0
$\begingroup$

I want to apologise in advance for my ignorance but I'm hitting my head in a wall here as I'm not sure how to proceed with my project and I've got no experience in that field even though I'm doing a masters in Data Science (don't slag me too much pls) So I've started a project with this small business that offer self-catering accommodation like Airbnb and what they want is to increase their occupancy. Basically they have properties which perform well and some that doesn't and they want to know why. I've got a list with variables but no clue how to use any other software than weka and I am not sure if it will be appropriate. I am also clueless how to proceed with the analysis so I would be grateful if you have any ideas. I've added a screenshot of the variables to give you an idea.enter image description here My university supervisor is not very helpful so I've decided to reach out for help here and any tips will be much appreciated.

Would it be more helpful to provide you with the excel sheet? I am not sure how to upload it

Improve this question
$\endgroup$
1
  • $\begingroup$ Be very careful with making the company data public, you would need their permission before doing that. $\endgroup$
    – Erwan
    Jul 19, 2019 at 16:09

1 Answer 1

Reset to default
0
$\begingroup$

I would start with a simple decision tree regression with the occupancy rate as target value. By visualizing the tree after training you can see which features impact the occupancy rate. You can do this in Weka with M5P regression: http://weka.sourceforge.net/doc.stable/weka/classifiers/trees/M5P.html

Improve this answer
$\endgroup$
1
  • $\begingroup$ That's great! Thank you so much and yes, I should've considered that first, my bad. $\endgroup$
    – Plamena Koleva
    Jul 20, 2019 at 18:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.