Heart failure predictor

This is project is a machine learning classification problem that aims to predict if a patient is prone to heart failure or not. Several machine learning models were trained for this project and bagging classifier had the best performace of amongs the model trained.

The following task were carried out in this project:

• Summary statistics to understand the data better.
• Data visualization of the classes of each feature in the datset to gain insight on how balanced and useable the dataset will be for building a machine learning model.
• Feature selection using Eli5 and mutual info classifier.
• Model development was done with flask and was deloyed to heroku.

View Live WebApp

U.S adult Salary Predictor

This is a classification machine learning project that predicts if an adult in the U.S earns less or more than $500.

The following task were carried out in this project:

• Data cleaning and outlier removal.
• Building of pipeline to codify and automate the ML workflow.
• Hyper parameter tuning to optimive the machine learning model for better performance.
• Integration of the built models into a webapp built with streamlit and deployed on streamlit server.

View Live WebApp

Analysis of customer Churn of a Telco Company

Telco is a company that provides home and internet service to 7043 customers in califonia.Data was provided by them on customers that have left their company(churned).Using this data we want to gain insight into the factors thats driving customer churn,which will help the company understand how to drive campaigns on customer retention and decision making to aid in reduction customer churn.

The following task were performed using the data:

• Data cleaning
• Exploratory data analysis to answer some business problems.(Full description to each step is on the python notebook containing this project on github).
• Visualizations.
• Building and evaluation of several machine learning models to predict customers that are likely to churn.

Exploratory Data Analysis on Amazon Books Dataset Using SQL

Amazon books is a chain of retail bookstores owned by online retailer Amazon. This projects data was gotten from kaggle and contains data of all the top 50 sold books from 2009-2019.Making it a total of 550 books.The data was cleaned using python and queried using SQL

The data was queried using SQL to answer the following questions:

• Authors with the highest number of best selling books between (2009-2019)
• Most expensive best selling book in amazon between(2009 - 2019)
• Average price of amazon best selling books
• What genre of book sold the most?
• Best selling book between the yaer(2009 - 2019)
• Average price of books by genre?

SuperStore Sales Dashboard

Dashboards shows all performance of a business at a glance.it is a collection of several views letting us compare a variety of data simulteneously.

This interactive dashboard was built with the aim of answering and visualizating the following business questions:

• KPI of customers,sales,average shipment time and total quantity.
• Total sales made.
• Sales per region.
• Sales per state.
• Sales growth per year(Quaterly).

View Dashboard

Nothing to see here

Lorem ipsum dolor sit amet, consectetur adipisicing elit. Mollitia neque assumenda ipsam nihil, molestias magnam, recusandae quos quis inventore quisquam velit asperiores, vitae? soluta, eos quod consequuntur itaque. Nam.