Questions tagged [recommender-system]

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.

Filter by
Sorted by
Tagged with
0 votes
0 answers
7 views

Matrix Factorization for Recommendation System vs just calculating Conditional Probabilities

I have a matrix factorization model (using Lightfm, without user-features) to recommend items to users, based on a user-item matrix. Context A particular recommendation stuck out to me as curious, ...
user avatar
  • 283
3 votes
1 answer
38 views

Matrix Factorization and Overfitting

I recently came accross the algorithm of Matrix Factorization for a recommendations system. One of the tutorials I followed can be found here. According to it given the initial matrix $R$ and the ...
user avatar
0 votes
0 answers
27 views

How to calculate Cosine Similarity from Keras model?

I'm trying to make hybrid recommender system that recommends movies to users from Movielens dataset. Its Content part is based on Doc2Vec model from gensim library and its Collaborative Filtering part ...
user avatar
0 votes
0 answers
23 views

How to choose the best recommender system? What evaluation metrics to use?

I want to build a recommender system to suggest similar songs to continue a playlist (similar to what Spotify does by recommending similar songs at the end of a playlist). I want to build two models: ...
user avatar
  • 1
0 votes
1 answer
15 views

Recommend items to complete a set

I'm trying to predict/recommend items to add to an incomplete basket. In this case, the basket is a project — e.g., you detect a customer trying to fix the AC in their car, hence you suggest things ...
user avatar
  • 109
0 votes
0 answers
19 views

Recommender system with regression output

So basically I have a some products (basket A, which may differ from store to store) which are sold and some other products tend to be sold according to what happens in basket A. I Need to forecast ...
user avatar
  • 209
0 votes
1 answer
26 views

weighted average in collaborative filtering

I studied on collaborative filtering recently and found that for ratings at last most methods applied were weighted average, no matter what they had proposed, similarity, time, etc. Thus I am ...
user avatar
0 votes
0 answers
17 views

How would we set the initial parameters of beta distribution of Thompson sampling if we want to start the model with the existing data?

This was one of the business-related questions from my technical interview last week for a data science position in a recommender system team at a search engine company focusing on advertisement ...
user avatar
  • 221
1 vote
1 answer
27 views

What is the general right call to make from a marginal difference in A/B test results in recommender system?

This was one of the business-related questions from my technical interview last week for a data science position in a recommender system team at a search engine company focusing on advertisement ...
user avatar
  • 221
1 vote
0 answers
29 views

Statistical test for recommendation algorithms

I've developed 2 different image based recommendation systems for an e-commerce company and now it's time to evaluate them using some statistical tests. Both algorithms work in a way in which if fed ...
user avatar
  • 329
2 votes
1 answer
32 views

Sample size calculation for precision at $k$: a/b-testing a recommender system

I would like to conduct an online experiment to compare two different versions of a recommender system. The system returns a list of $r$ ranked recommendations. I would like to evaluate the ...
user avatar
0 votes
0 answers
15 views

Why is performing a collaborative filtering or PCA not clustering?

Today I had a very brief discussion with someone who claimed that performing collaborative filtering (CF) is a way of performing clustering. The reasoning that was given for CF to be clustering, was ...
user avatar
  • 521
0 votes
0 answers
93 views

Funk SVD for binary data - product like or dislike

Assume the following situation: you have a user-item sparse matrix. However, instead of the usual 1 to 5 rating scale, items can only receive a positive (1) or negative (-1) feedback. Thus, the matrix ...
user avatar
  • 133
1 vote
0 answers
26 views

Content-based filtering recommendation system for a group

Is there any example of content-based filtering recommendation for a group of observations? For example, user can choose a group of 10 movies they like the most, and the recommendation engine will ...
user avatar
0 votes
0 answers
25 views

Can I restrict action space based on state in DDPG?

I am working on a RL based recommender system using the DDPG algorithm and wanted to restrict the action space (possible recommedations) based in the user. I have tried using the restricted action ...
user avatar
0 votes
1 answer
29 views

Sequential recommendation: how to effective encoding output item?

Now I am learning about sequential recommendation - session based recommendation. I have understood that User-item interactions may be viewed as sequential action (first I clicked item A, then click ...
user avatar
  • 244
1 vote
1 answer
36 views

Is there one splitting strategy for both K-NN and Matrix Factorization recommender systems?

I am researching several different recommender systems, some of which are based on a user K-Nearest Neighbour algorithm and some of which are based on a matrix factorization algorithm. My dataset is ...
user avatar
1 vote
1 answer
37 views

What's the difference between a recommender system and a decision tree?

We learned in Machine Learning that both of those techniques try to predict an output (whether person A likes a specific product or whether person A has a high default risk) based on data of other ...
user avatar
0 votes
0 answers
23 views

efficient way to create negative samples in recommender system

I am trying to build a binary classification model that predicts whether user will buy a product when recommended with card transaction dataset. Therefore data are only stored when transaction ...
user avatar
  • 101
0 votes
0 answers
24 views

How many variables for a recommender system?

I'm creating a survey for 60 people. I'd like to measure their music, tv-series and purchasing tastes. The survey will show a number of songs. People will be asked to say if they like it (1) or not (0)...
user avatar
0 votes
0 answers
76 views

Next best offer models in recommendation system

I'm trying to so some research on "Next Best Offer" or "Next Best Action" models used in recommendation systems. Searched on google but didn't find any article with detailed ...
user avatar
  • 289
0 votes
0 answers
13 views

What is the linearity assumption used in the paper - Off-policy evaluation for slate recommendation

I am reading the paper, Off-policy evaluation for slate recommendation. I am quite confused by their statement, "Specifically, we posit a linearity assumption, stating that the slate-level reward ...
user avatar
0 votes
1 answer
39 views

Search, rank and recommend in large text datasets

Imagine you are Spotify and you have billions of songs. Assume that each of these songs are transcribed into text. How do you design your search and recommendation pipeline such that when somebody ...
user avatar
  • 118
0 votes
0 answers
16 views

Evaluating recommender system implicit vs explicit

I have rating data that allows me to create a recommender system based on these explicit rankings. Moreover, I want to test some methods that utilize implicit feedback. I can obtain it by binarizing ...
user avatar
  • 75
0 votes
0 answers
16 views

Combining two very different prediction models

My entire dataframe looks like this: ...
user avatar
0 votes
0 answers
20 views

Values overshooting for sparse matrix factorization (recommendation system)

Using this article as reference for ease of replicability, I noticed that when expanding the pivot matrix R with many missing values, the final recommendation matrix tends to have values overshooting ...
user avatar
0 votes
0 answers
24 views

Constructing a User Profile for Music Taste

My goal is to construct User profiles based on positive (and maybe also negative) interactions with songs. A User has the option to like a song. This would give me a list of likes for each user. With ...
user avatar
0 votes
0 answers
26 views

Why does the RMSE stay the same regardless of the algorithm that I use?

I have a dataframe with users, items, and ratings that are either 0 or 1. There are more items than users, some users might rate lots of common items, and some not any common items at all. Here is a ...
user avatar
0 votes
0 answers
22 views

Using Classification using User Profile for Product Recommendation

I want to build a ML model which will recommend products to users (Currently I have 3 products to recommend A,B and C) So I have some user profile data, like address, gender, age, and few other ...
user avatar
0 votes
0 answers
19 views

Measuring similarity between students to build an exam recommender system

I am playing around with the following problem: I have various groups of students from various schools and they receive online questions. A group receives the same questions, but different groups in ...
user avatar
0 votes
0 answers
15 views

Correct metrics for recommender system models based on explicit and implicit data respectively

For modeling explicit ratings, regression metrics like RMSE, MSE and MAE should be used for evaluation. For modeling implicit ratings, ranking metrics like precision, recall, MAP@K should be used for ...
user avatar
  • 643
2 votes
2 answers
64 views

What are the pros and cons of using a supervised learning approach for building a recommender system compared to traditional approaches?

Pretty much all articles I read about recommender systems use Collaborative Filtering, Content Based Filtering and Hybrid approaches. Virtually no one mentioned about supervised learning approaches. ...
user avatar
  • 643
2 votes
2 answers
121 views

The role of bias terms in binary recommender systems

I realize that a recommender system applied to, for example, the Movielens dataset needs to account for bias. That is, one needs to adjust for the varying popularity of movies, and that users have ...
user avatar
  • 1,042
0 votes
0 answers
24 views

Should the intercept be omitted when fitting a pairwise Learn to Rank Model?

In a typical Learning to Rank setting for a recommender system, suppose we have a number of feature scores $\mathbf{x} = x_1, ..., x_n$ for each document, for a given query. The pairwise paradigm (e.g....
user avatar
0 votes
0 answers
18 views

Why does Non-Negative Matrix Factorization reconstructs exactly the same matrix?

I'm trying recently to get into recommender systems and almost all tutorials I find mention collaborative filtering done with matrix factorization. I found this tutorial that describes how to build ...
user avatar
0 votes
0 answers
14 views

Predicting user ratings based on ratings and demographic data on the fly

would like to predict the rating (0-5 stars) a user would give to an item. My data looks like this: for every user I have the age-group, gender-group, and two other factors right after the user ...
user avatar
  • 101
1 vote
1 answer
17 views

Latent factors are the same in both decomposed matrices?

This question is in the context of recommendation systems. We can use matrix factorization techniques to decompose a user-product explicit/implicit matrix(R) into two matrices(U, P). Let's say R is a ...
user avatar
  • 643
1 vote
1 answer
35 views

Recommender System without collaborative or content filtering

Lets say I have a library of items. I want to show some of these items to the users and I want to show what I think would be the best items to the user. I don't have access to the preference of the ...
user avatar
0 votes
0 answers
25 views

Best model for dataset with grouped data

I'm working on a recommender system for our website. Users visit the website and choose content to consume. Users tend to consume the same content at the same hour of the day and on the same day of ...
user avatar
  • 137
0 votes
0 answers
19 views

Recommender System without Ratings but Duration instead

I'm currently working on a recommender system without ratings variable. I only have the watch duration for streamers and I should be able to recommend a list of streamers with importance on its order. ...
user avatar
1 vote
0 answers
92 views

How does Neural Collaborative filtering (NCF) handle new users (cold start)?

I am building model with NCF and i know NCF has great power to handle sparse data, and find great representation for users and items. However, I don't know how would NCF handle new users? my structure ...
user avatar
0 votes
0 answers
22 views

Sort items based on frequency and recency [duplicate]

I'm working on a problem which requires me to sort a list of static items for each user. I understand best way to solve this problem would be to come up with a function that captures both the ...
user avatar
  • 109
0 votes
0 answers
16 views

Function to sort a static list of items by recency and frequency

I'm working on a problem which requires me to sort a list of static items for each user. I understand best way to solve this problem would be to come up with a function that captures both the ...
user avatar
  • 109
0 votes
0 answers
23 views

Best way to pre-process this data for a recommender?

I've created this data set, time series, with timestamps, and 3 columns, user_id, events, and next_hotel. I want to use next_hotel as a response variable. user_id: ...
user avatar
  • 3
0 votes
1 answer
54 views

Recommender system for matching user input keywords to objects that have different keywords assigned to them (and getting the matching weights) [closed]

I'm looking for some tips in the right direction as to what to look into for this recommender system: We have a predefined set of objects, each with a few keywords assigned to them. We can call the ...
user avatar
0 votes
1 answer
128 views

why do we calculating dot product between a matrix and its transpose to capture interaction of data?

I am wondering why in this work https://github.com/facebookresearch/dlrm, authors are calculating dot product of embeddings by its transpose. Here is the sentence from their paper, the last paragraph ...
user avatar
0 votes
0 answers
31 views

Listwise Learning to rank for partially labelled list of ranking data

I was reading about the Listwise approach for LTR. The first Listwise LTR paper ( refer page 3, left column, para 2-3) explicitly mentions "n(i)" which implies the number of ranked Documents ...
user avatar
0 votes
1 answer
52 views

Evaluate recommender system based on profit margin generated?

Let's say that I have two recommendation system models built, Model A and Model B. Now I track the performance of both the models for 5 days from 1st Jan to 5th Jan. Each model has been assigned a ...
user avatar
  • 1,792
1 vote
0 answers
50 views

Cosine similarity with recentering - collaborative filtering

I don’t know much about stats (nor maths), so I’m sorry if I’m not being very clear on this... I’m trying to build a simple recommender system for books using collaborative filtering item by item. I’...
user avatar
0 votes
0 answers
57 views

Product cannibalization - Would affinity analysis help?

We send an email to our customers to advertise 3-4 products every day. Supposing that today's email is advertising three products (product A, product B, product C), the customer is given four options: ...
user avatar

1
2 3 4 5
9