NathanNam
commented
7 years ago Dave, thanks for your feedback. While I understand you think the match (2.7M hours) is not compelling, I'd like to keep this number to emphasize a potential impact of this project. I will look for more compelling numbers later.
Thanks, Nathan
The concept: Provide a tool for apartment/house hunters to help them understand how rents are trending, and where they might go. This prediction will help future renters know if the offering is well priced.
I really like the idea, knowing how difficult it is to find an apartment in a sellers market. I'd definitely keep the description of the travel site comparison. This resonated and I immediately thought of the kayak predictor that tells you to buy now or wait.
On the flip side, the idea of a recommender system implies that you are doing more than just predicting the future price -- rather comparing that prediction against relative rents for similar profiles and suggesting if the rent is a good price. I think you could add some description to further articulate this value.
I understand the basis for calculated time savings, however I didn't find the math compelling. Rather, I would have been more influenced by knowing the tool helps renters find better deals and better homes.
Very interesting idea!