Jayasurya
Tumurouthu

Data Analyst well versed in SQL, Python, Power BI, Excel

Customer Churn Analysis using PowerBI

It is beneficial for banks to understand what motivates a client to leave the company. Churn prevention enables businesses to create loyalty programmes and retention campaigns in order to retain as many customers as possible. This visualisation was created with PowerBI, and various features such as slicers, actions, intellisense, powerpivot, powerquery, and powermodelling were used.

Predicting number of retweets
based on total followers

In this project, we made use of the 'tweepy' package and api's to extract the relevant data from the twitter database. A model to estimate the number of retweets based on the number of followers has been developed by performing EDA initially, followed by training and testing with a decision tree classifier algorithm.

Predicting Heart Disease based on symptoms

We conducted exploratory data analysis on components in this research that may be suspected of causing heart disease in the near future. A model has been created based on these variables to determine whether all of these symptom-inducing elements can eventually result in the development of heart disease.

Sentimental Analysis
British Airways

In this virtual work experience programme, web scraping is done on the skytrax website, which contains airline reviews. The data pertaining to British Airways was scraped and cleaned in preparation for further analysis. It is concluded that 60% of the reviews are positive, 39% are negative, and the remainder are neutral.

Predicting Diabetes risk based on symptoms

This project puts to the test several models that are good at predicting the likelihood of developing diabetes in the near future. Several features are considered, and EDA is performed initially. Different classification algorithms were tested, and the MLP classifier was found to have the highest classification accuracy.

Bike Sales Dashboard using Excel

This project contains the Bike Sales Dashboard, which was created entirely in Excel. After cleaning the data by removing duplicates, several operations such as IF,WHATIF, SUMIF, and VLOOKUP are used to get the most out of it. Pivot Tables are used to generate more accurate summaries based on gender, and a dashboard is created at the end with slicers and interactions between various graphs.

BellaBeat Fitness Visualization using Tableau

This Dashboard displays several features through the use of various charts, with actions between the charts incorporated to show how different variables correlate. This is entirely based on Tableau of Fitness Tracker App data in order to gain insights into how consumers use the FitBit app and discover trends and insights for the Bellabeat marketing team.