Embedded Files
Kate Njoki Mbugua
Highlights
Highlights
The goal of this challenge was to write a python modular code to fetch and visualize LIDAR high definition elevation data - USGS 3DEP .
The USGS recently released high resolution elevation data as a lidar point cloud in a public dataset on Amazon.
Conducted exploratory data analysis on the Telecom data to generate meaningful insights. Used K-means clustering to analyze customers' engagement with the company's products. Carried out Linear Regression to predict satisfaction scores of Telecom customers.
Online users were grouped into 2 groups, a control group and an exposed group and shown a dummy Ad and Smart Ad respectively.
To test how effective the SmartAd was, I first carried out Classical A/B p-value testing. Next I used machine learning algorithms such as Logistic Regression, Decision Tree Classifier and XGBoost Classifier to generate even more complex insights.
Check out the deployed app here!
Twitter COVID19 Data Analysis
Twitter COVID19 Data Analysis
Carried out topic modelling and sentiment analysis on Twitter Data using the gensim package. Built a dashboard using Streamlit
The store sales were forecasted using Facebook Prophet, Machine Learning algorithms such as Random Forest, SGD and Decision Tree regression models. LSTM, a Deep Learning Neural Network was also implemented.
I also built a streamlit dashboard here!
Page updated
Google Sites
Report abuse