Letter recognition using diff ML algorithms - use of bagging and boosting methods

ML models for HR classification problem. For more information please visit the link: https://datahack.analyticsvidhya.com/contest/wns-analytics-hackathon-2018-1/

Implementation of bagging-based ensemble for solar irradiance prediction. Base learners used in ensemble learning is stacked-LSTM

Python Notebook for Ensemble Learning on Wisconsin Breast Cancer Database

Given a string statement, the aim is to identify the B-Protein entity in the statement.

Application of various text classification algorithms on multiple datasets.

Notebooks on ensemble techniques.

Ensemble model which uses supervised machine learning algorithm to predict whether or not the patients in the dataset have diabetes

Reducing imbalanced dataset (Undersampling) by Consensus Clustering (Simple Majority Voting function) and validating the changes using different classifier model with bagging and boosting techniques.

This project explains why and how are the Bagged Models better than the Complete Model. Bagged Model parameters have tighter confidence interval and a lower bias.

churn prediction for telecom company

Implementation of Machine Learning Projects

Comparing the perfomance of different ensembling and feature engineering techniques to predict the outcome of a game of chess.

Machine Learning algorithms from-scratch implementation. It covers most Supervised and Unsupervised algorithms. Homework assignments and Projects for graduate level Machine Learning Course taught by Dr Manfred Huber at UTA during Spring 21

Using machine learning to detect whether an email is a scam or not.

Optimizing LinearSVC models trained on the MNIST Handwritten Digits dataset, includes ensemble methods and bayesian optimization.

The objective of this project is to build a “Risk Analytics model” to understand the renewal potential and claim propensity of Existing Customers under Personal Auto Insurance Lines. The final models for claims and renewals were able to accurately identifies over 70% of claims & 85% renewals.

In this project, we reduced an imbalanced dataset to a balanced dataset using Under-sampling approach by applying Consensus Clustering using 'Simple Majority Voting' consensus function and further saw the increase in the accuracy of disease prediction by running multiple classifiers with bagging and boosting technique.

https://teacher.bupt.edu.cn/zhuchuang/en/index.htm