Binary Classification: Programming Exercise

For this implementation uses a decision tree classifier to predict the label of a fruit based on its color and weight. The data is stored in a pandas DataFrame and split into training and testing sets. The classifier is trained on the training data, and its accuracy is tested on the testing data.

Remember to encode the variable 'color' using, for example, 'get_dummies'. The models only understand numbers, not words or strings.

Split the dataset by using the function train_test_split(). You need to pass 3 parameters features, target, and test_set size. Split the dataset 30% test and 70% train and set using a random state=0. Store output of the split in X_train, Y_train, X_test and Y_test.

Then implement the decision tree using only random_state=0. Store the model in the variable cf, and finnaly estimate the accuracy of the model using the test dataset.

# Create the data for the fruit classifier
data = {'fruit': ['apple', 'banana', 'apple', 'banana', 'banana','apple','apple','apple'],
        'color': ['red', 'yellow', 'green', 'yellow', 'yellow','green', 'green', 'red'],
        'weight': [200, 100, 150, 90, 85,95,99,102],
        'label': [0, 1, 0, 1, 1,0,0,0]}
df = pd.DataFrame(data)
df.head()

The results of the accuracy calculation should be stored in the variables train_accuracy and test_accuracy for the training and testing sets, respectively.

We continue working with previous dataset, but for this task train a KNN classifier to predict the label of a fruit based on its color and weight.

Split the dataset by using the function train_test_split(). You need to pass 3 parameters features, target, and test_set size. Split the dataset 30% test and 70% train and set using a random state=0. Store output of the split in X_train, Y_train, X_test and Y_test. Remember that KNN is sensitive to the scale of the features, so you should use StandardScaler to standardize the features and store the results in the variables X_train_scaler and X_test_scaler.

Then implement the KNN using the argument for default. Store the model in the variable knn, and finnaly estimate the accuracy of the model using the test dataset.

# Create the data for the fruit classifier
data = {'fruit': ['apple', 'banana', 'apple', 'banana', 'banana','apple','apple','apple'],
        'color': ['red', 'yellow', 'green', 'yellow', 'yellow','green', 'green', 'red'],
        'weight': [200, 100, 150, 90, 85,95,99,102],
        'label': [0, 1, 0, 1, 1,0,0,0]}
df = pd.DataFrame(data)

The results of the accuracy calculation should be stored in the variables train_accuracy and test_accuracy for the training and testing sets, respectively.

What are the advantages of the decision tree?

Choose the correct statement from below.

A student has a dataset with 500 data points that he want to use to train a KNN classifier. He trains 4 kNN classifiers (k={1,3,5,10}) using all the data points. Then you randomly select 300 data points from the 500, and classify them using each of the 4 classifiers.

Which classifier will come out as the best one?

Based on the following figure, identify the class of the black point if you train a K-NN algorithm with k=2.

Choose the correct statement from below.

For this task, we will use the following simulated dataset to train a decision tree and KNN models.

The data consists of information about 20 individuals, including their age, income, student status, and credit rating. The target variable, class, indicates whether an individual earns more or less than 50,000 a year (1 for more, 0 for less).

# Load the sample data
data = pd.DataFrame({'age': [23, 25, 22, 21, 24, 26, 20, 22, 19, 23, 25, 27, 21, 24, 22, 25, 26, 29, 31, 28],
                     'income': [50000, 60000, 55000, 65000, 65000, 70000, 45000, 62000, 48000, 50000, 67000, 72000, 49000, 55000, 65000, 62000, 72000, 75000, 85000, 90000],
                     'student': [0, 0, 1, 1, 0, 0, 1, 0, 1, 1, 0, 0, 1, 0, 1, 1, 0, 0, 1, 1],
                     'credit_rating': [620, 630, 600, 675, 635, 700, 625, 650, 575, 645, 725, 675, 550, 575, 600, 650, 720, 775, 800, 850],
                     'class': [0, 0, 0, 1, 1, 0, 1, 0, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1]})

The data is split into training and testing sets, and a decision tree and a KNN classifier were trained on the training data. The accuracy of the classifiers were evaluated on the testing data using the accuracy score and confusion matrix.

# Split the data into training and testing sets
X = data.drop(["class"], axis=1)
y = data["class"]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=0)

Your objective is to identify which models performed best in terms of the evaluation metric.