Posts

Introduction According to the Wikipedia, “Anomaly detection (also referred to as outliers detection and sometimes as novelty detection) is generally understood to be the identification of rare items, events or observations which deviate significantly from the majority of the data and do not conform to a well defined notion of normal behavior.

Introduction We all heard about the Central Limit Theorem (CTL) in out basis statistics class. It is one of the fundamental theorem in statistics and is applied in numerous analytics and statistics analysis.

Introduction Gradient Boosting (GB) is one of the most popular algorithms in the Machine Learning/Data Science community. You might heard about various version of boosting algorithms like Ada-boost, XG Boost, LG Boost, CatBoost.

Stationarity is one of the major concept that needed to be understood in order to model any time series. The Wikipedia defines stationary process as “a stochastic process whose unconditional joint probability distribution does not change when shifted in time.

The class imbalance problem is very common scenario that we encounter in various domains like Churn Analysis in HR, Fraud Analysis in Finance, Customer Retention in Marketing. Class Imbalance simply means that in our multi-class target variable the proportion are not equally close to each other.

The notion of Regularization arises because of overfitting. So, let’s start with overfitting then. I will keep aside all fancy definition. Overfitting simply means when your model does good on the training data but performs poor on the testing data.

In my last post, I described the math behind the Gradient Descent. In there, I explained about the definition, working principles and how does it works in terms of Multiple Linear Regression.

Gradient descent is one of the most popular optimization algorithms in the data science and machine learning. For definition, as always, I investigated the Wikipedia. It says, “In mathematics gradient descent (also often called steepest descent) is a first-order iterative optimization algorithm for finding a local minimum of a differentiate function.

We all have heard about the Linear Regression from our basic statistics class or some newspaper or somewhere here and there. What I mean to say is that Linear Regression is one of the most common algorithm that we will see in many kinds of predictive analysis as well as finding linear relationship between the dependent and independent variables.

In this article we will discuss the basic of the data manipulation – slicing and selecting. As we have seen, majority of time, we need to work with a subset of the data for the analysis.