Power of Ensemble models



This session will brief you about the significant Machine Learning topic and give great insights towards each of the techniques. Ensemble techniques are being leveraged predominantly during Hackathons as it is a powerful method to build the model.

Choosing the right ensembles is more of an art. You will understand the nuances about which ensemble learner to be used in different kinds of scenario as you solve more problems. Ensemble modeling can exponentially boost the performance of your model.

Here, we will cover various ensemble learning techniques and execute few problems to get to know, how these techniques are applied in machine learning algorithms.


• Basics of Ensemble Learning

          Why Ensemble Models?
          Predictive power 
          Significant use cases

• Ensemble Techniques

          Basic - Max Voting/Averaging/Weighted Average
          Advanced -Blending/Stacking/Bagging/Boosting

• Various Algorithms

• Pros and cons of Ensembles

• Ensemble models using Python


Basics of Programming and Interested in deriving insights from Data

Content URLs:

Content urls: Slides as pdf



My short video of the presentation: https://www.youtube.com/watch?v=9BVU-2fKsCQ&feature=youtu.be

Speaker Info:

Padmapriya Mohankumar, Sr. Technical Product Manager at PayPal.

She is a Tech Evangelist with demonstrated expertise in diverse stack of Data and Business Intelligence Products, Data Warehousing, Tech Architecture, Big Data adoption solutions, Machine Learning, Analytics and Governance. She has been a Technology Influencer of various significant project implementations for Cisco, Apple, Bank of America and spearheaded many significant projects for them in the past. She is highly passionate about the emerging trends and building futuristic products. She is a Computer Engineering Graduate from Vellore Institute of Technology and currently pursuing Post Graduate program in AIML

Speaker Links:

Speaker Links



Id: 1240
Section: Data Science, Machine Learning and AI
Type: Talks
Target Audience: Beginner
Last Updated: