Proposed a model that helps to predict the churning of customers based on the analysis of their historical data using Machine learning algorithms

1. Performing data exploration, data cleaning, feature engineering, feature elimination and built various machine learning models on the dataset.
2. Evaluating models by comparing the accuracy score. Created a single page dashboard using PowerPoint showcasing all the analysis important for the business.

Designed machine learning models - Linear Regression, RandomForest, XGBoost and LightGBM for predicting the price for Airbnb listing using Python from the ground up. Enhanced model performance by performing Python hyperparametric optimization by over 19%. Performed deep end-to-end analysis, applied statistical concepts and performed data visualization using Python and Tableau. Visualized a dashboard in Tableau for the business containing factors impacting the listing price. Evaluated the performance of models based on their RMSE values and the best performance was given by the LightGBM model having an RMSE value of 0.4491.

Implemented regression model and decision tree algorithm to analyze and predict Customer churn in the telco up to an accuracy of 94.75% using Python. Implemented an End to End Analytics Lifecycle including data mining, data cleansing, analysis, feature importance score calculation and feature reduction to visualization. Designed a machine learning model for significant variables and evaluated the performance of the models using accuracy score code.

Designed a fully automated prediction and risk-scoring platform for text mining of global news articles using text mining techniques.

Performed sentiment analysis of Twitter data for political parties using R. Denoising, categorizing and classifying sentiments into positive, negative and neutral. Created frequency charts, word clouds, identified common words. Applied statistical modelling techniques to develop and evaluate algorithms for a dataset using R.