Assignment

data_science_assignment.py

import pandas as pd
import numpy as np
from sklearn.preprocessing import MinMaxScaler
from sklearn import metrics

df = pd.read_csv('mammographic_masses.data.txt',feature_names=['BI_RADS','age','shape','margin','density','severity'])
df = df.replace('?', np.nan)
#check data is not baised and equally distributed 
df.describe(include='all')
df.dropna(inplace=True)
X = df[['age','shape','margin','density']].values
scaler = MinMaxScaler()
#check differnece betwee fit and fit_transform
X = scaler.fit_transform(X)
y = df.severity
from sklearn.datasets import load_iris
from sklearn.tree import DecisionTreeClassifier
from sklearn.model_selection import train_test_split
#check random_state
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=1)
clf = DecisionTreeClassifier(random_state=1)
clf = clf.fit(X_train,y_train)
y_pred = clf.predict(X_test)

metrics.accuracy_score(y_test, y_pred)
# 0.7269076305220884

from sklearn.model_selection import KFold,cross_val_score
cv = KFold(n_splits=10, random_state=1, shuffle=True)
scores = cross_val_score(clf, X, y, scoring='accuracy', cv=cv)
np.mean(scores)

from sklearn.ensemble import RandomForestClassifier
rf_clf = RandomForestClassifier(random_state=1)
rf_clf.fit(X_train,y_train)
# RandomForestClassifier(bootstrap=True, ccp_alpha=0.0, class_weight=None,
#                        criterion='gini', max_depth=None, max_features='auto',
#                        max_leaf_nodes=None, max_samples=None,
#                        min_impurity_decrease=0.0, min_impurity_split=None,
#                        min_samples_leaf=1, min_samples_split=2,
#                        min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=2,
#                        oob_score=False, random_state=0, verbose=0,
#                        warm_start=False)
rf_clf.predict(X_test[0:10])

scores = cross_val_score(clf, X, y, scoring='accuracy', cv=cv)
np.mean(scores)
# 0.7759036144578313

from sklearn import svm
svm_clf = svm.SVC(kernel='linear', C = 1.0)
scores = cross_val_score(svm_clf, X, y, scoring='accuracy', cv=cv)
np.mean(scores)
#0.7915662650602409

from sklearn.neighbors import KNeighborsClassifier
knn_clf = KNeighborsClassifier(n_neighbors=3)
scores = cross_val_score(knn_clf, X, y, scoring='accuracy', cv=cv)
np.mean(scores)
#0.7674698795180723

from sklearn.naive_bayes import MultinomialNB
nb_clf = MultinomialNB()
scores = cross_val_score(nb_clf, X, y, scoring='accuracy', cv=cv)
np.mean(scores)

model = LogisticRegression()
scores = cross_val_score(nb_clf, X, y, scoring='accuracy', cv=cv)
np.mean(scores)

mammographic_masses.data.txt

5,67,3,5,3,1
4,43,1,1,?,1
5,58,4,5,3,1
4,28,1,1,3,0
5,74,1,5,?,1
4,65,1,?,3,0
4,70,?,?,3,0
5,42,1,?,3,0
5,57,1,5,3,1
5,60,?,5,1,1
5,76,1,4,3,1
3,42,2,1,3,1
4,64,1,?,3,0
4,36,3,1,2,0
4,60,2,1,2,0
4,54,1,1,3,0
3,52,3,4,3,0
4,59,2,1,3,1
4,54,1,1,3,1
4,40,1,?,?,0
?,66,?,?,1,1
5,56,4,3,1,1
4,43,1,?,?,0
5,42,4,4,3,1
4,59,2,4,3,1
5,75,4,5,3,1
2,66,1,1,?,0
5,63,3,?,3,0
5,45,4,5,3,1
5,55,4,4,3,0
4,46,1,5,2,0
5,54,4,4,3,1
5,57,4,4,3,1
4,39,1,1,2,0
4,81,1,1,3,0
4,77,3,?,?,0
4,60,2,1,3,0
5,67,3,4,2,1
4,48,4,5,?,1
4,55,3,4,2,0
4,59,2,1,?,0
4,78,1,1,1,0
4,50,1,1,3,0
4,61,2,1,?,0
5,62,3,5,2,1
5,44,2,4,?,1
5,64,4,5,3,1
4,23,1,1,?,0
2,42,?,?,4,0
5,67,4,5,3,1
4,74,2,1,2,0
5,80,3,5,3,1
4,23,1,1,?,0
4,63,2,1,?,0
4,53,?,5,3,1
4,43,3,4,?,0
4,49,2,1,1,0
5,51,2,4,?,0
4,45,2,1,?,0
5,59,2,?,?,1
5,52,4,3,3,1
5,60,4,3,3,1
4,57,2,5,3,0
3,57,2,1,?,0
5,74,4,4,3,1
4,25,2,1,?,0
4,49,1,1,3,0
5,72,4,3,?,1
4,45,2,1,3,0
4,64,2,1,3,0
4,73,2,1,2,0
5,68,4,3,3,1
5,52,4,5,3,0
5,66,4,4,3,1
5,70,?,4,?,1
4,25,1,1,3,0
5,74,1,1,2,1
4,64,1,1,3,0
5,60,4,3,2,1
5,67,2,4,1,0
4,67,4,5,3,0
5,44,4,4,2,1
3,68,1,1,3,1
4,57,?,4,1,0
5,51,4,?,?,1
4,33,1,?,?,0
5,58,4,4,3,1
5,36,1,?,?,0
4,63,1,1,?,0
5,62,1,5,3,1
4,73,3,4,3,1
4,80,4,4,3,1
4,67,1,1,?,0
5,59,2,1,3,1
5,60,1,?,3,0
5,54,4,4,3,1
4,40,1,1,?,0
4,47,2,1,?,0
5,62,4,4,3,0
4,33,2,1,3,0
5,59,2,?,?,0
4,65,2,?,?,0
4,58,4,4,?,0
4,29,2,?,?,0
4,58,1,1,?,0
4,54,1,1,?,0
4,44,1,1,?,1
3,34,2,1,?,0
4,57,1,1,3,0
5,33,4,4,?,1
4,45,4,4,3,0
5,71,4,4,3,1
5,59,4,4,2,0
4,56,2,1,?,0
4,40,3,4,?,0
4,56,1,1,3,0
4,45,2,1,?,0
4,57,2,1,2,0
5,55,3,4,3,1
5,84,4,5,3,0
5,51,4,4,3,1
4,43,1,1,?,0
4,24,2,1,2,0
4,66,1,1,3,0
5,33,4,4,3,0
4,59,4,3,2,0
4,76,2,3,?,0
4,40,1,1,?,0
4,52,?,4,?,0
5,40,4,5,3,1
5,67,4,4,3,1
5,75,4,3,3,1
5,86,4,4,3,0
4,60,2,?,?,0
5,66,4,4,3,1
5,46,4,5,3,1
4,59,4,4,3,1
5,65,4,4,3,1
4,53,1,1,3,0
5,67,3,5,3,1
5,80,4,5,3,1
4,55,2,1,3,0
4,48,1,1,?,0
4,47,1,1,2,0
4,50,2,1,?,0
5,62,4,5,3,1
5,63,4,4,3,1
4,63,4,?,3,1
4,71,4,4,3,1
4,41,1,1,3,0
5,57,4,4,4,1
5,71,4,4,4,1
4,66,1,1,3,0
4,47,2,4,2,0
3,34,4,4,3,0
4,59,3,4,3,0
5,55,2,?,?,1
4,51,?,?,3,0
4,62,2,1,?,0
4,58,4,?,3,1
5,67,4,4,3,1
4,41,2,1,3,0
4,23,3,1,3,0
4,53,?,4,3,0
4,42,2,1,3,0
5,87,4,5,3,1
4,68,1,1,3,1
4,64,1,1,3,0
5,54,3,5,3,1
5,86,4,5,3,1
4,21,2,1,3,0
4,39,1,1,?,0
4,53,4,4,3,0
4,44,4,4,3,0
4,54,1,1,3,0
5,63,4,5,3,1
4,62,2,1,?,0
4,45,2,1,2,0
5,71,4,5,3,0
5,49,4,4,3,1
4,49,4,4,3,0
5,66,4,4,4,0
4,19,1,1,3,0
4,35,1,1,2,0
4,71,3,3,?,1
5,74,4,5,3,1
5,37,4,4,3,1
4,67,1,?,3,0
5,81,3,4,3,1
5,59,4,4,3,1
4,34,1,1,3,0
5,79,4,3,3,1
5,60,3,1,3,0
4,41,1,1,3,1
4,50,1,1,3,0
5,85,4,4,3,1
4,46,1,1,3,0
5,66,4,4,3,1
4,73,3,1,2,0
4,55,1,1,3,0
4,49,2,1,3,0
3,49,4,4,3,0
4,51,4,5,3,1
2,48,4,4,3,0
4,58,4,5,3,0
5,72,4,5,3,1
4,46,2,3,3,0
4,43,4,3,3,1
?,52,4,4,3,0
4,66,2,1,?,0
4,46,1,1,1,0
4,69,3,1,3,0
2,59,1,1,?,1
5,43,2,1,3,1
5,76,4,5,3,1
4,46,1,1,3,0
4,59,2,4,3,0
4,57,1,1,3,0
5,43,4,5,?,0
3,45,2,1,3,0
3,43,2,1,3,0
4,45,2,1,3,0
5,57,4,5,3,1
5,79,4,4,3,1
5,54,2,1,3,1
4,40,3,4,3,0
5,63,4,4,3,1
2,55,1,?,1,0
4,52,2,1,3,0
4,38,1,1,3,0
3,72,4,3,3,0
5,80,4,3,3,1
5,76,4,3,3,1
4,62,3,1,3,0
5,64,4,5,3,1
5,42,4,5,3,0
3,60,?,3,1,0
4,64,4,5,3,0
4,63,4,4,3,1
4,24,2,1,2,0
5,72,4,4,3,1
4,63,2,1,3,0
4,46,1,1,3,0
3,33,1,1,3,0
5,76,4,4,3,1
4,36,2,3,3,0
4,40,2,1,3,0
5,58,1,5,3,1
4,43,2,1,3,0
3,42,1,1,3,0
4,32,1,1,3,0
5,57,4,4,2,1
4,37,1,1,3,0
4,70,4,4,3,1
5,56,4,2,3,1
3,76,?,3,2,0
5,73,4,4,3,1
5,77,4,5,3,1
5,67,4,4,1,1
5,71,4,3,3,1
5,65,4,4,3,1
4,43,1,1,3,0
4,40,2,1,?,0
4,49,2,1,3,0
5,76,4,2,3,1
4,55,4,4,3,0
5,72,4,5,3,1
3,53,4,3,3,0
5,75,4,4,3,1
5,61,4,5,3,1
5,67,4,4,3,1
5,55,4,2,3,1
5,66,4,4,3,1
2,76,1,1,2,0
4,57,4,4,3,1
5,71,3,1,3,0
5,70,4,5,3,1
4,35,4,2,?,0
5,79,1,?,3,1
4,63,2,1,3,0
5,40,1,4,3,1
4,41,1,1,3,0
4,47,2,1,2,0
4,68,1,1,3,1
4,64,4,3,3,1
4,65,4,4,?,1
4,73,4,3,3,0
4,39,4,3,3,0
5,55,4,5,4,1
5,53,3,4,4,0
5,66,4,4,3,1
4,43,3,1,2,0
5,44,4,5,3,1
4,77,4,4,3,1
4,62,2,4,3,0
5,80,4,4,3,1
4,33,4,4,3,0
4,50,4,5,3,1
4,71,1,?,3,0
5,46,4,4,3,1
5,49,4,5,3,1
4,53,1,1,3,0
3,46,2,1,2,0
4,57,1,1,3,0
4,54,3,1,3,0
4,54,1,?,?,0
2,49,2,1,2,0
4,47,3,1,3,0
4,40,1,1,3,0
4,45,1,1,3,0
4,50,4,5,3,1
5,54,4,4,3,1
4,67,4,1,3,1
4,77,4,4,3,1
4,66,4,3,3,0
4,71,2,?,3,1
4,36,2,3,3,0
4,69,4,4,3,0
4,48,1,1,3,0
4,64,4,4,3,1
4,71,4,2,3,1
5,60,4,3,3,1
4,24,1,1,3,0
5,34,4,5,2,1
4,79,1,1,2,0
4,45,1,1,3,0
4,37,2,1,2,0
4,42,1,1,2,0
4,72,4,4,3,1
5,60,4,5,3,1
5,85,3,5,3,1
4,51,1,1,3,0
5,54,4,5,3,1
5,55,4,3,3,1
4,64,4,4,3,0
5,67,4,5,3,1
5,75,4,3,3,1
5,87,4,4,3,1
4,46,4,4,3,1
4,59,2,1,?,0
55,46,4,3,3,1
5,61,1,1,3,1
4,44,1,4,3,0
4,32,1,1,3,0
4,62,1,1,3,0
5,59,4,5,3,1
4,61,4,1,3,0
5,78,4,4,3,1
5,42,4,5,3,0
4,45,1,2,3,0
5,34,2,1,3,1
5,39,4,3,?,1
4,27,3,1,3,0
4,43,1,1,3,0
5,83,4,4,3,1
4,36,2,1,3,0
4,37,2,1,3,0
4,56,3,1,3,1
5,55,4,4,3,1
5,46,3,?,3,0
4,88,4,4,3,1
5,71,4,4,3,1
4,41,2,1,3,0
5,49,4,4,3,1
3,51,1,1,4,0
4,39,1,3,3,0
4,46,2,1,3,0
5,52,4,4,3,1
5,58,4,4,3,1
4,67,4,5,3,1
5,80,4,4,3,1
3,46,1,?,?,0
3,43,1,?,?,0
4,45,1,1,3,0
5,68,4,4,3,1
4,54,4,4,?,1
4,44,2,3,3,0
5,74,4,3,3,1
5,55,4,5,3,0
4,49,4,4,3,1
4,49,1,1,3,0
5,50,4,3,3,1
5,52,3,5,3,1
4,45,1,1,3,0
4,66,1,1,3,0
4,68,4,4,3,1
4,72,2,1,3,0
5,64,?,?,3,0
2,49,?,3,3,0
3,44,?,4,3,0
5,74,4,4,3,1
5,58,4,4,3,1
4,77,2,3,3,0
4,49,3,1,3,0
4,34,?,?,4,0
5,60,4,3,3,1
5,69,4,3,3,1
4,53,2,1,3,0
3,46,3,4,3,0
5,74,4,4,3,1
4,58,1,1,3,0
5,68,4,4,3,1
5,46,4,3,3,0
5,61,2,4,3,1
5,70,4,3,3,1
5,37,4,4,3,1
3,65,4,5,3,1
4,67,4,4,3,0
5,69,3,4,3,0
5,76,4,4,3,1
4,65,4,3,3,0
5,72,4,2,3,1
4,62,4,2,3,0
5,42,4,4,3,1
5,66,4,3,3,1
5,48,4,4,3,1
4,35,1,1,3,0
5,60,4,4,3,1
5,67,4,2,3,1
5,78,4,4,3,1
4,66,1,1,3,1
4,26,1,1,?,0
4,48,1,1,3,0
4,31,1,1,3,0
5,43,4,3,3,1
5,72,2,4,3,0
5,66,1,1,3,1
4,56,4,4,3,0
5,58,4,5,3,1
5,33,2,4,3,1
4,37,1,1,3,0
5,36,4,3,3,1
4,39,2,3,3,0
4,39,4,4,3,1
5,83,4,4,3,1
4,68,4,5,3,1
5,63,3,4,3,1
5,78,4,4,3,1
4,38,2,3,3,0
5,46,4,3,3,1
5,60,4,4,3,1
5,56,2,3,3,1
4,33,1,1,3,0
4,?,4,5,3,1
4,69,1,5,3,1
5,66,1,4,3,1
4,72,1,3,3,0
4,29,1,1,3,0
5,54,4,5,3,1
5,80,4,4,3,1
5,68,4,3,3,1
4,35,2,1,3,0
4,57,3,?,3,0
5,?,4,4,3,1
4,50,1,1,3,0
4,32,4,3,3,0
0,69,4,5,3,1
4,71,4,5,3,1
5,87,4,5,3,1
3,40,2,?,3,0
4,31,1,1,?,0
4,64,1,1,3,0
5,55,4,5,3,1
4,18,1,1,3,0
3,50,2,1,?,0
4,53,1,1,3,0
5,84,4,5,3,1
5,80,4,3,3,1
4,32,1,1,3,0
5,77,3,4,3,1
4,38,1,1,3,0
5,54,4,5,3,1
4,63,1,1,3,0
4,61,1,1,3,0
4,52,1,1,3,0
4,36,1,1,3,0
4,41,?,?,3,0
4,59,1,1,3,0
5,51,4,4,2,1
4,36,1,1,3,0
5,40,4,3,3,1
4,49,1,1,3,0
4,37,2,3,3,0
4,46,1,1,3,0
4,63,1,1,3,0
4,28,2,1,3,0
4,47,2,1,3,0
4,42,2,1,3,1
5,44,4,5,3,1
4,49,4,4,3,0
5,47,4,5,3,1
5,52,4,5,3,1
4,53,1,1,3,1
5,83,3,3,3,1
4,50,4,4,?,1
5,63,4,4,3,1
4,82,?,5,3,1
4,54,1,1,3,0
4,50,4,4,3,0
5,80,4,5,3,1
5,45,2,4,3,0
5,59,4,4,?,1
4,28,2,1,3,0
4,31,1,1,3,0
4,41,2,1,3,0
4,21,3,1,3,0
5,44,3,4,3,1
5,49,4,4,3,1
5,71,4,5,3,1
5,75,4,5,3,1
4,38,2,1,3,0
4,60,1,3,3,0
5,87,4,5,3,1
4,70,4,4,3,1
5,55,4,5,3,1
3,21,1,1,3,0
4,50,1,1,3,0
5,76,4,5,3,1
4,23,1,1,3,0
3,68,?,?,3,0
4,62,4,?,3,1
5,65,1,?,3,1
5,73,4,5,3,1
4,38,2,3,3,0
2,57,1,1,3,0
5,65,4,5,3,1
5,67,2,4,3,1
5,61,2,4,3,1
5,56,4,4,3,0
5,71,2,4,3,1
4,49,2,2,3,0
4,55,?,?,3,0
4,44,2,1,3,0
0,58,4,4,3,0
4,27,2,1,3,0
5,73,4,5,3,1
4,34,2,1,3,0
5,63,?,4,3,1
4,50,2,1,3,1
4,62,2,1,3,0
3,21,3,1,3,0
4,49,2,?,3,0
4,36,3,1,3,0
4,45,2,1,3,1
5,67,4,5,3,1
4,21,1,1,3,0
4,57,2,1,3,0
5,66,4,5,3,1
4,71,4,4,3,1
5,69,3,4,3,1
6,80,4,5,3,1
3,27,2,1,3,0
4,38,2,1,3,0
4,23,2,1,3,0
5,70,?,5,3,1
4,46,4,3,3,0
4,61,2,3,3,0
5,65,4,5,3,1
4,60,4,3,3,0
5,83,4,5,3,1
5,40,4,4,3,1
2,59,?,4,3,0
4,53,3,4,3,0
4,76,4,4,3,0
5,79,1,4,3,1
5,38,2,4,3,1
4,61,3,4,3,0
4,56,2,1,3,0
4,44,2,1,3,0
4,64,3,4,?,1
4,66,3,3,3,0
4,50,3,3,3,0
4,46,1,1,3,0
4,39,1,1,3,0
4,60,3,?,?,0
5,55,4,5,3,1
4,40,2,1,3,0
4,26,1,1,3,0
5,84,3,2,3,1
4,41,2,2,3,0
4,63,1,1,3,0
2,65,?,1,2,0
4,49,1,1,3,0
4,56,2,2,3,1
5,65,4,4,3,0
4,54,1,1,3,0
4,36,1,1,3,0
5,49,4,4,3,0
4,59,4,4,3,1
5,75,4,4,3,1
5,59,4,2,3,0
5,59,4,4,3,1
4,28,4,4,3,1
5,53,4,5,3,0
5,57,4,4,3,0
5,77,4,3,4,0
5,85,4,3,3,1
4,59,4,4,3,0
5,59,1,5,3,1
4,65,3,3,3,1
4,54,2,1,3,0
5,46,4,5,3,1
4,63,4,4,3,1
4,53,1,1,3,1
4,56,1,1,3,0
5,66,4,4,3,1
5,66,4,5,3,1
4,55,1,1,3,0
4,44,1,1,3,0
5,86,3,4,3,1
5,47,4,5,3,1
5,59,4,5,3,1
5,66,4,5,3,0
5,61,4,3,3,1
3,46,?,5,?,1
4,69,1,1,3,0
5,93,1,5,3,1
4,39,1,3,3,0
5,44,4,5,3,1
4,45,2,2,3,0
4,51,3,4,3,0
4,56,2,4,3,0
4,66,4,4,3,0
5,61,4,5,3,1
4,64,3,3,3,1
5,57,2,4,3,0
5,79,4,4,3,1
4,57,2,1,?,0
4,44,4,1,1,0
4,31,2,1,3,0
4,63,4,4,3,0
4,64,1,1,3,0
5,47,4,5,3,0
5,68,4,5,3,1
4,30,1,1,3,0
5,43,4,5,3,1
4,56,1,1,3,0
4,46,2,1,3,0
4,67,2,1,3,0
5,52,4,5,3,1
4,67,4,4,3,1
4,47,2,1,3,0
5,58,4,5,3,1
4,28,2,1,3,0
4,43,1,1,3,0
4,57,2,4,3,0
5,68,4,5,3,1
4,64,2,4,3,0
4,64,2,4,3,0
5,62,4,4,3,1
4,38,4,1,3,0
5,68,4,4,3,1
4,41,2,1,3,0
4,35,2,1,3,1
4,68,2,1,3,0
5,55,4,4,3,1
5,67,4,4,3,1
4,51,4,3,3,0
2,40,1,1,3,0
5,73,4,4,3,1
4,58,?,4,3,1
4,51,?,4,3,0
3,50,?,?,3,1
5,59,4,3,3,1
6,60,3,5,3,1
4,27,2,1,?,0
5,54,4,3,3,0
4,56,1,1,3,0
5,53,4,5,3,1
4,54,2,4,3,0
5,79,1,4,3,1
5,67,4,3,3,1
5,64,3,3,3,1
4,70,1,2,3,1
5,55,4,3,3,1
5,65,3,3,3,1
5,45,4,2,3,1
4,57,4,4,?,1
5,49,1,1,3,1
4,24,2,1,3,0
4,52,1,1,3,0
4,50,2,1,3,0
4,35,1,1,3,0
5,?,3,3,3,1
5,64,4,3,3,1
5,40,4,1,1,1
5,66,4,4,3,1
4,64,4,4,3,1
5,52,4,3,3,1
5,43,1,4,3,1
4,56,4,4,3,0
4,72,3,?,3,0
6,51,4,4,3,1
4,79,4,4,3,1
4,22,2,1,3,0
4,73,2,1,3,0
4,53,3,4,3,0
4,59,2,1,3,1
4,46,4,4,2,0
5,66,4,4,3,1
4,50,4,3,3,1
4,58,1,1,3,1
4,55,1,1,3,0
4,62,2,4,3,1
4,60,1,1,3,0
5,57,4,3,3,1
4,57,1,1,3,0
6,41,2,1,3,0
4,71,2,1,3,1
4,32,2,1,3,0
4,57,2,1,3,0
4,19,1,1,3,0
4,62,2,4,3,1
5,67,4,5,3,1
4,50,4,5,3,0
4,65,2,3,2,0
4,40,2,4,2,0
6,71,4,4,3,1
6,68,4,3,3,1
4,68,1,1,3,0
4,29,1,1,3,0
4,53,2,1,3,0
5,66,4,4,3,1
4,60,3,?,4,0
5,76,4,4,3,1
4,58,2,1,2,0
5,96,3,4,3,1
5,70,4,4,3,1
4,34,2,1,3,0
4,59,2,1,3,0
4,45,3,1,3,1
5,65,4,4,3,1
4,59,1,1,3,0
4,21,2,1,3,0
3,43,2,1,3,0
4,53,1,1,3,0
4,65,2,1,3,0
4,64,2,4,3,1
4,53,4,4,3,0
4,51,1,1,3,0
4,59,2,4,3,0
4,56,2,1,3,0
4,60,2,1,3,0
4,22,1,1,3,0
4,25,2,1,3,0
6,76,3,?,3,0
5,69,4,4,3,1
4,58,2,1,3,0
5,62,4,3,3,1
4,56,4,4,3,0
4,64,1,1,3,0
4,32,2,1,3,0
5,48,?,4,?,1
5,59,4,4,2,1
4,52,1,1,3,0
4,63,4,4,3,0
5,67,4,4,3,1
5,61,4,4,3,1
5,59,4,5,3,1
5,52,4,3,3,1
4,35,4,4,3,0
5,77,3,3,3,1
5,71,4,3,3,1
5,63,4,3,3,1
4,38,2,1,2,0
5,72,4,3,3,1
4,76,4,3,3,1
4,53,3,3,3,0
4,67,4,5,3,0
5,69,2,4,3,1
4,54,1,1,3,0
2,35,2,1,2,0
5,68,4,3,3,1
4,68,4,4,3,0
4,67,2,4,3,1
3,39,1,1,3,0
4,44,2,1,3,0
4,33,1,1,3,0
4,60,?,4,3,0
4,58,1,1,3,0
4,31,1,1,3,0
3,23,1,1,3,0
5,56,4,5,3,1
4,69,2,1,3,1
6,63,1,1,3,0
4,65,1,1,3,1
4,44,2,1,2,0
4,62,3,3,3,1
4,67,4,4,3,1
4,56,2,1,3,0
4,52,3,4,3,0
4,43,1,1,3,1
4,41,4,3,2,1
4,42,3,4,2,0
3,46,1,1,3,0
5,55,4,4,3,1
5,58,4,4,2,1
5,87,4,4,3,1
4,66,2,1,3,0
0,72,4,3,3,1
5,60,4,3,3,1
5,83,4,4,2,1
4,31,2,1,3,0
4,53,2,1,3,0
4,64,2,3,3,0
5,31,4,4,2,1
5,62,4,4,2,1
4,56,2,1,3,0
5,58,4,4,3,1
4,67,1,4,3,0
5,75,4,5,3,1
5,65,3,4,3,1
5,74,3,2,3,1
4,59,2,1,3,0
4,57,4,4,4,1
4,76,3,2,3,0
4,63,1,4,3,0
4,44,1,1,3,0
4,42,3,1,2,0
4,35,3,?,2,0
5,65,4,3,3,1
4,70,2,1,3,0
4,48,1,1,3,0
4,74,1,1,1,1
6,40,?,3,4,1
4,63,1,1,3,0
5,60,4,4,3,1
5,86,4,3,3,1
4,27,1,1,3,0
4,71,4,5,2,1
5,85,4,4,3,1
4,51,3,3,3,0
6,72,4,3,3,1
5,52,4,4,3,1
4,66,2,1,3,0
5,71,4,5,3,1
4,42,2,1,3,0
4,64,4,4,2,1
4,41,2,2,3,0
4,50,2,1,3,0
4,30,1,1,3,0
4,67,1,1,3,0
5,62,4,4,3,1
4,46,2,1,2,0
4,35,1,1,3,0
4,53,1,1,2,0
4,59,2,1,3,0
4,19,3,1,3,0
5,86,2,1,3,1
4,72,2,1,3,0
4,37,2,1,2,0
4,46,3,1,3,1
4,45,1,1,3,0
4,48,4,5,3,0
4,58,4,4,3,1
4,42,1,1,3,0
4,56,2,4,3,1
4,47,2,1,3,0
4,49,4,4,3,1
5,76,2,5,3,1
5,62,4,5,3,1
5,64,4,4,3,1
5,53,4,3,3,1
4,70,4,2,2,1
5,55,4,4,3,1
4,34,4,4,3,0
5,76,4,4,3,1
4,39,1,1,3,0
2,23,1,1,3,0
4,19,1,1,3,0
5,65,4,5,3,1
4,57,2,1,3,0
5,41,4,4,3,1
4,36,4,5,3,1
4,62,3,3,3,0
4,69,2,1,3,0
4,41,3,1,3,0
3,51,2,4,3,0
5,50,3,2,3,1
4,47,4,4,3,0
4,54,4,5,3,1
5,52,4,4,3,1
4,30,1,1,3,0
3,48,4,4,3,1
5,?,4,4,3,1
4,65,2,4,3,1
4,50,1,1,3,0
5,65,4,5,3,1
5,66,4,3,3,1
6,41,3,3,2,1
5,72,3,2,3,1
4,42,1,1,1,1
4,80,4,4,3,1
0,45,2,4,3,0
4,41,1,1,3,0
4,72,3,3,3,1
4,60,4,5,3,0
5,67,4,3,3,1
4,55,2,1,3,0
4,61,3,4,3,1
4,55,3,4,3,1
4,52,4,4,3,1
4,42,1,1,3,0
5,63,4,4,3,1
4,62,4,5,3,1
4,46,1,1,3,0
4,65,2,1,3,0
4,57,3,3,3,1
4,66,4,5,3,1
4,45,1,1,3,0
4,77,4,5,3,1
4,35,1,1,3,0
4,50,4,5,3,1
4,57,4,4,3,0
4,74,3,1,3,1
4,59,4,5,3,0
4,51,1,1,3,0
4,42,3,4,3,1
4,35,2,4,3,0
4,42,1,1,3,0
4,43,2,1,3,0
4,62,4,4,3,1
4,27,2,1,3,0
5,?,4,3,3,1
4,57,4,4,3,1
4,59,2,1,3,0
5,40,3,2,3,1
4,20,1,1,3,0
5,74,4,3,3,1
4,22,1,1,3,0
4,57,4,3,3,0
4,57,4,3,3,1
4,55,2,1,2,0
4,62,2,1,3,0
4,54,1,1,3,0
4,71,1,1,3,1
4,65,3,3,3,0
4,68,4,4,3,0
4,64,1,1,3,0
4,54,2,4,3,0
4,48,4,4,3,1
4,58,4,3,3,0
5,58,3,4,3,1
4,70,1,1,1,0
5,70,1,4,3,1
4,59,2,1,3,0
4,57,2,4,3,0
4,53,4,5,3,0
4,54,4,4,3,1
4,53,2,1,3,0
0,71,4,4,3,1
5,67,4,5,3,1
4,68,4,4,3,1
4,56,2,4,3,0
4,35,2,1,3,0
4,52,4,4,3,1
4,47,2,1,3,0
4,56,4,5,3,1
4,64,4,5,3,0
5,66,4,5,3,1
4,62,3,3,3,0

mammographic_masses.names.txt

1. Title: Mammographic Mass Data

2. Sources:

   (a) Original owners of database:
        Prof. Dr. R�diger Schulz-Wendtland
        Institute of Radiology, Gynaecological Radiology, University Erlangen-Nuremberg
        Universit�tsstra�e 21-23
        91054 Erlangen, Germany
        
   (b) Donor of database:
        Matthias Elter
        Fraunhofer Institute for Integrated Circuits (IIS)
        Image Processing and Medical Engineering Department (BMT) 
        Am Wolfsmantel 33
        91058 Erlangen, Germany
        [email protected]
        (49) 9131-7767327 
        
   (c) Date received: October 2007
 
3. Past Usage:
    M. Elter, R. Schulz-Wendtland and T. Wittenberg (2007)
    The prediction of breast cancer biopsy outcomes using two CAD approaches that both emphasize an intelligible decision process.
    Medical Physics 34(11), pp. 4164-4172

4. Relevant Information:
    Mammography is the most effective method for breast cancer screening
    available today. However, the low positive predictive value of breast
    biopsy resulting from mammogram interpretation leads to approximately
    70% unnecessary biopsies with benign outcomes. To reduce the high
    number of unnecessary breast biopsies, several computer-aided diagnosis
    (CAD) systems have been proposed in the last years.These systems
    help physicians in their decision to perform a breast biopsy on a suspicious
    lesion seen in a mammogram or to perform a short term follow-up
    examination instead.
    This data set can be used to predict the severity (benign or malignant)
    of a mammographic mass lesion from BI-RADS attributes and the patient's age.
    It contains a BI-RADS assessment, the patient's age and three BI-RADS attributes
    together with the ground truth (the severity field) for 516 benign and
    445 malignant masses that have been identified on full field digital mammograms
    collected at the Institute of Radiology of the
    University Erlangen-Nuremberg between 2003 and 2006.
    Each instance has an associated BI-RADS assessment ranging from 1 (definitely benign)
    to 5 (highly suggestive of malignancy) assigned in a double-review process by
    physicians. Assuming that all cases with BI-RADS assessments greater or equal
    a given value (varying from 1 to 5), are malignant and the other cases benign,
    sensitivities and associated specificities can be calculated. These can be an
    indication of how well a CAD system performs compared to the radiologists.

5. Number of Instances: 961

6. Number of Attributes: 6 (1 goal field, 1 non-predictive, 4 predictive attributes)

7. Attribute Information:
   1. BI-RADS assessment: 1 to 5 (ordinal)  
   2. Age: patient's age in years (integer)
   3. Shape: mass shape: round=1 oval=2 lobular=3 irregular=4 (nominal)
   4. Margin: mass margin: circumscribed=1 microlobulated=2 obscured=3 ill-defined=4 spiculated=5 (nominal)
   5. Density: mass density high=1 iso=2 low=3 fat-containing=4 (ordinal)
   6. Severity: benign=0 or malignant=1 (binominal)

8. Missing Attribute Values: Yes
    - BI-RADS assessment:    2
    - Age:                   5
    - Shape:                31
    - Margin:               48
    - Density:              76
    - Severity:              0

9. Class Distribution: benign: 516; malignant: 445