
#-----------------------------------------------------------------------------------------------------------------------------
#------------------------------------------------[ Libraries Import ]---------------------------------------------------------
#-----------------------------------------------------------------------------------------------------------------------------


from sklearn.preprocessing import LabelEncoder
from pandas.plotting import scatter_matrix
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
from sklearn.decomposition import PCA
from adjustText import adjust_text
import seaborn as sns


#-----------------------------------------------------------------------------------------------------------------------------
#---------------------------------------------------[ Import DataBase ]-------------------------------------------------------
#-----------------------------------------------------------------------------------------------------------------------------


dataTelecom=pd.read_excel('S:\\TunisieTelecom\\TelecomData.xlsx')
dataTelecom.head(100)


selected_variables = [
    'DUREE_APPEL_TOT',
    'nb_sms_tot' ,
    'revenu_cdr_c',
    'MNT_RECH',
    'FREQ_ACT_OUT',
    'NB_JOUR_APPEL_TAXE' ,
    'NB_SMS_TAXE' ,
    'NB_RECH_SUP5',
    'MNT_TRANSFERT_OUT' ,
    'revenu_voix' ,
    'MNT_FORFAIT_DATA',
    'NB_FORFAIT_VOIX' ,
    'OFFRE',
    'VOLUME_SESSION',
    'P_revenu_data',
    'MNT_FORFAIT',
    'FREQ_USSD',
    'P_FF_Data' ,
    'Duree_onnet_tot',
    'Duree_offnet_tot' ,
]
# Select the desired variables from the dataframe
dataTelecomSelected = dataTelecom[selected_variables].copy()
dataTelecomSelected.head(10)


dataTelecomSelectedSample=dataTelecomSelected.sample(100)
dataTelecomSelectedSample.head(150)


#-----------------------------------------------------------------------------------------------------------------------------
#---------------------------------------------------[ Label Encoder ]---------------------------------------------------------
#-----------------------------------------------------------------------------------------------------------------------------


label_encoder = LabelEncoder()
encoded_offre = label_encoder.fit_transform(dataTelecomSelectedSample['OFFRE'])
dataTelecomSelectedSample['OFFRE']=encoded_offre
dataTelecomSelectedSample.head(100)


#-----------------------------------------------------------------------------------------------------------------------------
#----------------------------------------------[ PCA In Different Colors ]----------------------------------------------------
#-----------------------------------------------------------------------------------------------------------------------------


X = dataTelecomSelectedSample.iloc[:,0:20]
X


# Les 20 colonnes de mesures qui décrivent différents dataTelecomSelectedSample

X = dataTelecomSelectedSample.iloc[:,0:20]

colors = dataTelecomSelectedSample.OFFRE.astype('category')

# Les couleurs déduites de chaque espèce de dataTelecomSelectedSample

y = colors.cat.codes

print(X) ; print(y)


scatter_matrix(X,c=y) ; plt.show()

       DUREE_APPEL_TOT  nb_sms_tot  revenu_cdr_c   MNT_RECH  FREQ_ACT_OUT  \
19257        30.090072    8.036033      6.628774  21.796702      4.390850   
38166       577.035853    8.792087      8.468283   6.180827      1.926081   
79907       126.227433    2.056120      9.733714   4.687255      4.980074   
9564          0.042172    3.331580     10.712672  12.834082     13.015438   
21204        39.975826    4.348320      9.195774   3.873290      7.742715   
...                ...         ...           ...        ...           ...   
87907        10.833286   22.632752      7.434685  16.818580      1.330702   
69701       439.174973    6.930242     10.579883   3.869375      3.062471   
1612         34.254143    6.868878     14.567883   1.299701      0.257181   
43159        81.030834    1.408071      3.985414   1.817113      5.775719   
34600         2.058911    8.345012      3.530926   2.438283      7.900239   

       NB_JOUR_APPEL_TAXE  NB_SMS_TAXE  NB_RECH_SUP5  MNT_TRANSFERT_OUT  \
19257            3.572585     2.230938          0.33           6.576157   
38166            7.041788     0.025432          5.00           0.692279   
79907            8.250577     0.144793          0.00           5.995859   
9564            10.310605     6.283435          2.67           6.088506   
21204            4.286317     3.506429          0.00           0.665510   
...                   ...          ...           ...                ...   
87907           30.000000     2.258939          0.00           8.323257   
69701            4.274253     0.701469          3.67           3.708045   
1612             3.908668     2.735732          2.33           1.125163   
43159            7.573036     4.426431          0.00           4.559158   
34600            2.923418     2.510042          0.00           1.223616   

       revenu_voix  MNT_FORFAIT_DATA  NB_FORFAIT_VOIX  OFFRE  VOLUME_SESSION  \
19257     5.973346          8.614453         2.382774      3    96933.695160   
38166     3.000000          0.417457        12.181873     26    77383.665104   
79907     3.000000          3.374781         8.150515      4   186211.083505   
9564     10.611291         13.975665         7.204972     24   120468.526004   
21204     7.944436         10.043437         1.146157     16    53403.671898   
...            ...               ...              ...    ...             ...   
87907     6.718734          4.107238         1.454253     20    14172.653991   
69701     8.752228          8.264411         5.081549      9    96018.498183   
1612      3.000000          4.784903        18.786315     24    74526.004536   
43159     3.000000         11.190860         0.234017     21    70646.213006   
34600     3.000000         10.885480         0.874752      8    85687.964862   

       P_revenu_data  MNT_FORFAIT   FREQ_USSD  P_FF_Data  Duree_onnet_tot  \
19257       0.596794    10.756903  143.642413   0.014595        44.241075   
38166       0.158712     3.796006   57.790428   0.516101        39.859481   
79907       0.935933     1.266953   46.308896   0.645644        12.738402   
9564        0.154198     1.421667   37.609093   0.708598        41.733081   
21204       0.089654     4.067660  190.000000   0.841939        74.725692   
...              ...          ...         ...        ...              ...   
87907       0.825286     1.269155   47.655680   0.445803         5.349727   
69701       0.689052     3.429793   94.129045   0.630098        71.752719   
1612        0.463658     3.038143   14.416576   0.491777         7.667571   
43159       0.939057    10.895926  174.699935   0.606876        50.623577   
34600       0.818186     4.716647   71.896790   0.599819        73.497484   

       Duree_offnet_tot  
19257         34.738155  
38166          8.263161  
79907         15.707098  
9564           3.447804  
21204         22.666880  
...                 ...  
87907          9.497032  
69701          6.520351  
1612          16.154592  
43159         58.188738  
34600          6.421968  

[100 rows x 20 columns]
19257     3
38166    26
79907     4
9564     24
21204    16
         ..
87907    20
69701     9
1612     24
43159    21
34600     8
Length: 100, dtype: int8


mypca = PCA(n_components=3)  # Here we set the number of components to keep as 3

# PCA Model

mypca.fit(X)

# Percentage of variance explained by each of the selected components.

print(mypca.singular_values_)  # Variance values

print(mypca.explained_variance_ratio_)  # Percentages

# Principal axes in the feature space, representing the directions of maximum variance in the data. Components are sorted by explained variance.

print(mypca.components_)

# PCA Results

data_output = mypca.fit_transform(X)

# Estimated noise related to covariance

print(mypca.noise_variance_)

[564941.22927323   1149.42806145    599.59527821]
[9.99993781e-01 4.13956159e-06 1.12643767e-06]
[[ 2.88108368e-04  3.12680649e-06  5.26615536e-06  4.54496005e-06
   5.67742455e-06  6.40142272e-06 -8.68600113e-07 -2.65358195e-06
  -1.44663598e-06  3.38948997e-06  9.45387932e-06 -1.54292830e-05
  -2.23363931e-05  9.99999946e-01 -7.45822317e-07  6.61593278e-06
  -1.47328229e-04  3.32983634e-07 -2.25310724e-05  2.94022681e-05]
 [ 9.97311085e-01  7.22821679e-03  1.92414007e-02 -7.17861907e-03
   2.10169687e-03 -1.08034265e-02 -2.76084222e-03  3.43901500e-03
   1.11028400e-03  1.89063725e-02 -4.02245801e-03  2.16942301e-03
   8.66008066e-03 -2.91240162e-04 -4.64452049e-05 -6.84484088e-03
  -3.14723181e-02 -2.98461637e-04 -3.56400744e-02 -4.46507837e-02]
 [ 4.08648966e-02 -2.19314886e-04 -5.59803698e-03  2.65465205e-02
   5.30864593e-04 -1.05352589e-02 -1.05896720e-03 -5.80377396e-03
  -8.62801219e-03  5.52831256e-03 -9.71655462e-03  6.03946452e-03
  -7.12011394e-03  1.34764393e-04  4.04697045e-04  1.59891469e-02
   9.75491868e-01  4.55087713e-04  2.05332097e-01  5.62102713e-02]]
180.69185527562942


pca = PCA(n_components=2)
pca.fit(dataTelecomSelectedSample)

data_mean = np.mean(dataTelecomSelectedSample, axis=0)
data_std = np.std(dataTelecomSelectedSample, axis=0)
data_standardized = (dataTelecomSelectedSample - data_mean) / data_std

pca_result=pca.fit_transform(data_standardized)
# Get the coordinates of the variables on the first two principal components
variable_coordinates = pca.components_.T

# Calculate the explained variance ratio
explained_variance_ratio = pca.explained_variance_ratio_

# Create the figure and axis
fig, ax = plt.subplots(figsize=(12, 12))

# Plot the correlation circle
circle = plt.Circle((0, 0), radius=1, edgecolor='black', facecolor='None')
ax.add_patch(circle)
ax.set_xlim(-1.1, 1.1)
ax.set_ylim(-1.1, 1.1)

sns.scatterplot(x=pca_result[:, 0], y=pca_result[:, 1], color='yellow', alpha=0.5, legend=False)

ax.axhline(0, color='black', linewidth=0.5)
ax.axvline(0, color='black', linewidth=0.5)

# Add variable labels to the plot
texts = []
colors = ['red', 'blue', 'green', 'orange', 'purple', 'brown', 'gray', 'cyan', 'magenta', 'yellow']

for i, variable in enumerate(dataTelecomSelectedSample.columns[:]):
    color=colors[i % len(colors)]
    arrow = ax.arrow(0, 0, variable_coordinates[i, 0], variable_coordinates[i, 1], head_width=0.05, head_length=0.1, fc=color, ec=color)
    texts.append(plt.text(variable_coordinates[i, 0] + 0.05, variable_coordinates[i, 1] + 0.05, variable, fontsize=12, rotation=45,color=color))

# Adjust the position of variable labels to avoid overlap
adjust_text(texts, arrowprops=dict(arrowstyle="-", color='black'))

# Add axis labels
ax.set_xlabel('PC1 ({}%)'.format(round(explained_variance_ratio[0] * 100, 2)), fontsize=14)
ax.set_ylabel('PC2 ({}%)'.format(round(explained_variance_ratio[1] * 100, 2)), fontsize=14)

# Set plot title
plt.title('PCA Correlation Circle', fontsize=16)

# Equal aspect ratio
ax.set_aspect('equal', adjustable='box')

# Show the plot
plt.show()

	STATUT	OFFRE	ANC_M	HANDSET	revenu_voix	revenu_inter	NB_JOUR_ACTIVITE_TAXE	NB_JOUR_APPEL_TAXE	DUREE_APPEL_TOT	DUREE_APPEL_TAXEE	...	FREQ_USSD_VOIX	FREQ_USSD_SMS	VOLUME_SESSION	VOLUME_SESSION_WEEKEND	REVENU_VAS	ARPU	P_revenu_data	P_revenu_voix_c	P_revenu_vas_c	id_client
0	Active	Offre30	123	2G	12.709375	1.464447	10.366042	10.658842	27.261836	2.125971	...	28.679570	146.930634	200000.000000	54.025577	1.059822	4.325398	0.136685	0.943095	0.279541	1.0
1	Active	Offre8	98	2G	3.000000	0.756078	1.330736	7.079320	19.883099	7.735475	...	28.490597	169.219363	3337.992419	24.789260	1.411803	2.677563	0.533695	0.540374	0.712010	2.0
2	Active	Offre24	90	4G	32.514156	0.681197	0.158160	4.367702	133.476368	7.681088	...	190.000000	44.403308	107082.775926	160.031496	18.195224	1.954007	0.799606	0.071368	0.376754	3.0
3	Active	Offre10	226	2G	3.821551	5.265345	4.003452	3.086766	35.053364	8.540951	...	19.412661	190.000000	128700.752169	102.668979	17.746873	2.496790	0.426526	0.314082	0.676112	4.0
4	Active	Offre1	139	2G	60.009385	1.957144	5.689241	0.403300	0.844956	12.355853	...	81.567146	6.365788	75654.384291	0.341570	7.280861	2.291381	0.319736	0.600057	0.719450	5.0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
95	Active	Offre8	159	2G	3.000000	2.250293	21.961465	2.872333	1.826910	27.907662	...	12.954468	23.226754	32664.259485	21.276030	6.134130	0.689345	0.857627	0.392768	0.657034	96.0
96	Suspended	Offre2	145	2G	3.000000	3.473084	3.575555	0.298674	80.858548	5.349497	...	153.474792	190.000000	82909.748831	26.213769	1.504337	0.164892	0.980014	0.467268	0.150409	97.0
97	Active	Offre10	185	2G	4.531545	2.142349	3.470616	6.400014	107.330103	5.199869	...	157.487581	36.176280	141996.977868	42.321968	1.637815	1.848815	0.772398	0.407095	0.545033	98.0
98	Active	Offre15	197	3G	11.762926	3.474145	2.842419	10.629819	86.011908	0.124836	...	159.332900	77.858228	12052.670191	46.685004	1.422138	1.944146	0.152620	0.243845	0.387906	99.0
99	Suspended	Offre30	178	2G	17.606096	0.019640	1.850818	0.184009	89.864928	5.858613	...	18.339092	47.120168	134081.459601	10.649038	0.640634	2.204618	0.893336	0.293854	0.121368	100.0

	DUREE_APPEL_TOT	nb_sms_tot	revenu_cdr_c	MNT_RECH	FREQ_ACT_OUT	NB_JOUR_APPEL_TAXE	NB_SMS_TAXE	NB_RECH_SUP5	MNT_TRANSFERT_OUT	revenu_voix	MNT_FORFAIT_DATA	NB_FORFAIT_VOIX	OFFRE	VOLUME_SESSION	P_revenu_data	MNT_FORFAIT	FREQ_USSD	P_FF_Data	Duree_onnet_tot	Duree_offnet_tot
0	27.261836	3.685763	19.941537	12.306756	10.667904	10.658842	1.624059	2.00	8.877510	12.709375	2.059265	8.589441	Offre30	200000.000000	0.136685	10.882398	190.000000	0.498091	30.583118	27.219712
1	19.883099	1.476121	3.908449	10.348607	13.650949	7.079320	7.059615	0.67	1.633333	3.000000	20.025208	5.685491	Offre8	3337.992419	0.533695	1.049276	18.623378	0.096339	13.426765	32.571522
2	133.476368	6.162948	32.541849	3.939266	7.079317	4.367702	1.638128	0.00	0.011593	32.514156	5.397060	4.417241	Offre24	107082.775926	0.799606	4.041068	66.449461	0.667540	6.217262	24.861880
3	35.053364	1.448782	4.145009	2.408488	17.423588	3.086766	0.016726	0.33	2.003034	3.821551	0.534564	0.107662	Offre10	128700.752169	0.426526	6.441283	19.259807	0.407815	5.837191	37.779013
4	0.844956	1.131308	60.496821	10.870592	0.067246	0.403300	3.934964	0.33	2.755649	60.009385	0.031219	2.276588	Offre1	75654.384291	0.319736	0.805758	38.034894	0.828712	10.395156	2.831625
5	26.779409	1.477695	3.625786	40.974329	1.374602	3.346885	1.038573	1.67	3.748974	3.000000	2.137682	1.144439	Offre16	65253.482895	0.274582	1.253000	23.161888	0.335963	42.091494	90.670997
6	2.113332	4.376941	33.872104	13.929977	0.730859	5.357250	0.488122	0.00	1.547834	31.953274	0.830859	2.403743	Offre10	45157.810902	0.566405	2.536099	190.000000	0.211775	19.369493	32.897501
7	5.922602	0.773370	15.836382	10.793736	1.060993	0.100707	4.178468	1.00	3.345468	10.331387	1.547844	15.417804	Offre6	37165.463625	0.620034	0.385122	21.880660	0.205227	0.059797	4.935726
8	68.691901	1.314796	21.616162	0.516740	4.607737	2.281113	3.947526	6.67	11.943487	14.262607	19.898847	0.068199	Offre23	200000.000000	0.087045	6.721631	82.203454	0.203176	21.697246	0.627422
9	37.872839	0.901089	9.153639	8.129576	15.158324	11.425205	0.109140	1.00	1.932114	5.870684	17.095836	1.245533	Offre8	65233.389597	0.708388	1.570219	49.499533	0.247614	26.167306	59.049667

	DUREE_APPEL_TOT	nb_sms_tot	revenu_cdr_c	MNT_RECH	FREQ_ACT_OUT	NB_JOUR_APPEL_TAXE	NB_SMS_TAXE	NB_RECH_SUP5	MNT_TRANSFERT_OUT	revenu_voix	MNT_FORFAIT_DATA	NB_FORFAIT_VOIX	OFFRE	VOLUME_SESSION	P_revenu_data	MNT_FORFAIT	FREQ_USSD	P_FF_Data	Duree_onnet_tot	Duree_offnet_tot
19257	30.090072	8.036033	6.628774	21.796702	4.390850	3.572585	2.230938	0.33	6.576157	5.973346	8.614453	2.382774	Offre12	96933.695160	0.596794	10.756903	143.642413	0.014595	44.241075	34.738155
38166	577.035853	8.792087	8.468283	6.180827	1.926081	7.041788	0.025432	5.00	0.692279	3.000000	0.417457	12.181873	Offre7	77383.665104	0.158712	3.796006	57.790428	0.516101	39.859481	8.263161
79907	126.227433	2.056120	9.733714	4.687255	4.980074	8.250577	0.144793	0.00	5.995859	3.000000	3.374781	8.150515	Offre13	186211.083505	0.935933	1.266953	46.308896	0.645644	12.738402	15.707098
9564	0.042172	3.331580	10.712672	12.834082	13.015438	10.310605	6.283435	2.67	6.088506	10.611291	13.975665	7.204972	Offre5	120468.526004	0.154198	1.421667	37.609093	0.708598	41.733081	3.447804
21204	39.975826	4.348320	9.195774	3.873290	7.742715	4.286317	3.506429	0.00	0.665510	7.944436	10.043437	1.146157	Offre24	53403.671898	0.089654	4.067660	190.000000	0.841939	74.725692	22.666880
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
87907	10.833286	22.632752	7.434685	16.818580	1.330702	30.000000	2.258939	0.00	8.323257	6.718734	4.107238	1.454253	Offre29	14172.653991	0.825286	1.269155	47.655680	0.445803	5.349727	9.497032
69701	439.174973	6.930242	10.579883	3.869375	3.062471	4.274253	0.701469	3.67	3.708045	8.752228	8.264411	5.081549	Offre18	96018.498183	0.689052	3.429793	94.129045	0.630098	71.752719	6.520351
1612	34.254143	6.868878	14.567883	1.299701	0.257181	3.908668	2.735732	2.33	1.125163	3.000000	4.784903	18.786315	Offre5	74526.004536	0.463658	3.038143	14.416576	0.491777	7.667571	16.154592
43159	81.030834	1.408071	3.985414	1.817113	5.775719	7.573036	4.426431	0.00	4.559158	3.000000	11.190860	0.234017	Offre3	70646.213006	0.939057	10.895926	174.699935	0.606876	50.623577	58.188738
34600	2.058911	8.345012	3.530926	2.438283	7.900239	2.923418	2.510042	0.00	1.223616	3.000000	10.885480	0.874752	Offre17	85687.964862	0.818186	4.716647	71.896790	0.599819	73.497484	6.421968