A. AWS S3 & Sentiment Analysis.ipynb
-
Objective: Engineer model for sentiment analysis of product reviews.
-
Data: Download Amazon Product Reviews from AWS’s Registry of Open Data.
-
Data prep: With pandas, seaborn, and sklearn to clean, transform, and export the data in preparation to train the data for a sentiment analysis algorithm.
-
Visualize sentiment metrics with word cloud, violin plots, & bar charts.
-
Take the first step in Data Lake formation, using AWS Glue & Athena to catalog metadata & query it using Athena. Use Jupyter’s custom magic capabilities/SQL to incorporate Athena’s query abilities.
B. Basic Plotting with Matplotlib.ipynb
C. Supervised Learning & Machine Learning with Sklearn.ipynb
-
Objective: Use machine learning models to determine if observations are of a benign or malignant tumor.
-
Data: Breast cancer data. Exploratory data analysis with visualization. Data preparation: Create a dummy classifier, train/test split.
-
Models: Fit logistic regression, decision tree, & random forest. Analyze & visualize models' performance: Confusion matrix, accuracy, precision, recall, & ROC/AUC.
