import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

#This is used to scale the data from the set.
from sklearn.preprocessing import StandardScaler

#GridSearchCV is used to help tune hyperparameters
from sklearn.model_selection import train_test_split, GridSearchCV

#Used for Decision Tree algorithms and plotting.
from sklearn.tree import DecisionTreeClassifier, plot_tree

#Used for SVM algorithms.
from sklearn.svm import SVC

#Metrics which are are used to evaluate each model.
from sklearn import metrics
from sklearn.metrics import confusion_matrix, classification_report, precision_recall_curve, roc_auc_score, silhouette_score

#Clustering Algorithms.
from sklearn.cluster import KMeans, AgglomerativeClustering

import warnings
warnings.filterwarnings("ignore")

#This loads the Gaming Behaviour Dataset.
df = pd.read_csv('online_gaming_behavior_dataset.csv')

#This shows the first 5 rows of the loaded dataset.
df.head()

#This returns information about the dataset, including the column names, non-null count and the data type of each column.
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 40034 entries, 0 to 40033
Data columns (total 13 columns):
 #   Column                     Non-Null Count  Dtype  
---  ------                     --------------  -----  
 0   PlayerID                   40034 non-null  int64  
 1   Age                        40034 non-null  int64  
 2   Gender                     40034 non-null  object 
 3   Location                   40034 non-null  object 
 4   GameGenre                  40034 non-null  object 
 5   PlayTimeHours              40034 non-null  float64
 6   InGamePurchases            40034 non-null  int64  
 7   GameDifficulty             40034 non-null  object 
 8   SessionsPerWeek            40034 non-null  int64  
 9   AvgSessionDurationMinutes  40034 non-null  int64  
 10  PlayerLevel                40034 non-null  int64  
 11  AchievementsUnlocked       40034 non-null  int64  
 12  EngagementLevel            40034 non-null  object 
dtypes: float64(1), int64(7), object(5)
memory usage: 4.0+ MB

#This shows how many unique values are in each column of the dataset.
df.nunique()

PlayerID                     40034
Age                             35
Gender                           2
Location                         4
GameGenre                        5
PlayTimeHours                40034
InGamePurchases                  2
GameDifficulty                   3
SessionsPerWeek                 20
AvgSessionDurationMinutes      170
PlayerLevel                     99
AchievementsUnlocked            50
EngagementLevel                  3
dtype: int64

#This drops the PlayerID column from the dataset, since it is only an identifier.
df=df.drop(['PlayerID'],axis=1)

#This organises the columns into 2 groups: numerical columns, and categorical columns.
num_cols=['Age','PlayTimeHours','SessionsPerWeek','AvgSessionDurationMinutes','PlayerLevel','AchievementsUnlocked']
cat_cols=['Gender','Location','GameGenre','GameDifficulty', 'InGamePurchases']

#This gives summary statistics of all numerical columns.
df[num_cols].describe().T

#This creates histograms for each numerical column.
df[num_cols].hist(figsize=(12,12),color='lime', edgecolor='black', linewidth=1.5)
plt.tight_layout()
plt.show()

#This prints percentiles for each categorical column.
for i in cat_cols:
    print(df[i].value_counts (normalize=True))
    print('*'*40)

Gender
Male      0.598466
Female    0.401534
Name: proportion, dtype: float64
****************************************
Location
USA       0.399660
Europe    0.299845
Asia      0.202203
Other     0.098291
Name: proportion, dtype: float64
****************************************
GameGenre
Sports        0.201029
Action        0.200804
Strategy      0.200130
Simulation    0.199406
RPG           0.198631
Name: proportion, dtype: float64
****************************************
GameDifficulty
Easy      0.49995
Medium    0.30002
Hard      0.20003
Name: proportion, dtype: float64
****************************************
InGamePurchases
0    0.799146
1    0.200854
Name: proportion, dtype: float64
****************************************

#This creates a crosstab to show the relation between EngagementLevel and other categorical variables, then plots it to a bar graph.
for i in cat_cols:
    if i!='EngagementLevel':
         (pd.crosstab(df[i],df['EngagementLevel'])).plot(kind='bar',figsize=(8,4))
         plt.ylabel('Engagement')
         #Shows the crosstab results as percentiles.
         show_percent = pd.crosstab(df[i], df['EngagementLevel'], normalize='index') * 100
         display(show_percent)

#This shows the mean of numerical variables grouped by the engagement level.
df.groupby(['EngagementLevel'])[num_cols].mean()

#This creates a heatmap to determine the correlation between numerical variables.
plt.figure(figsize=(16,8))
sns.heatmap(df[num_cols].corr(), annot=True, cmap='RdBu_r')
plt.show()

#This creates dummy columns for the Game Genre and Location variables.
df = pd.get_dummies(df, columns=["GameGenre", "Location"], drop_first=True)

#This maps the GameDifficulty, Gender and EngagementLevel variables.
map_difficulty = {'Easy': 1, 'Medium': 2, 'Hard': 3} 
map_gender = {'Male': 0, 'Female': 1}
map_engagement = {'Low': 1, 'Medium': 2, 'High': 3}

df['GameDifficulty'] = df['GameDifficulty'].map(map_difficulty)
df['Gender'] = df['Gender'].map(map_gender)
df['EngagementLevel'] = df['EngagementLevel'].map(map_engagement)

#This seperates our target variable, EngagementLevel, from the other variables in the dataset.
X = df.drop(columns=['EngagementLevel'])
Y = df['EngagementLevel']

#This splits the data into a 70% train and 30% test set.
X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size = 0.3, random_state = 50, stratify = Y)

#This scales our data.
scaler = StandardScaler()

#This fits and transforms the data for X_train.
X_train_scaled = scaler.fit_transform(X_train)
X_train_scaled=pd.DataFrame(X_train_scaled, columns=X.columns)

#This transforms the data for X_test.
X_test_scaled = scaler.transform(X_test)
X_test_scaled=pd.DataFrame(X_test_scaled, columns=X.columns)

#This function allows for a classification report to be printed and a confusion matrix to be plotted.
#This saves writing code for each model.
def metrics_score(actual, predicted):
    print(classification_report(actual, predicted))

    cm = confusion_matrix(actual, predicted)
    plt.figure(figsize=(8,5))
    
    sns.heatmap(cm, annot=True,  fmt='.2f', xticklabels=['Low', 'Medium', 'High'], yticklabels=['Low', 'Medium', 'High'])
    plt.ylabel('Actual')
    plt.xlabel('Predicted')
    plt.show()

#This creates a SVM using the linear kernel, and fits data to it.
svm = SVC(kernel='linear')
model = svm.fit(X= X_train_scaled, y = y_train)

#This checks performance on the train data for the Linear SVM model.
y_pred_train_svm = model.predict(X_train_scaled)

metrics_score(y_train, y_pred_train_svm)

              precision    recall  f1-score   support

           1       0.79      0.71      0.75      7227
           2       0.81      0.87      0.84     13561
           3       0.88      0.84      0.86      7235

    accuracy                           0.82     28023
   macro avg       0.83      0.81      0.82     28023
weighted avg       0.82      0.82      0.82     28023

#This checks performance on the test data for the Linear SVM model.
y_pred_test_svm = model.predict(X_test_scaled)

metrics_score(y_test, y_pred_test_svm)

              precision    recall  f1-score   support

           1       0.79      0.71      0.75      3097
           2       0.81      0.87      0.84      5813
           3       0.89      0.86      0.87      3101

    accuracy                           0.83     12011
   macro avg       0.83      0.81      0.82     12011
weighted avg       0.83      0.83      0.82     12011

#This shows the coefficients of the Linear SVM model, adding it to a DataFrame.
cols = X.columns
coef_svm = svm.coef_

pd.DataFrame(coef_svm, columns=cols).T.sort_values(by=0, ascending=False)

#This calculated the odds from the coefficient.
odds = np.exp(svm.coef_[0]) 
#This shows the odds by adding it to a DataFrame.
pd.DataFrame(odds, X_train_scaled.columns, columns = ['odds']).sort_values(by ='odds', ascending = False)

#This creates a SVM using the RBF kernel, and fits data to it.
svm_rbf=SVC(kernel='rbf',probability=True)
svm_rbf.fit(X_train_scaled,y_train)

SVC(probability=True)

SVC(probability=True)

#This checks performance on the train data for the RBF SVM model.
y_pred_train_rbf = svm_rbf.predict(X_train_scaled)
metrics_score(y_train, y_pred_train_rbf)

              precision    recall  f1-score   support

           1       0.91      0.85      0.88      7227
           2       0.89      0.95      0.92     13561
           3       0.93      0.87      0.90      7235

    accuracy                           0.91     28023
   macro avg       0.91      0.89      0.90     28023
weighted avg       0.91      0.91      0.90     28023

#This checks performance on the test data for the RBF SVM model.
y_pred_test_rbf = svm_rbf.predict(X_test_scaled)
metrics_score(y_test, y_pred_test_rbf)

              precision    recall  f1-score   support

           1       0.89      0.83      0.86      3097
           2       0.87      0.93      0.90      5813
           3       0.92      0.86      0.89      3101

    accuracy                           0.89     12011
   macro avg       0.89      0.88      0.88     12011
weighted avg       0.89      0.89      0.89     12011

#This calculates and prints the AUC-ROC score for the RBF SVM model.
y_scores_svm=svm_rbf.predict_proba(X_train_scaled)
auc_roc = roc_auc_score(y_train, y_scores_svm, multi_class='ovr')
print("AUC-ROC Score: ", auc_roc)

AUC-ROC Score:  0.9492544538462333

#This creates a SVM using the Polynomial kernel, and fits data to it.
svm_poly = SVC(kernel='poly', degree=3)
svm_poly.fit(X_train_scaled,y_train)

SVC(kernel='poly')

SVC(kernel='poly')

#This checks performance on the train data for the Polynomial SVM model.
y_pred_train_poly = svm_poly.predict(X_train_scaled)
metrics_score(y_train, y_pred_train_poly)

              precision    recall  f1-score   support

           1       0.90      0.80      0.85      7227
           2       0.85      0.95      0.90     13561
           3       0.92      0.83      0.87      7235

    accuracy                           0.88     28023
   macro avg       0.89      0.86      0.87     28023
weighted avg       0.88      0.88      0.88     28023

#This checks performance on the test data for the Polynomial SVM model.
y_pred_test_poly = svm_poly.predict(X_test_scaled)
metrics_score(y_test, y_pred_test_poly)

              precision    recall  f1-score   support

           1       0.88      0.78      0.83      3097
           2       0.84      0.93      0.88      5813
           3       0.91      0.83      0.87      3101

    accuracy                           0.86     12011
   macro avg       0.87      0.85      0.86     12011
weighted avg       0.87      0.86      0.86     12011

#This creates a SVM using the Sigmoid kernel, and fits data to it.
svm_sigmoid = SVC(kernel='sigmoid')
svm_sigmoid.fit(X_train_scaled,y_train)

SVC(kernel='sigmoid')

SVC(kernel='sigmoid')

#This checks performance on the train data for the Sigmoid SVM model.
y_pred_train_sigmoid = svm_sigmoid.predict(X_train_scaled)
metrics_score(y_train, y_pred_train_sigmoid)

              precision    recall  f1-score   support

           1       0.58      0.59      0.58      7227
           2       0.67      0.67      0.67     13561
           3       0.71      0.71      0.71      7235

    accuracy                           0.66     28023
   macro avg       0.65      0.66      0.66     28023
weighted avg       0.66      0.66      0.66     28023

#This checks performance on the test data for the Sigmoid SVM model.
y_pred_test_sigmoid = svm_sigmoid.predict(X_test_scaled)
metrics_score(y_test, y_pred_test_sigmoid)

              precision    recall  f1-score   support

           1       0.59      0.61      0.60      3097
           2       0.69      0.68      0.68      5813
           3       0.73      0.72      0.73      3101

    accuracy                           0.67     12011
   macro avg       0.67      0.67      0.67     12011
weighted avg       0.67      0.67      0.67     12011

#This creates a Decision Tree and fits the data to it.
dtree = DecisionTreeClassifier(random_state=50)
dtree.fit(X_train, y_train)

DecisionTreeClassifier(random_state=50)

DecisionTreeClassifier(random_state=50)

#This plots the first 3 layers of the Decision Tree.
plt.figure(figsize=(20,7))
plot_tree(dtree, feature_names=X.columns, class_names=['Low','Medium','High'], filled=True, rounded=True, max_depth = 2)
plt.show()

#This plots the entire decision tree.
plt.figure(figsize=(30,12))
plot_tree(dtree, feature_names=X.columns, class_names=['Low','Medium','High'], filled=True, rounded=True)
plt.show()

#This checks performance on the test data for the Decision Tree model.
y_pred = dtree.predict(X_test)

metrics_score(y_test, y_pred)

              precision    recall  f1-score   support

           1       0.80      0.80      0.80      3097
           2       0.87      0.86      0.87      5813
           3       0.81      0.83      0.82      3101

    accuracy                           0.84     12011
   macro avg       0.83      0.83      0.83     12011
weighted avg       0.84      0.84      0.84     12011

#This calculates and prints the AUC-ROC score for the Decision Tree model.
y_scores_dtree = dtree.predict_proba(X_test)

auc_roc_dtree = roc_auc_score(y_test,y_scores_dtree, multi_class='ovr')              
print("AUC-ROC Score: ", auc_roc_dtree)

AUC-ROC Score:  0.8724360935646601

#This creates a SVM using the RBF kernel, and fits data to it.
#This model has tuned hyperparameters, such as C, the regularization parameter, and gamma, the influence of each training point.
#class_weight='balanced' adjusts and balances the weights of each class for EngagementLevel.
svm_rbf_tuned = SVC(kernel='rbf',C=50, gamma=0.1, class_weight='balanced', probability=True)
svm_rbf_tuned.fit(X_train_scaled,y_train)

SVC(C=50, class_weight='balanced', gamma=0.1, probability=True)

SVC(C=50, class_weight='balanced', gamma=0.1, probability=True)

#This checks performance on the train data for the RBF SVM model using tuned hyperparameters.
y_pred_train_tuned = svm_rbf_tuned.predict(X_train_scaled)

metrics_score(y_train, y_pred_train_tuned)

              precision    recall  f1-score   support

           1       0.96      0.97      0.96      7227
           2       0.98      0.97      0.97     13561
           3       0.96      0.97      0.97      7235

    accuracy                           0.97     28023
   macro avg       0.97      0.97      0.97     28023
weighted avg       0.97      0.97      0.97     28023

#This checks performance on the test data for the RBF SVM model using tuned hyperparameters.
y_pred_test_tuned = svm_rbf_tuned.predict(X_test_scaled)

metrics_score(y_test, y_pred_test_tuned)

              precision    recall  f1-score   support

           1       0.76      0.79      0.78      3097
           2       0.85      0.83      0.84      5813
           3       0.82      0.82      0.82      3101

    accuracy                           0.82     12011
   macro avg       0.81      0.81      0.81     12011
weighted avg       0.82      0.82      0.82     12011

#This calculates and prints the AUC-ROC score for the RBF SVM model using tuned hyperparameters.
y_scores_tuned=svm_rbf_tuned.predict_proba(X_train_scaled)
auc_roc_tuned = roc_auc_score(y_train, y_scores_tuned, multi_class='ovr')
print("AUC-ROC Score: ", auc_roc_tuned)

AUC-ROC Score:  0.9913914716746848

#This is a parameter grid of different hyperparameters, which will be used for GridSearchCV.
tree_param = {
    'criterion' :['gini', 'entropy'],
    'max_depth': [10, 20, 30, None],
    'min_samples_split': [2, 5, 10],
    'min_samples_leaf': [1, 2, 4],
}

#This creates a Decision Tree.
dtree_tuned = DecisionTreeClassifier(random_state=50)

#This uses GridSearchCV to help find the best parameters to tune the Decision Tree model.
grid_search_dtree = GridSearchCV(estimator = dtree_tuned, param_grid=tree_param, cv=5, n_jobs=-1, verbose=True)

#This fits the train and test data.
grid_search_dtree.fit(X_train, y_train)

#This displays the best parameters found by GridSearchCV.
display(grid_search_dtree.best_params_)

#This applies the found parameters to the Decision Tree model.
dtree_tuned = grid_search_dtree.best_estimator_

Fitting 5 folds for each of 72 candidates, totalling 360 fits

{'criterion': 'entropy',
 'max_depth': 10,
 'min_samples_leaf': 4,
 'min_samples_split': 10}

#This plots the first 3 layers of the tuned Decision Tree.
plt.figure(figsize=(20,7))
plot_tree(dtree_tuned, feature_names=X.columns, class_names=['Low','Medium','High'], filled=True, rounded=True, max_depth = 2)
plt.show()

#This plots the entire tuned Decision Tree.
plt.figure(figsize=(30,12))
plot_tree(dtree_tuned, feature_names=X.columns, class_names=['Low','Medium','High'], filled=True, rounded=True)
plt.show()

#This checks performance on the test data for the tuned Decision Tree model.
y_pred_tuned = dtree_tuned.predict(X_test)

metrics_score(y_test, y_pred_tuned)

              precision    recall  f1-score   support

           1       0.89      0.87      0.88      3097
           2       0.91      0.94      0.92      5813
           3       0.92      0.89      0.90      3101

    accuracy                           0.91     12011
   macro avg       0.91      0.90      0.90     12011
weighted avg       0.91      0.91      0.91     12011

#This calculates and prints the AUC-ROC score for the tuned Decision Tree model.
y_scores_dtree_tuned = dtree_tuned.predict_proba(X_test)

auc_roc_dtree = roc_auc_score(y_test,y_scores_dtree_tuned, multi_class='ovr')              
print("AUC-ROC Score: ", auc_roc_dtree)

AUC-ROC Score:  0.9366403804622262

#This picks random sample data from the dataset, then fits and transforms it.
sample = df.sample(n=200, random_state=32)
sample_scale = scaler.fit_transform(sample[['PlayerLevel','AchievementsUnlocked']])
sample2 = df.sample(n=200, random_state=87)
sample2_scale = scaler.fit_transform(sample2[['Age','PlayTimeHours']])

#This creates a K-Means model using 3 clusters, and fits data to it.
kmeans = KMeans(n_clusters=3, random_state=7)
kmeans.fit(sample_scale)

#This applies cluster labels using K-Means.
kmeans_labels = kmeans.labels_

#This applies the labels from K-Means to the data, and plots it to a scatter graph.
sample['Cluster'] = kmeans_labels
sns.scatterplot(x=sample['PlayerLevel'],y=sample['AchievementsUnlocked'],hue=sample['Cluster'],data=sample,palette='Set1')

<Axes: xlabel='PlayerLevel', ylabel='AchievementsUnlocked'>

#This finds and prints both the Silhouette and Inertia scores for the K-Means model.
silhouette_km = silhouette_score(sample_scale, kmeans_labels)
inertia_km = kmeans.inertia_

print("Silhouette Score: ", silhouette_km)
print("Inertia Score: ", inertia_km)

Silhouette Score:  0.4013138016087001
Inertia Score:  147.4557070418625

#This finds the Within Cluster Sum of Squares to find the best amount of clusters, using the elbow method.
WCSS= []
K = range(1,9)
for i in K:
    Kmeanmodel=KMeans(n_clusters=i, init='k-means++', max_iter=300, n_init=10, random_state=42)
    Kmeanmodel.fit(df[['PlayerLevel','AchievementsUnlocked']])
    WCSS.append(Kmeanmodel.inertia_)

#This plots the WCSS to a graph.
plt.plot(range(1,9),WCSS)
plt.xlabel('Number of clusters')
plt.ylabel('WCSS')
plt.show()

#This creates a Agglomerative Clustering model using 3 clusters, and fits data to it.
agg_clustering = AgglomerativeClustering(n_clusters=3)

#This applies cluster labels using Agglomerative Clustering.
agg_cluster_labels = agg_clustering.fit_predict(sample_scale)

#This applies the labels from Agglomerative Clustering to the data, and plots it to a scatter graph.
sample['Cluster'] = agg_cluster_labels
sns.scatterplot(x=sample['PlayerLevel'],y=sample['AchievementsUnlocked'],hue=sample['Cluster'],palette='Set1')

<Axes: xlabel='PlayerLevel', ylabel='AchievementsUnlocked'>

#This finds and prints the Silhouette score for the Agglomerative Clustering model.
silhouette_agg = silhouette_score(sample_scale, agg_cluster_labels)
print("Silhouette Score:", silhouette_agg)

Silhouette Score: 0.3802371074372398

#This creates another Agglomerative Clustering model using 4 clusters, and fits data to it.
agg_clustering2 = AgglomerativeClustering(n_clusters=4)

#This applies cluster labels using Agglomerative Clustering.
agg_cluster_labels2 = agg_clustering2.fit_predict(sample2_scale)

#This applies the labels from Agglomerative Clustering to the data, and plots it to a scatter graph.
#The data here is the age of the player and the respective play time hours.
sample2['Cluster'] = agg_cluster_labels2
sns.scatterplot(x=sample2['Age'],y=sample2['PlayTimeHours'],hue=sample2['Cluster'],palette='Set1')

<Axes: xlabel='Age', ylabel='PlayTimeHours'>

#This finds and prints the Silhouette score for the Agglomerative Clustering model.
silhouette_agg2 = silhouette_score(sample2_scale, agg_cluster_labels2)
print("Silhouette Score:", silhouette_agg2)

Silhouette Score: 0.40824535777628673

	count	mean	std	min	25%	50%	75%	max
Age	40034.0	31.992531	10.043227	15.000000	23.000000	32.000000	41.000000	49.000000
PlayTimeHours	40034.0	12.024365	6.914638	0.000115	6.067501	12.008002	17.963831	23.999592
SessionsPerWeek	40034.0	9.471774	5.763667	0.000000	4.000000	9.000000	14.000000	19.000000
AvgSessionDurationMinutes	40034.0	94.792252	49.011375	10.000000	52.000000	95.000000	137.000000	179.000000
PlayerLevel	40034.0	49.655568	28.588379	1.000000	25.000000	49.000000	74.000000	99.000000
AchievementsUnlocked	40034.0	24.526477	14.430726	0.000000	12.000000	25.000000	37.000000	49.000000

EngagementLevel	High	Low	Medium
Location
Asia	26.324892	25.423101	48.252007
Europe	25.558147	26.182939	48.258914
Other	25.667090	26.327827	48.005083
USA	25.793750	25.543750	48.662500

EngagementLevel	High	Low	Medium
GameGenre
Action	25.674835	25.612638	48.712526
RPG	25.000000	26.370724	48.629276
Simulation	26.356007	26.155581	47.488413
Sports	25.882207	25.770378	48.347416
Strategy	26.173240	25.037444	48.789316

EngagementLevel	High	Low	Medium
GameDifficulty
Easy	25.640769	26.010492	48.348738
Hard	26.373626	25.924076	47.702298
Medium	25.743069	25.326784	48.930147

EngagementLevel	High	Low	Medium
InGamePurchases
0	25.636858	25.902541	48.460601
1	26.538988	25.332670	48.128342

	PlayerID	Age	Gender	Location	GameGenre	PlayTimeHours	InGamePurchases	GameDifficulty	SessionsPerWeek	AvgSessionDurationMinutes	PlayerLevel	AchievementsUnlocked	EngagementLevel
0	9000	43	Male	Other	Strategy	16.271119	0	Medium	6	108	79	25	Medium
1	9001	29	Female	USA	Strategy	5.525961	0	Medium	5	144	11	10	Medium
2	9002	22	Female	USA	Sports	8.223755	0	Easy	16	142	35	41	High
3	9003	35	Male	USA	Action	5.265351	1	Easy	9	85	57	47	Medium
4	9004	33	Male	Europe	Action	15.531945	0	Medium	2	131	95	37	Medium

EngagementLevel	High	Low	Medium
Gender
Female	26.102644	25.636081	48.261275
Male	25.627113	25.890062	48.482825

	Age	PlayTimeHours	SessionsPerWeek	AvgSessionDurationMinutes	PlayerLevel	AchievementsUnlocked
EngagementLevel
High	31.920085	12.069238	14.254547	131.921827	50.823723	25.095975
Low	31.896939	12.104915	4.530511	66.882119	46.101414	22.661565
Medium	32.082120	11.957503	9.553267	89.856405	50.926293	25.216424

	0	1	2
Location_Europe	0.031967	0.017114	-0.000226
Location_USA	0.021377	0.010977	0.011372
Location_Other	0.015172	-0.006107	0.011677
Gender	0.013748	0.003636	-0.001112
GameDifficulty	0.008906	-0.014510	-0.010850
PlayTimeHours	0.007896	0.006242	-0.013970
GameGenre_Sports	0.006009	0.007954	0.019745
GameGenre_Simulation	-0.002108	-0.001512	-0.011538
GameGenre_Strategy	-0.003431	0.008282	0.009721
GameGenre_RPG	-0.003809	-0.004110	0.001992
InGamePurchases	-0.009646	-0.009738	-0.029297
Age	-0.016962	0.004590	0.036140
AchievementsUnlocked	-0.237361	-0.093110	-0.087680
PlayerLevel	-0.238035	-0.128655	-0.100413
AvgSessionDurationMinutes	-1.055377	-0.970449	-1.808935
SessionsPerWeek	-1.636695	-1.234154	-1.954769

	odds
Location_Europe	1.032483
Location_USA	1.021607
Location_Other	1.015287
Gender	1.013842
GameDifficulty	1.008946
PlayTimeHours	1.007927
GameGenre_Sports	1.006027
GameGenre_Simulation	0.997895
GameGenre_Strategy	0.996575
GameGenre_RPG	0.996198
InGamePurchases	0.990400
Age	0.983181
AchievementsUnlocked	0.788707
PlayerLevel	0.788175
AvgSessionDurationMinutes	0.348061
SessionsPerWeek	0.194622