PyReco is a simple recommendation engine written in python 2.7 that given a list of people and their ratings corresponding to some items can return recommendation of items for a person. It can also give items that are similar to a given item and people that are similar to a given person.

It makes use of similarity ratings between the person for whom items have to be recommended and some other person to act as a weight for the rating of other person to predict the possible rating of that item for the person. It then takes the weighted sum average over all such other persons to get the expected rating for a item. This is done for all non rated movies of the person and then the expected ratings are sorted in descending order and returned as recommendations.

It also has functions for item based collabrative filtering provided in the file itembased.py

Deployed it on pythonanywhere Can see e.g of recommendations for Toby at http://siddharth.pythonanywhere.com/reco/

Update: Added support for retriving data from delicious in deliciousrec.py although it is slow due to delicious API limitations. You must install pydelicious first which is a python script providing functions to retrive data from delicious