run those files or notebooks in order:

*in the utils.py you can find all the necessary function that we used in other files and notebooks
*the fake_samples_generator generate fake data for our data set
*By running Generator.py we generate the dataset as well as the queries and the users set
*the utility_matrix_generator generates both the random and the pseudo-random utility matrix
*in the part A notebook you can find the main query recommendation algorithm as well as the experiments and the figures made
*in the part B notebook we propose a method to compute the utility of a new query for all users.The algorithm calculates the scores of a new query for all users by comparing it to previous queries and their associated scores in a
utility matrix. It also evaluate it by computing the rmse between every rated query and its predicted value
*in the experiment notebook you can find all the expiriments shown in the report