This repository contains programs and dataset used for developing Machine Learning Models for learning purpose. I have developed some of the initial level models which are listed listed below:
- Linear Regression: It is a supervised Machine Learning model used for prediction of a event or data based on past data na dis one of the most common machine learning models for prediction.
- Logistic Regression: It is also a supervised Mqachine Learning model but as per its name it is not used for regression rather it is used for classification of binary class or multiclass data. It is one of the basic classification models.
- KNN(k-Nearest Neighbor): KNN is a supervised machine learning algorithm which is used for clasification based on the similarity of a particular point. It is effective with small datasets but need other algorithms for classification for large dataset.
- Decision Tree: Decision Tree is a supervised Machine Learning algorithm which chooses the important parameters from the dataset and tries to reduce the entropy and increase the information gain at a particular split and if results are not satisfactory then it changes the order of parameters which are contributing to dataset and creates a new tree and this process keeps on going till algorithm converges.
- K-Means: K-Means is an unsupervised Machine Learning algorithm used for clustering of data. This model assumes different centroids and tried to separate the data based on similarity and dissimilarity of nearby points. The centroids keep on changing till every point in dataset is considered as a centroid and finally the centroids which provides best separation of data are the final points and a new point can be part of a cluster based on its similarity with its nearby clusters.