Project about the building and the execution of a classificator
This is a university project developed for a challenge held in the course Sistemi Intelligenti per Internet (Smart Systems for Internet). It was required the development of a tool that can predict the rate of a place given by a particular user, through the data provided by Yelp.
We decided to use a mixed approach between a user-based prediction and a item-based one. In particular we computed a linear combination of these two different predictions calculating the lambda starting from the neighboroods of the user and the item (for each item we computed how many users were in common with the current-item, and for the user we computed how many items were in common with the current-user).
More detailed information about the project request can be found it here, while as regards the development see at this link (both located in italian)
Date
January 2014
See on GitHub
https://github.com/Pausa90/ChallengeSII
Authors and Contributors
This Project is developed by @V1LL0 and @Pausa90