In which of the following situations will a collaborative filtering system be th
ID: 3727452 • Letter: I
Question
In which of the following situations will a collaborative filtering system be the most appropriate learning algorithm (compared to linear or logistic regression)?
You're an artist and hand-paint portraits for your clients. Each client gets a different portrait (of themselves) and gives you 1-5 star rating feedback, and each client purchases at most 1 portrait. You'd like to predict what rating your next customer will give you.
You run an online bookstore and collect the ratings of many users. You want to use this to identify what books are "similar" to each other (i.e., if one user likes a certain book, what are other books that she might also like?)
You manage an online bookstore and you have the book ratings from many users. You want to learn to predict the expected sales volume (number of books sold) as a function of the average rating of a book. You own a clothing store that sells many styles and brands of jeans.
You have collected reviews of the different styles and brands from frequent shoppers, and you want to use these reviews to offer those shoppers discounts on the jeans you think they are most likely to purchase
Explanation / Answer
Lets first understand what is collaborative filtering and regression , in regression based on the datasets collected we plot a regression model based on whether it is multivariate or single variate graph is created the prediction will only revolve around this graph which means that it would predict a statistical value for the dataset based on the best fit of the data points.The basic idea of collaborative filtering algorithm is that when we get one of the data, we derive the other one, and then we deduct it and then deduct it to optimize it. After optimization, we continue to deduce and continue to optimize, so we can derive the cycle and coordinate.:-
1. Here as mentioned above we don't have multiple agents to predict, we only the stars given by users in previous dataset in this case since we don't have other parameters or features related to the purchase data we can't apply collaborative filtering rather the regression would be able to predict the stars that the next customer will provide.
2. You run an online bookstore and collect the ratings of many users. You want to use this to identify which books are similar (that is, if a user likes a book, what book might she like?) There are many features, so we use collaborative filtering here.
3. This is again similar to first case we have just one vraiable rating which affects the volume sales so simle linear regression would do.
4. You have a clothing store that sells many styles and brands of jeans. You have collected reviews of different styles and brands from different shoppers. You want to use these reviews to discount customers who you think are most likely to buy jeans. A lot of features , so in this case we use collaborative filtering.
Note:- if this answer was useful for you please give a thumbsup and comment your queries if any.