#2020-07-18 12:51:41
kNN is so simple to understand. It’s the beauty of it! When you need an intuitive, accurate Machine Learning algorithm without all the bells and whistles, k Nearest Neighbors is ready to be your bestie.
Since it’s so unique and special it’s also a very welcome fact that it can thrive doing both classification and regression. Yay!
#SUPERVISED
#CLASSIFICATION
#REGRESSION
K-nearest Neighbors
- Dataset: Large and Small
- Speed: Slow
- Ease of Use: Easy
- Normalization: Yes
- Predictor: Numeric
- Primary Problem: Multiclass or Binary
- Mixed-type: No
- Missing Data Handling: No
- Popularity: 70%
k-Nearest Neighbor tutorials in python
1- Simple Implementation
When the inspiration knocks your door, don't wait. Act immediately!
Here is a very simple kNN implementation which everyone can understand and use immediately, right now.
2- Step by Step
In case you are not clear with any part of the simple kNN implementation, here is a step by step walk-through of the fundamental pieces of a kNN algorithm application in scikit-learn.
3- Optimization
Once you understand the basics, it can make a difference to know and understand the different parameters for fine tuning your data science tasks. Here are the most common kNN parameters for optimization, fine tuning, improved results and efficient machine learning implementations.
4- Pros & Cons
Knowing the Pros & Cons can mean the difference between right decision and wrong decision. Don't worry, we have prepared straightforward guides for your a quick learning. Here are all the advantages and disadvantages of a kNN algorithm fairly laid out.
5- History
Read about the history of k-Nearest Neighbor Algorithm.