoran

4.3

Jambrut Inn Hotel

4.1

BI Executive Hotel

4.5

d'primahotel WTC Mangga Dua

4.3

U Stay Hotel Asem Baris Jakarta

4.4

V Hotel Tebet

4.3

OYO 148 Cempaka Place Homestay

4.0

Sun Rise House

4.3

data.describe()

Ratingfloat64

Price_IDRfloat64

count

1271.0

mean

4.270731707317074

462930.98111723055

std

0.29543937586711777

543601.8297884194

min

2.0

54401.0

25%

4.1

201057.0

50%

4.3

311850.0

75%

4.5

503834.0

max

4.9

7152628.0

data_fill = data[['Area', 'AirConditioner', 'Bar', 'Pool', 'Spa', 'FitenessCenter', 'Restaurant', 'HotTub', 'Price_IDR', 'hotelClass']]

data2 = data_fill[['Price_IDR', 'hotelClass']] data2

Price_IDRint64

54401 - 7152628

hotelClassint64

2 - 5

113133

972

219843

971

324589

970

340632

969

171027

968

189022

967

260411

973

282534

966

315410

964

547646

df = data2.loc[data2['hotelClass'] == 2] rslt_df = df.sort_values(by=['Price_IDR']) # remove rows using the drop() function data_fill.drop(data_fill[(data_fill['Price_IDR'] > 580_000) & (data_fill['hotelClass'] == 3)].index, inplace=True) rslt_df

/shared-libs/python3.9/py/lib/python3.9/site-packages/pandas/core/frame.py:4308: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  return super().drop(

Price_IDRint64

54401 - 1374843

hotelClassint64

2 - 2

119

102586

103592

123

103970

723

104214

741

104545

795

104545

792

104545

699

104545

778

104545

df = data2.loc[data2['hotelClass'] == 3] rslt_df = df.sort_values(by=['Price_IDR']) threshold_price3 = (rslt_df.values[121])[0] # remove rows using the drop() function data_fill.drop(data_fill[(data_fill['Price_IDR'] < threshold_price3) & (data_fill['hotelClass'] == 3)].index, inplace=True) print(f' Kuartil 1: {threshold_price3}') data_fill

 Kuartil 1: 235564
/shared-libs/python3.9/py/lib/python3.9/site-packages/pandas/core/frame.py:4308: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  return super().drop(

Areaobject

Jakarta38.3%

Bandung24.5%

2 others37.2%

AirConditionerbool

True91.3%

False8.7%

173

Jakarta

True

174

Jakarta

True

175

Jakarta

True

176

Jakarta

True

178

Jakarta

True

179

Jakarta

True

181

Jakarta

True

182

Jakarta

True

183

Jakarta

True

184

Jakarta

True

df = data2.loc[data2['hotelClass'] == 4] rslt_df = df.sort_values(by=['Price_IDR']) threshold_price4 = (rslt_df.values[54])[0] # remove rows using the drop() function data_fill.drop(data_fill[(data_fill['Price_IDR'] < threshold_price4) & (data_fill['hotelClass'] == 4)].index, inplace=True) print(f' Kuartil 1: {threshold_price4}') data_fill

 Kuartil 1: 455678
/shared-libs/python3.9/py/lib/python3.9/site-packages/pandas/core/frame.py:4308: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  return super().drop(

Areaobject

Jakarta38.5%

Bandung24.2%

2 others37.3%

AirConditionerbool

True90.9%

False9.1%

223

Jakarta

True

224

Jakarta

True

226

Jakarta

True

227

Jakarta

True

228

Jakarta

True

229

Jakarta

True

230

Jakarta

True

231

Jakarta

True

232

Jakarta

True

233

Jakarta

True

df = data2.loc[data2['hotelClass'] == 5] rslt_df = df.sort_values(by=['Price_IDR']) threshold_price5 = (rslt_df.values[11])[0] # remove rows using the drop() function data_fill.drop(data_fill[(data_fill['Price_IDR'] < threshold_price5) & (data_fill['hotelClass'] == 5)].index, inplace=True) # filter data outlier # print(f' Kuartil 1: {threshold_price5[0]}') data_fill

/shared-libs/python3.9/py/lib/python3.9/site-packages/pandas/core/frame.py:4308: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  return super().drop(

Areaobject

Jakarta38.5%

Bandung24.3%

2 others37.2%

AirConditionerbool

True90.9%

False9.1%

Jakarta

True

Jakarta

True

Jakarta

True

Jakarta

False

Jakarta

True

Jakarta

True

Jakarta

True

Jakarta

True

Jakarta

True

Jakarta

True

data_fill

Areaobject

Jakarta38.5%

Bandung24.3%

2 others37.2%

AirConditionerbool

True90.9%

False9.1%

1267

Bandung

True

1268

Bandung

True

1269

Bandung

True

1270

Bandung

True

data_list = data_fill.values # dari pandas dataframe ke numpy array data_list[:10]

# Area # 0 = Jakarta, 1 = Surabaya, 2 = Jogjakarta, 3 = Bandung dom = { 'Jakarta': 0, 'Surabaya': 1, 'Jogjakarta': 2, 'Bandung': 3, } for data in data_list: data[0] = dom[data[0]] data_list

# Air conditioner # 0 = False, 1 = True for data in data_list: data[1] = 0 if not data[1] else 1 data_list

# Bar # 0 = False, 1 = True for data in data_list: data[2] = 0 if not data[2] else 1 data_list

# Pool # 0 = False, 1 = True for data in data_list: data[3] = 0 if not data[3] else 1 data_list

# Spa # 0 = False, 1 = True for data in data_list: data[4] = 0 if not data[4] else 1 data_list

# FitenessCenter # 0 = False, 1 = True for data in data_list: data[5] = 0 if not data[5] else 1 data_list

# Restaurant # 0 = False, 1 = True for data in data_list: data[6] = 0 if not data[6] else 1 data_list

#Hot Tub # 0 = False, 1 = True for data in data_list: data[7] = 0 if not data[7] else 1 data_list

# Price IDR for i in data_list: if i[8] < 200_000: i[8]= 'Kelas Harga 1 : < 200_000' elif 200_000 <= i[8] < 450_000: i[8]= 'Kelas Harga 2 : 200.000 - 450.000' elif 450_000 <= i[8] < 700_000: i[8]= 'Kelas Harga 3 : 450.000 - 700.000' else: i[8]= 'Kelas Harga 4 : >= 700.000' data_list[:10]

#Hotel Class dom = { 2: 0, 3: 1, 4: 2, 5: 3, } for data in data_list: data[9] = dom[data[9]] data_list

x = data_list[:, [0, 1, 2, 3, 4, 5, 6, 7, 9]] y = data_list[:, -2]

len(x), len(y)

from sklearn.model_selection import train_test_split X_train, X_val, y_train, y_val = train_test_split(x, y, test_size=0.25)

data_fill

Areaobject

Jakarta38.5%

Bandung24.3%

2 others37.2%

AirConditionerbool

True90.9%

False9.1%

Jakarta

True

Jakarta

True

Jakarta

True

Jakarta

False

Jakarta

True

Jakarta

True

Jakarta

True

Jakarta

True

Jakarta

True

Jakarta

True

X_train.dtype, y_train.dtype # print(type(X_train))

from sklearn.naive_bayes import CategoricalNB clf = CategoricalNB() clf.fit(X_train, y_train)

clf.score(X_val, y_val)

X_val, y_val

pred_data = clf.predict(X_val) pred_data

from sklearn.metrics import confusion_matrix, classification_report from seaborn import heatmap cm = confusion_matrix(pred_data, y_val) heatmap(cm, annot=True)

print(classification_report(pred_data, y_val))

                                   precision    recall  f1-score   support

        Kelas Harga 1 : < 200_000       0.91      0.73      0.81        85
Kelas Harga 2 : 200.000 - 450.000       0.71      0.78      0.75        96
Kelas Harga 3 : 450.000 - 700.000       0.36      0.54      0.43        35
       Kelas Harga 4 : >= 700.000       0.80      0.62      0.70        45

                         accuracy                           0.70       261
                        macro avg       0.70      0.67      0.67       261
                     weighted avg       0.75      0.70      0.72       261

import numpy as np data_pred = np.array([[2, 1, 1, 1, 1, 1, 1, 1, 3]]) print(data_pred) clf.predict(data_pred)

[[2 1 1 1 1 1 1 1 3]]