Machine Learning & data modelling

- Prepare raw data to ensure it is clean and ready to model as follows: Clean the data as appropriate for a textual data source

- Apply exploratory methods as appropriate to textual data. Marks will be
awarded as follows:
1. Explore the dataset to check if the data is balanced
2. Extract features from the data using Count Vectorizor
3. Identify the most common words for spam and ham sms messages

- Apply the correct machine learning/modelling approach to model the
data appropriately.
- Model training
o Split the dataset into training and test sets, and ensure the training
dataset is balanced (using SMOTE)
- Apply two machine learning/modelling approaches to model the data
o Correctly and completely train both models, including the selection of
appropriate hyperparameter values (including use of hyperparameter
tuning using grid search)
o Run the model using the best parameters
- Model evaluation
o Successfully and fully test the models, using appropriate measures for
evaluation (e.g. accuracy, balanced accuracy, training time, prediction
time)