A recommendation engine tries to predict how much a user will enjoy certain goods (movies, books, songs, etc) and makes recommendations. They are often used by online vendors to suggest new purchases.

learn more… | top users | synonyms

0
votes
0answers
5 views

How to optimize a SVD recommender regarding number of factors?

I am using R to do recommendation based on pureSVD. It is basically to choose the number of factors and then SVD the user-item matrix and then restore the matrix and provide top-N recommendations for ...
0
votes
0answers
11 views

Recommending Items based on Multiple Similarity Metrics

I'm building a recommendation systems that suggests items to a user based on items chosen by similar users. It's similar to collaborative filtering, although I am using multiple dimensions to describe ...
0
votes
0answers
6 views

Recommendation System with Graph Database

Hi I'm looking into creating a recommendation system based on a graph database for an ecommerce website. Basically something like amazon, "if my customer's have bought these products recommend some ...
0
votes
0answers
14 views

how to compare a user-based collaborative filtering with item-based

Apart from prediction accuracy, what are good evaluation metrics for comparing user-based and item-based collaborative filtering
0
votes
0answers
15 views

Collaborative Filtering, matrix decomposition, and incorporate various kinds of data

Just about every matrix factorization (e.g., SVD++) has some matrix that includes a n users and m (e.g.,) ratings. Here is my question, how do you include information like demographic information, ...
1
vote
0answers
25 views

clustering in Item-based collaborative filtering

I read about item-based collaborative filtering in the paper from Sawar et al. I want to apply clustering on items to find the most similar items and then apply the prediction. Is this a good ...
2
votes
2answers
105 views

Are there any probabilistic models for graph-based recommender systems?

All I can find now is somehow based on random walks or graph kernels, which is nice, but I want to have a more or less solid probabilistic foundation for my recommender system for bounds and ...
0
votes
0answers
32 views

The difference between SVD and SVD++

What if any is the connection between SVD (the one you learn about in your linear algebra course) and SVD++ (the one from the Netflix prize)? I know they both want to find latent factor spaces. But ...
0
votes
1answer
27 views

Compute the user and item features in SVD++

I have a sparse matrix. There is lots of missing data. Hence, I can't use SVD naively. I read Koren's SVD++ paper. I'm confused as to how the $q_i$ and $p_u$ vectors are determined. $q_i^Tp_u$ is ...
2
votes
1answer
34 views

What does “aspect model” refer to in machine learning

Hopefully this is the right place to ask my question. I am reading this paper about cold-start recommendations: http://dl.acm.org/citation.cfm?id=1352837 the expression "aspect model" is used a lot ...
1
vote
0answers
29 views

Confidence estimation for data points in a recommender system

I have a 100 by 100 matrix. Each cell is either 0, 1, or missing (denoted NA). Rows denote 'users' and columns denote 'items'. My goal is to impute the missing values, and provide a confidence level ...
0
votes
0answers
29 views

Cosine Distance and Recommender Systems

Suppose that a computer is rated according to three numerical features: processor speed, disk size, main-memory size. Consider three computers $A,B$ and $C$ with values: $$A(3.06,500,6)$$ $$B(2.68, ...
0
votes
0answers
16 views

How to find communities?

I know how to implement community detection algorithms in code. Of course, I know how to google. I know how to find groups on social networks. But I feel like it's only tip of an iceberg. Sometimes ...
1
vote
0answers
76 views

What is a good and efficient algorithm for a content based recommender?

I want to build a content based recommender in a restricted environment regarding cpu power and memory (to be specific: a mobile device, but it is not acceptable to build the recommender on a remote ...
-1
votes
1answer
75 views

What does ItemAverage Recommender do in Mahout

The description sounds to me as if it makes a MostPopular recommendation. But the MostPopular recommendation I did myself got much better results. So what does this recommender really return? It is a ...
1
vote
0answers
172 views

Is there an implementation of Item-Item Collaborative Filtering in R? [closed]

I was reading about item-item collaborative filtering at: http://guidetodatamining.com/guide/ch3/DataMining-ch3.pdf But I did't find an implementation in R of this algorithm. Is there one?
1
vote
1answer
120 views

Matrix Factorization Model for recommender systems how to determine number of latent features?

I am trying to design a matrix factorization technique for a simple user-item, rating recommender system. I have 2 questions about this. First in a simple implementation that I saw of matrix ...
1
vote
2answers
71 views

How to integrate users' profile information into a recommender system

I need to build a recommendation system with lots of information about users (age, sex, location, income etc), but very sparse information about users' prefernces (i.e 1-2 products consumed out of 100 ...
2
votes
0answers
30 views

Given $n$ ratings from 0 to 1, how to calculate a weighted “average” estimating the “true” rating?

I've read How Not to Sort By Average Rating regarding how to average binary positive/negative ratings in a way that takes the number of ratings into account. The author uses the "lower bound of Wilson ...
1
vote
1answer
103 views

Why Bayes Rule in Naive Bayes compared to simple P(class|features)

I would like to improve on my recommendation system. Imagine I have training data of $M=7,000,000$ samples. Each training sample contains a variable number of words in the body, and a variable amount ...
1
vote
0answers
108 views

Incremental SVD in Collaborative Filtering

In the so-called incremental SVD used for collaborative filtering: http://www.machinelearning.org/proceedings/icml2007/papers/407.pdf http://www2.research.att.com/~volinsky/papers/ieeecomputer.pdf ...
5
votes
1answer
114 views

Conditional Logit for recommender systems?

Are conditional multinomial logits used for recommendation engines? Although they are commonly used in econometrics, I've never heard it used or discussed in the context of recommender systems. ...
5
votes
2answers
95 views

Simple recommender system - where to start?

Without going into specifics, I'm currently working on a system that involves 20-25 questions being answered as either Green, Yellow, Orange or Red. After completing a subset of these questions (many ...
2
votes
0answers
29 views

How to build event recommender based only on events' descriptions

I built a simple app that grabs events using eventbrite and meetup.com APIs and displays it based on your zip code. Now I'm trying to build a simple event recommender that will use events that you ...
1
vote
1answer
54 views

Item correlation for recommender system

I just made an implementation of P(A|B)/P(¬A|B) for a "people who bought this also bought..." algorithm. I'm doing it by ...
0
votes
1answer
95 views

How to deal with giant sparse matrices?

I'm looking to do some heavy-duty manipulation of some really large and often very sparse matrices and I'm looking for the right tool for the job. These matrices will be much, much larger than the RAM ...
0
votes
2answers
49 views

algorithm for finding similar items

I have a group of items S, say there are 100 items in S, and I know some features of these items, i.e., color, size, ...Now, I have another group of items P, say there are 10000 items in P. What can I ...
1
vote
2answers
85 views

Support Vector Machines and Recommender Algorithms

The recommendation problem is this: Suppose that we have a matrix whose columns are items (e.g., movies) and whose rows are users. A small part of the matrix is filled with rating values. That is ...
0
votes
4answers
234 views

Algorithm for a 5 star rating system

I'm creating a reporting tool that produces a data set and calculates a metric for each item in the data set. I have two algorithms that I'm working with: an algorithm that produces the metric and an ...
1
vote
1answer
195 views

recommenderlab, why does the similarity function return only 0's and 1's? [closed]

In a matrix with user ratings I want to calculate similarities between the first ten users and the rest of the users. I use the method "jaccard" for the similarity calculation in recommenderlab. My ...
1
vote
1answer
106 views

How to compute K and n? [Item-based Collaborative Filtering]

I'm currently studying an item-based collaborative filtering algorithm described in Ul Haq, Raza - Hybrid Recommender System Towards User Satisfaction. I've formulated the algorithm below based on it. ...
1
vote
1answer
68 views

What is the standard procedure for evaluating a user-based CF algorithm with a dataset offline?

I have read some papers and other materials about the evaluation of recommender systems (RS). Most of them discuss the various properties of RS (e.g. accuracy, diversity, etc.), and metrics for ...
3
votes
1answer
92 views

Any public available evaluation frameworks for recommender systems?

I would like to know if there are any evaluation frameworks of recommender systems which are capable of evaluating rating prediction and topN recommendation (Precision and recall etc.). Maybe I need ...
3
votes
0answers
77 views

Does it make sense to use Lasso regression for recommendation systems?

I'm really intrigued by Lasso Regression, and it seems like a potential candidate for estimating “objective” ratings for items based on the users who watched them. The user data is sparse, and the ...
2
votes
3answers
513 views

Best machine learning approach for recommendation engine?

I am given a dataset where there are people profiles and the types of beer each person likes, given in a list. What is the best way to find relevant beers given a specific beer based on this data? ...
2
votes
2answers
147 views

How to treat rare / new items in the validation of a recommender system?

I'm working on movie recommendation algorithm. The data set consists of about 40 million ratings (user, film, rating). I want to separate the ratings into two groups - training set and probe set. The ...
2
votes
2answers
143 views

Finding similar items based on user likes and dislikes

I have data on content-based recommendations of movies and their attributes. Suppose a user likes x,y and z and also dislikes c and d movies. I want to predict movies that he will like based on his ...
1
vote
2answers
85 views

Affinity analysis dealing with power law distribution

How do you account for top selling/most central items when analyzing affinities? For example, let's say you're analyzing affinities among movies based on people who rented both. How do you deal with ...
0
votes
2answers
154 views

Adding content information to matrix factorization-based recommender

I'm currently using a matrix factorization method to generate recommendations (for info on this, check: Matrix Factorization Techniques for Recommender Systems). At the moment, my rating estimate is ...
0
votes
0answers
142 views

alternating least squares and all zero rows-columns

I'm trying to implement the algorithm described in Large-scale parallel collaborative filtering for the Netflix prize and I'm having some trouble understanding how to deal with a rating matrix that's ...
0
votes
1answer
41 views

Calculating user preferences based on purchases: how to incorporate different variables for comparison

I am trying to come up with a metric to calculate user preference correlation for my final project (a web-shop for shoes) at school. Originally I intended to include user ratings and use Pearson's ...
1
vote
1answer
66 views

When should I update a recommendation engine?

[I asked this on StackOverflow and was told it would be a better fit here] I am including a basic recommendation engine in a very small project for my final exam. I understand the code and the math ...
3
votes
2answers
162 views

SVD in movie recommendation

Assume that there is a 5 $\times$ 6 matrix that records the ratings of six users on five movies. I have computed the singular value decomposition (SVD) for such a matrix. Suppose I add another ...
2
votes
0answers
227 views

Matrix factorization and gradient descent for recommender systems; user bias?

I've been reading about using Matrix Factorization techniques to do collaborative filtering. A popular thing to do seems to be to add user and item biases into the ratings prediction. What I don't ...
1
vote
0answers
51 views

MAE/MSE with or without square root

I read some papers about recommender systems and information retrieval, where Mean Absolut Error and Mean Squared Error are mentioned. But I've found some differences between the formal definition of ...
2
votes
1answer
417 views

Precision - Recall: Graphical Representation

I'm a little bit confused with precision recall. I read some papers about recommender systems, where in one paper they have a graphical representation and in other papers they don't (they just have ...
1
vote
1answer
121 views

Calculate boundary for MAE given RMSE

E.g. from the Netflix prize I know that the best RMSE = 0.8563 where the test dataset has a size of n=1,408,789. Can I calculate a boundary for the MAE. If not, why can't I calculate a boundary? I ...
1
vote
0answers
88 views

recommender system implicit rating to ordinal scale

There are 4 ways a user can show preference for an article within my news app: number of times an article was viewed; for how long was the article viewed; whether it was favorited or not; and number ...
2
votes
1answer
63 views

Using collaborative filtering to “clean” data and the other way around

I am considering two types of systems - which might have more appropriate names: Recommender systems: These recommender systems are based on collaborative filtering methods, both model-based and ...
3
votes
1answer
126 views

Most relevant algorithms for Collaborative Filtering to test against

I am working on algorithms for collaborative filtering (CF). As part of this work, I want to compare a new algorithm to previous approaches to the problem. I am also surveying the most important ...