This is a repository for my work in the Google Advanced Data Analytics course on Coursera.

Todo

• Get a link for above
• Polish

Details

1. Course 2: Python Introduction

• Loaded data from a csv file with Pandas
• Retrieved preliminary information and statistics
• Analyzed some features, especially in relation to the target variable, claim_status
• Concluded that there is a large difference in likes, comments, views, etc. between different claim_status

2. Course 3: Exploratory Data Analysis

• Constructed visualizations using Seaborn (histograms, box plots, pie charts, bar graphs, scatterplots, etc.)
• Analyzed distribution of features by class (of the target variable)
• Checked for statistical outliers
• Made conclusions about author_ban_status and other variables in relation to claim_status

3. Course 4: Hypothesis Testing

• Checked for missing data
• Prepared hypotheses for hypothesis testing
• Conducted a t-test to determine stastical significance

4. Course 5: Regression Modeling

• Initial analysis of target variable, verification_status
• Check correlation of features to satisfy model assumptions
• Split data into train and test sets
• Encode data using one-hot encoding
• Trained a logistic regression model
• Evaluated the model on metrics using a confusion matrix

5. Course 6: Machine Learning Models

• Split data into train, validation, and test sets
• Trained a Random Forest Model (from sklearn) on the data
• Trained an XGBoost Model (from xgboost) on the data
• Used Grid Search to tune hyperparameters
• Evaluated the model using metrics and confusion matrices