Projects

This is a rest application to manage the organization of events such as seminars, lectures, and what have you. The application is a collection of APIs that has been published via postman and deployed on Heroku.

The application was built using Spring Boot a Java Framework and a"dockerised" Postgres for persistence of data

This is a Spring Boot application used to manage a pet store.

It allows for various CRUD operations such as add/delete/update/retrieve a pet and store.

I t was built using Spring Boot, Hibernate and MySQL

I worked with the PyMC library and Priesman group library to build a Bayesian model to forecast the spread of covid-19 in Ghana. The mathematical framework upon which the model was built is the SE(I)R epidemic model; Susceptible-(Exposed)-Infected and Recovered.

The output of this project was a report on the analytics and modelling task.

I built a supervised machine learning predictive model to classify customers who are likely purchase a bank's term sheet.

The data was an imbalanced one hence I had to apply Synthetic Minority Oversampling Technique to the minority class using the Imblearn library.

I used various algorithms in building the model; Logistic regression, SVM, Random Forest Classifier and Catboost. I used GridSearch for the hyper-parameter tuning and selected the best model with the lowest logloss. Catboost model performed better than the rest of the models

I also performed exploratory data analysis on the data with Tableau

A Machine Learning Web App that predicts sales and customer churn for Rossman pharmaceuticals

The Web App was built for users to be able to interact with the model that was built and make predictions over the web.

The Web App was built with Flask; a Python framework and HTML/CSS

The model was built to solve a regression and time-series problem. Link to the code on GitHub can be found here.

In this project, I applied my skills in data modeling with PostgreSQL and build an ETL pipeline using Python. In completing the project, I defined fact and dimension tables for a star schema for a particular analytic focus, and wrote an ETL pipeline that transfers data from files in two local directories into these tables in PostgreSQL using Python and SQL.