Description
Instructions:
1. Go to Kaggle.com. Find a text classification data set that interests you. Divide into train/test.
Create a graph showing the distribution of the target classes. Describe the data set and what the
model should be able to predict.
2. Create a sequential model and evaluate on the test data
3. Try a different architecture like RNN, CNN, etc and evaluate on the test data
4. Try different embedding approaches and evaluate on the test data
5. Write up your analysis of the performance of various approaches
Grading Rubric:
– Each part is worth 0 to 20 points
– Your grade is not determined by the accuracy achieved, but by how much work and thought you
put into it
