## Precision-Recall Curve for Classification model analysis

Accuracy is something which gives an intuition of model performance. i.e. ratio of number of correct predictions with respect to total sample present. But what in case of unbalanced data.Imagine a case where we need to make a model to predict click through rate over a display rate. The click trough rate used to be […]

## The ROC – AUC curves for Classification based Model Performance Analysis

Accuracy is something which gives an intuition of model performance. i.e. ratio of number of correct predictions with respect to total sample present. But what in case of unbalanced data.Imagine a case where we need to make a model to predict click through rate over a display rate. The click trough rate used to be […]

## Logistic Regression using Python

Logistic Regression is a method to create Machine Learning model for two class problems. It came out of linear regression but used to generate binary output (0 and 1) for making classifications. For example In Linear Regression we use simple linear equation as follows :- Yh = b0 + b1X1 Where X combines linearly with […]

## KNN Algorithm and Maths behind

KNN or K – Nearest Neighbours is one the powerful algorithm used in classification based problems to successfully make categorical predictions. Scikit-Learn gives us built in library to use and make the process easier for us if we are having data. But here we will write KNN code mathematically without any inbuilt library to figure […]

## Mean, Variance, Standard Deviation, Standard Score, Covariance & Data Projection

Variance – It is the measure of squared difference from the Mean. To calculate it we follow certain steps mentioned below: Calculate average of numbers For each numbers subtract the mean and square the result Calculate the average of those squared differences i.e. Variance Editorial Team

## Eigenvector represents greatest variance in case of PCA

In case of Principal Component Analysis we project our data points on a vector in a direction of maximum variance to decrease the number of existing components. In this case we consider the direction eigenvector generated using covariance matrix as the direction of maximum variance. In this article we look into the proof of why […]

## Eigen Vector and Eigen Values

Eigen vector is the direction in a coordinate space defined by a metrics which doesn’t change its direction with metrics transformation. Eigen value is a scaler number which is multiplied with Eigen vector to give same result as Eigen vector multiplier with existing metrics. Editorial Team