Modelling, Evaluation and Interpretation

The task in hand demands us to predict whether a visitor will transact or not based on the features available. The target variable in this dataset is ‘Revenue’ column which makes it the dependent variables and the other variables as independent variables. Models helps us understand the relationship between the dependent variables and independent variables. This section is concerned with the task of modelling, evaluation of those models, and interpretations of those models. The metric that will be used for evaluation will be the same that is used in the original papers by the aithors i.e. the F1 score. For interpretation, techniques depending upon the model will be used. Following are models that are used in the project:

  1. Logistic Regression

  2. Support Vector Machine - Support Vector Classifier

  3. Decision Tree Classifier

  4. Adaboost Classifier

  5. Random Forest Classifier

Exploratory Data Analysis - Multivariate Logistic Regression