A brief description of the projects worked on during my Data Science Diploma at BrainStation.
Strategized the factors that influence the volume of usage, popular stations and overall business growth using SQL and data visualizations using Tableau on the Bixi Bikes dataset.
Showed an increase revenue by 30% from past year via memberships, timing of promotions and changes in price model.
Performed EDA and Data analysis for analyzing data from previous United States presidential elections.
Linear model developed to predict the popular vote in the U.S. presidential elections from 2016 showed 62% accuracy.
Performed sentiment analysis (counter vectorization) of hotel reviews to determine how to increase positive reviews
Used different model fits and performed feature engineering to determine improvement in prediction accuracy assessed using confusion matrix.
Worked with a multi-disciplinary, cross-country team making food delivery sustainability goals realistic to customers and restaurants.
Food menu items should include a curated rating system called ‘Smart Score’ - factors influencing the rating - locally grown crops (total commodity supply), health/nutritional needs, distance delivered from and LCA emissions.
Applied skills learnt in the Big Data Fundamentals unit to load, filter, and visualize a large real-world dataset in cloud-based distributed computing environment using Hadoop, Spark, and the S3 filesystem.
Followed a Big Data analysis workflow where we filter and reduce data down to a manageable size, and then do some analysis locally on our machine after extracting data from the Cloud and processing it using Big Data tools.
The castone project was completed as a requirement for the Data Science Diploma and aims to use data science modeling tools on the available data to capture the trends of the CO2 exchange (flux) and stocks (storage) for Boreal Forests in Canada.
Available on github at https://github.com/rtherego/BrainStation-Capstone-Project