The best location for YOU to live is a tool that allows anyone to find their ideal residence place in a given city by specifying their personal level of importance for the location environmental parameters (noise,pollution...) and visual appeal.
The model is based on multiple layers of machine learning techniques applied to images of the city streets obtained with Google StreetView's API.
While environmental characteristics are strapped from multiple sources, the images appeal is derived from the results of a survey, in the form of a mini-battle between two images.
By means of an object recognition algorithm, a given view is associated to its vector of features containing the proportion of pixels taken by each object within the image. A regression analysis then allows to find what features are the most significant for users and to make a prediction of the most visually appealing neighborhood in another city for which no survey results are available.