K-Means & KNN in my life ????....

KNN (K-Nearest Neighbors)

KNN is a supervised learning algorithm, which is used for classifying the data based on its neighbors. K denotes number of neighbors.

To classify a datapoint based on its neighbor, we need to select k neighbors and calculate the distance between the neighbors and the required datapoint to be classified. Eventually, finding the neighbor with the nearest distance, we will classify the data point.

Note: In this, the classification is done based on the labeled data(i.e., labeled neighbors).

Uffff ?? We all know this, but what's that in your life ??? this is the question right !!!

This is the answer to your question and actually, this is in every one of your life dudes... ??

Eg: people decide one person based on his/her friend circle ?? and even sometimes, decide based on the friend among those friends group they move so close with.

Hope you got the answer, I think this happens with every one !!!

K-Means

K-Means is an Unsupervised machine learning algorithm which is used for clustering datapoints based on similarities without any labels provided. K denotes number of clusters.

To makes K number of clusters, we need to follow the following steps :

1. Choose number of clusters(i.e., K values)
2. Initialize K number of centroids
3. Based on the nearest distance from the centroid to the data point, assign the data point to the cluster.
4. Update the centroid by calculating the man of the datapoints within the cluster.
5. Repeat the above steps until all the datapoints are assigned to the clusters.

In this way, the distance is like similarities among the data points. Less the distance, the more the similarity, and vice versa.,

Its okk, we do even know this process even. What does it do in your life ????This is your question right!! Then ,this is my answer for you.

Eg: I have a basket of fruits which contains some green apples, guavas and some avocados. I am a small kid and don't know their names, but can identify the difference in their view. Now,I found 3 kinds of fruits and pick each of then(Step 1). I decide them as centroids (Step 2)and start finding similar fruits as the chosen centroids(Step 3). Sometimes, the fruit size may be small, but the fruit is the same, so I will know that some fruits are there which are similar to the centroid, but different in size. so, I update myself with this new information(Step 4) and start clustering until all the fruits in the basket are assigned to the required cluster(Step 5).

Not only this, but , if you could observe, in online shopping platforms, the clustering of the clothes based on the fabric is also an example.

The definition is the same everywhere, but understand it through your daily life routine.

Suspense is??That student from 2020's is me ????