Homework 3 Review

[akamlani]

Master Check List:

• [x] session duration time (in minutes) calculated for each row entry
• [x] regression model

Nice job on a couple particulars:

• Identifying outliers via percentile rather than value filters
• String to binary vector via 'get_dummies'
• Visualization of scatter plots
• Scatter Matrix for Correlation Feature Comparisons
• Regression Model as base vs w/class additions for model comparisons
• Defining new features (e.g. student class average session duration)

  General Comments

• show head of student_class_avg_session_duration, instead of entire array
• how about a regression model comparison with other features

  Questions

• any particular reason you chose to identify session time in minutes vs seconds/hours/days?
• do you find the visualization more helpful than numbers for correlation purposes?