Insight - The Data Science Society of IMI, New Delhi

10/05/2020

Random Forest can be said to be the most powerful supervised machine learning algorithm. Microsoft used it in X-box to develop posture sensitive games. Here, a Random Forest Classification has been explained in detail and a model has been developed using a bank dataset where we have the details of customers who bought a credit card from the bank.

https://insightimi.wordpress.com/2020/05/09/developing-random-forest-classification-in-r/

In supervised machine learning algorithms, Random Forest stands apart as it is arguably the most powerful classification model. When Microsoft developed their X-box game which enables you to play a…

27/04/2020

Learn to develop SVM algorithm and Solve a Real-World Business Problem by implementing this classification algorithm in R.

https://insightimi.wordpress.com/2020/04/25/svm-to-solve-the-business-problem-in-r/

SVM has been explained from the ground up in a simple and intuitive manner with graphical representations of the hyperplanes, multi-dimensional datasets, and how the kernel approach works. All this has been beautifully explained by following step by step approach to divide the complex parts into subparts and covering them one at a time.

For the problem faced by the business, we try to find the solution by implementing the SVM algorithm in R. And in this process you also get to know how to logically approach a problem as a data scientist would do. We also covered the oversampling technique called SMOTE, various methods of data pre-processing and how to do a preliminary analysis of the data.

Support Vector Machine (SVM) is a supervised machine learning algorithm which can be used for both classification or regression challenges. However, it is mostly used in classification problems. S…

19/04/2020

Hierarchical Clustering - Everything you need to know about it including step by step algorithm and implementation in R.

https://insightimi.wordpress.com/2020/04/19/hierarchical-clustering-from-scratch-in-r/

Hierarchical Clustering has been explained with an example of how types of seeds can be segmented by similar attributes (like area, perimeter etc). While implementing the algorithm, we have not only used the traditional Dendograms approach, but also used heat maps, tanglegrams and correlation maps to give a better understanding and beautiful representation of clusters and their visualization.

It will also give you a better understanding of various metrics, linkage methods, evaluation indexes and how to use them efficiently.

Hierarchical Clustering – Everything you need to know about it.

12/04/2020

K-Means Clustering - Everything you need to know about it including step by step algorithm and implementation in R.

https://insightimi.wordpress.com/2020/04/12/unsupervised-learning-k-means-clustering/

In this blog, K Means Clustering has been explained with an example of how customers can be segmented by similar attributes (Taking income and their spending score). K-Means algorithm is also popular with fraud detection, making profitable portfolios, documents classification and more.

This K-Means example will particularly be helpful for all the marketeers wanting to segment huge chunks of data to gain valuable insights.

Everything you need to know about K-Means Clustering

05/04/2020

Naïve Bayes classifiers are a family of simple "probabilistic classifiers" based on Bayes' theorem .It works on the assumption of independence of features
to classify the dependent variables. In this blog, we are representing a comprehensive view of naïve Bayes classifier from scratch with a hands-on example in R.

This post covers the naïve Bayes model, its functionality, and a hands-on example in R programming language.

Do give it a read!

https://insightimi.wordpress.com/2020/04/04/naive-bayes-classifier-from-scratch-with-hands-on-examples-in-r/

Naïve Bayes classifiers are a family of simple “probabilistic classifiers” based on applying Bayes’ theorem with strong independence assumptions between the features. In this post…

29/03/2020

C4.5 is the most widely used Decision Tree Algorithm. It is a successor to ID3 algorithm, and gets rid of the bottlenecks of ID3. This post will take you deep into how C4.5 works and how it is better than ID3. Furthermore, we have talked about C5.0 as well, which is the successor to C4.5.

This post covers the C4.5 algorithm, C5.0 algorithm, and gives a comparative analysis of the most popular Decision Tree Algorithms.

Do give it a read!
https://insightimi.wordpress.com/2020/03/29/c4-5-in-detail-and-comparative-analysis-of-decision-tree-algorithms/



.5 .0

Previously, we have talked about 2 of the Decision Tree Algorithms: 1. Gini Index (while implementing CART – Classification) 2. ID3 – Iterative Dichotomiser 3 (while implementing CART –…

22/03/2020

Logistic regression is used to predict a categorical outcome based on variables which may be discrete, continuous or mixed. Thus, when the dependent variable has two categorical outcomes, logistic regression is a commonly used technique. In this blog, we are representing a comprehensive view of logistic regression from scratch with a hands-on example in R.

This post covers the Logistic Regression model, its functionality, and a hands-on example in R programming language.

Do give it a read!

https://insightimi.wordpress.com/2020/03/22/logistic-regression/

Introduction Researchers are often interested in setting up a model to analyse the relationship between some predictors (i.e., independent variables) and a response (i.e., dependent variable). Line…

21/03/2020

We have always been asked as to who are we and what do we stand for. This is our brief introduction.

15/03/2020

Regression Trees are fundamental decision-making models which are used by companies to determine the best-fit salary based upon several parameters, companies to determine the price of products and many more things.

Greetings from Insight - The Data Science Society of IMI, New Delhi

This week's post covers Regression Trees the second part of CART (the first having being posted last week)

https://insightimi.wordpress.com/2020/03/15/cart-regression-tree-from-scratch-with-a-hands-on-examplein-r/

Do give it a read!



#
#

Decision trees are made with the objective of creating a model that predicts the value of a target or dependent variable based on the values of several input or independent variables. The CART algo…

08/03/2020

Greetings from Insight - The Data Science Society of IMI, New Delhi!!!

Classification and Regression Trees (CART) is the fundamental decision tree algorithm in which powerful models like Random Forest, Bagged or Boosted Decision Trees are built.

This blog covers the Classification part of it.

To study more about what is CART is all about, how it is used to make decisions given certain criterias, visit the blog.

https://insightimi.wordpress.com/2020/03/08/classification-and-regression-tree/

The blog aims at explaining the Classification part of the CART algorithm in as much detail as possible. Give a read to gain a real, hands-on experience on a real-life data.

01/03/2020

K Nearest Neighbours algorithm is one of the simple, easy to understand and topmost machine learning algorithms that has it's applications in different branches.

To study more about what KNN Classification is all about, how it is used in the field of Finance, Healthcare, Image recognition and many others, visit the blog.

https://insightimi.wordpress.com/2020/03/01/k-nearest-neighboursk-nn-algorithm-from-scratch-with-a-hands-on-example-in-r/

Part A: K-NN Algorithm from scratch An Introduction: Classification algorithms have developed over time from basic models to complex ones providing much more flexibility in dealing with the data an…

22/02/2020

Insight is here and again with its weekly edition of the tutorial.
Presenting Linear Regression on Cross-sectional Data
https://insightimi.wordpress.com/2020/02/23/linear-regression-for-cross-sectional-data/

Linear Regression is one of the 1st algorithms that everyone learns in Machine Learning, Statistics, Financial Econometrics and Data Science. In this post you will learn and go through Types of Dat…