In this project I will be attempting to analyse how different music artists in Kenya are performing on youtube, spotify, apple music and boomplay. I plan on getting the relevant data by utilising the APIs of the different platforms. Currently I have been able to pull data from youtube. Some metrics I will be looking at will be total monthly plays, likes to dislikes ratio and subscriber growth. After completing the data gathering and analysis scripts, I plan on releasing monthly updates on how these artist have been perfoming. Stay tuned as many great things are about to happen in this project.

The aim of this project was to rank different twitter users according to their influence. The influence was calculated using different metrics eg follower count and retweets. The usernames were scraped from ... the tweets were collected using scrapy. A report of my findings can be found here.

This project can be extended to a companies digital campaigns so as to know which users can be beneficial to raising awareness about the brand.

In this project I applied machine learning algorithims to find out which customers would subscribe to a particular service offered by a Portugeese bank. This will assist the bank to focus more on customers who are more likely to subscribe to this service hence saving time and resources for the bank

In this project I had to analyse the sales of Rossman group a pharmaceutical company and create a model that could predict the sales for six months. The model was hosted on heroku and we created a dashboard that summarised the sales in their different stores

In this project, I analysed moodle logs of students and created a tableau dashboard to deliver insights. The dashboard can be found here and also a report of my findings can be found here

The main objective of this project was to analyse the effect of COVID 19 on food quantity in Africa. I did this project aspart of a team of 5 talented data scientist. We focused on the following countries South Africa, Nigeria and Kenya. The data was collected from twitter and we analysed it by looking at which tweets had the most retweets and also we analysed the sentiments of all the tweets. A report of our findings can be found here.

The objective of this project is to develop a multi-class classification model to classify news content according to their specific categories specified.The model can be used by Swahili online news platforms to automatically group news according to their categories and help readers find the specific news they want to read.

In this competition, I leveraged data and ML methods to improve market outcomes for insurance provider Zimnat, by matching consumer needs with product offerings in the Zimbabwean insurance market. Zimnat wants an ML model to use customer data to predict which kinds of insurance products to recommend to customers. The company has provided data on nearly 40,000 customers who have purchased two or more insurance products from Zimnat.

The challenge: for around 10,000 customers in the test set, you are given all but one of the products they own, and are asked to make predictions around which products are most likely to be the missing product. This same model can then be applied to any customer to identify insurance products that might be useful to them given their current profile.