Question: Build up a collaborative filtering-based recommender system to recommend hotels to users. The following table includes the ratings from 4 users to 3 hotels.
Build up a collaborative filtering-based recommender system to recommend hotels to users. The following table includes the ratings from 4 users to 3 hotels. Each rating is an integer from 1 to 5. "?" means that the user has not rated the hotel. Rating User 1 User 2 User 3 User 4 Hotel 1 2 5 ? 2 Use gradient descent algorithm for the collaborative filtering model. Some settings are as follows. The learning rate is a = 0.0003. The regularization parameter is =0.01. The dimension for feature/parameter vectors is K = 2. The initial values of the feature vector for hotels are x = parameter vector for users are T Hotel 2 3 ? 3 4 = [0.13 0.191 0.83 0.32 0.23 0.04 0.27] L0.95 Hotel 3 ? 1 2 1 [0.64 0.14 0.181 L0.23 0.75 0.32] and the Please answer the following questions: c) Calculate the values of x 2) (i.e., the first element in the feature vector for Hotel 2) and () (i.e., the first element in the parameter vector for User 1) after the first iteration. [15 marks] d) If we obtain x() [1.1983 2.3214] and () what is the rating of User 3 on Hotel 1? [5 marks] = = [1.8742 1.1234] after training,
Step by Step Solution
There are 3 Steps involved in it
To calculate the values of x2 the first element in the feature vector for Hotel 2 and 1 the first element in the parameter vector for User 1 after the ... View full answer
Get step-by-step solutions from verified subject matter experts
