1.The EY AI & Data Challenge Program(ZINDI.AFRICA)-Frog Presence prediction
The goal of the challenge is to build a machine learning classification model to predict the presence of frog species based on TerraClimate variables extracted from the Microsoft Planetary Computer data catalogue.
2. Yield Estimation project:PULA.IO
This involved data for a location in Nigeria where the objective was to perform a data analysis during a planting to the harvest season using box placement data to perform a yield prediction of the area a year later and to a larger irrigation area. Used excel for data cleaning, Tableau for visualization and ArcGIS for generation of Spatial distribution maps.
4.Electricity Consumption project:PULA.IO
This involved prediction of electric power consumption analysis for a power supply company where date ,hour of the day, power supplied and power consumption for a location was used to predict the amount of power to be supplied for that region. In this scenario I employed excel for data cleaning and sorting and Random forest algorithm using python in jupyter notebooks to perform the prediction analysis.
5.Assessment of postfire vegetation recovery dynamics and it’s driving factors for Mount Kenya Forest: JKUAT
This involved application of Gis and Remote sensing project to analyze satellite imagery for a fire occurrence in mount Kenya in 2012.Google earth Engine (alias GEE)was used to generate the classified burn severity maps using JavaScript scripts were used to perform the driving factors influence using Random forest algorithm as well as the multicollinearity between the most influencing variables. Random forest prediction in Python was finally applied for prediction of the vegetation recovery in the forest