10. List the characteristics of k-nearest neighbour algorithm.

 The k-Nearest Neighbors (k-NN) algorithm is a popular supervised machine learning algorithm used for classification and regression tasks. Here are some key characteristics of the k-NN algorithm:

Instance-Based Learning:

k-NN is an instance-based learning algorithm. It does not explicitly learn a model during the training phase but memorizes the entire training dataset.

Lazy Learning:

k-NN is considered a lazy learning algorithm because it defers the processing of training data until the prediction phase. It doesn't generalize a model from the training data; instead, it stores the training instances and makes predictions based on the nearest neighbors during testing.

Non-Parametric:

k-NN is a non-parametric algorithm, meaning it makes no assumptions about the underlying distribution of the data. It directly uses the training dataset for predictions.

Classification and Regression:

k-NN can be used for both classification and regression tasks. In classification, the majority class of the k-nearest neighbors is assigned to the test instance, while in regression, the average or weighted average of the k-nearest neighbors' values is used as the prediction.

Distance Metric:

The choice of a distance metric (e.g., Euclidean distance, Manhattan distance, Minkowski distance) is a critical aspect of k-NN. It determines how the "closeness" of neighbors is measured.

Hyperparameter k:

The parameter 

�

k represents the number of nearest neighbors considered during the prediction. The value of 

�

k is a hyperparameter that needs to be specified by the user. A larger 

�

k makes the model less sensitive to noise but may lead to smoother decision boundaries.

Voting Mechanism:

For classification, k-NN uses a voting mechanism to determine the class of a test instance. The class with the majority of votes among the k-nearest neighbors is assigned to the test instance.

Local Decision Boundaries:

k-NN creates local decision boundaries based on the distribution of training instances. This can result in decision boundaries that follow the intricacies of the data, making the algorithm suitable for complex and non-linear datasets.

Sensitive to Irrelevant Features:

k-NN is sensitive to irrelevant or redundant features. Feature scaling is often necessary to ensure that all features contribute equally to the distance computation.

Computational Cost:

The computational cost of k-NN can be relatively high, especially for large datasets, as it involves calculating distances between the test instance and all training instances. Efficient data structures like KD-trees or ball trees can be used to speed up the search for nearest neighbors.

Understanding these characteristics is important when deciding whether to use the k-NN algorithm for a particular task and how to fine-tune its parameters for optimal performance.