fahimabrar/machine-learning

Python implementation of different Machine Learning Algorithms. Principal Component Analysis, DBSCAN, Kmeans, KNN, Tree, SVM, Hierarchical Clustering

Machine-Learning

This Repository Covers,

Unsupervised Learning

Principal Component Analysis

PCA is a technique for dimensionality reduction, where a reduced diemnsional data extract as much as insights possible from the higher dimensional data. After Extracting (2 and 3) Principal component from 30 columns we visualized it. Still the Malignent and Benign Cancer data are distinguishable form each other.

K means Clustering

Clustering country based on unlabelled data

DBSCAN

Hirerchical Clustering

Supervised Learning

Classification (Decision Tree, Random Forest, SVM, KNN)

we tried to predict if a student get placement after his/her study. The accuracy for different algorithms are,

• Decision Tree - 0.81
• Random Forest - 0.80
• K Nearest Neighbour - 0.80
• Suppor Vector Classifier - 0.78

Regression (Decision Tree, Random Forest, SVM, KNN)

We Tried to predict the salary of the students who got placed. The Mean Squared Error for different algorithms are,

• Decision Tree - 0.02
• Random Forest - 0.019
• K Nearest Neighbour - 0.016
• Suppor Vector Classifier - 0.012