Data mining assignment 2

Detailed task description

- Business understanding

• Your task is to predict what hotels properties listed as a result of a hotel search a user is most likely to click on.
• Can you find some other people that have tried to make such predictions (e.g. from the Kaggle competition)?
• And what have they used as most prominent predictors?
• Have other people that participate in the competition mentioned anything about their approaches?
• Please spend a couple of paragraphs on this topic (i.e. related work) in your report.

Data understanding

• EDA

Data preparation

• There are certain attributes with a large amount of missing values, do they still provide useful information? And how will you handle a missing value if this shows to be the case?

Modeling and Evaluation

• To test how your model is compared to other, you can upload your answers for the test set on the in class Kaggle website: https://www.kaggle.com/c/vu-dmt-2assignment/ . Note that the score shown there is only for part of the test set, the score on the rest of the test set will only be disclosed on the final lecture and will form part of your final grade.

To do:

• Read the file with the explanation
• Explorating data analysis (for 7/5)
  • Plot --> a lot